2025-in-review

It’s strange to measure every year against a concept developed by a science fiction writer, but William Gibson’s line “The future is already here—it’s just not evenly distributed”1. has been my north star for my recent year-in-review essays. Gibson meant that the future was unevenly distributed by class: the wealthy receive high-tech healthcare while the world’s poorest live in squalor—though one might ask which of these is really our future. Yet the quote has been repeatedly misread as a claim about time andspace: that the future arrives somewhere first, perhaps unseen, while the rest of the world catches up. But this misreading is more productive than Gibson’s intent. Gibson’s critique of inequality is fair enough, but we all know this, decry it, and go on about our business. The misreading, on the other hand, is a theory of historical change.

With the release of ChatGPT in late 2022, a temporal rift opened, shattering the post-Covidean present. But many tried the early tools, encountered hallucinations, read articles about slop and imminent environmental ruin, and reasonably concluded there was nothing to see. By 2025, a cursory examination of news in AI would have assured them that AI had proved a bust. OpenAI’s long-awaited updates disappointed, and the company flailed, turning to social media with Sora, a TikTok clone for AI. Meta seemed to abandon its efforts to create a competitive AI and instead turned to content generation for Instagram and Facebook, something nobody on earth wanted. Talk of a bubble started among Wall Street pundits. The hype-to-disappointment cycle is familiar, and the dismissals were not unreasonable.

But again, the future isn’t evenly distributed, and if you don’t know where to look, you would be excused for believing it’s all hype. Looking past such failures, 2025 was actually a year of breakneck progress. Anthropic’s Claude emerged as the most capable system for complex tasks, Google’s Gemini became highly competitive, while DeepSeek and Moonshot AI proved that China was not far behind. More significant than any single model was the emergence of agentic AI—systems that can take on multi-step tasks, act, navigate filesystems, write and execute code, and work across documents. Claude Code was the year’s groundbreaking innovation. While “slop” was Merriam-Webster’s word of the year, “vibe coding”—using agents to write programs—was much more important. Not only could programmers use them to accelerate their work, it also became possible for non-programmers to realize their ideas without any knowledge of code, a radical change in access I explored in “What Did Vibe Coding Just Do to the Commons?”.

By any first-world standards, at least, these tools are remarkably democratic and inexpensive. A basic Claude subscription costs about as much as a month of streaming, and even the $200 maximum usage account costs less than a monthly car payment. For many, however, the barrier is not price but something deeper—a resistance approaching revulsion. These tools provoke fear in a way that earlier technologies did not. It’s not the apocalyptic dread of the doomers or the Dark Mountain sensibility that apocalypse is near. Rather, it’s a threat to the sense that thought itself is what makes us distinct. The unevenness of the future is no longer about access; it’s now about willingness to engage.

As a scholar, thinking about the very short term is strange for me. I have always been suspicious of claims that radical change was upon us. I would rather align myself with the French Annales school concept of la longue durée, as defined by the great Fernand Braudel, the long-term structures of geography and climate. Faster than that were the medium-term cycles of economies and states, while he dismissed the short-term événements of rulers and political events as “surface disturbances, crests of foam that the tides of history carry on their strong backs.”2. Events, he wrote elsewhere, “are the ephemera of history; they pass across its stage like fireflies, hardly glimpsed before they settle back into darkness and as often as not into oblivion.”3. The real forces operate beneath, slowly, often imperceptibly.

Curiously, Braudel himself embraced technological change in his own work. In the 1920s and 30s, he adapted an old motion-picture camera to photograph archival documents—2,000 to 3,000 pages per day across Mediterranean archives from Simancas to Dubrovnik. He later claimed to be “the first user of microfilms” for scholarly historical research.4. His wife Paule spent years reading the accumulated reels through what Braudel called “a simple magic lantern.”5. Captured in 1940, he spent five years as a prisoner of war and wrote the entire first draft of The Mediterranean—some 3,000 to 4,000 pages—from memory. Paule, meanwhile, retained access to the microfilm and notes in Paris, and after the war, they reconstructed the text, taking his manuscript, verifying it and adding footnotes and references from the microfilm.6.

In 1945, the same year Braudel was liberated, Vannevar Bush published “As We May Think,” in which he imagined a device he called the “Memex”: a mechanized desk storing a researcher’s entire library, indexed and cross-referenced, expandable through associative trails.7. The vision remained speculative for decades. Now the world’s archives are being digitized; AI systems translate, summarize, and search across them in seconds and can translate any language. To take one example, earlier this year, I used Google’s Gemini to translate the Hierosolymitana Peregrinatio of Mikalojus Kristupas Radvila Našlaitėlis, a sixteenth-century pilgrimage narrative from an online scan of the Latin first edition. The result is not a polished scholarly translation, but a working text that allowed me to gain a good sense of a text that was previously unreadable to anyone without proficiency in Latin or Polish (the only language into which, to my knowledge, it had been translated). The role of the intellectual is being transformed—not replaced, but augmented in ways Bush could only sketch. This feels like something other than foam.

How to account for such a rapid shift? Manuel DeLanda offers one answer in A Thousand Years of Nonlinear History. Working in Braudel’s materialist tradition and drawing on Gilles Deleuze and complexity theory, DeLanda describes how flows—of trade, energy, and information—accumulate and concentrate until they cross a threshold, undergo a phase transition, radically reorganizing into a new stable state. But here is the key insight: intensification is la longue durée. The accumulation of flows that began with the Industrial Revolution—or perhaps with writing, agriculture, or even symbolic representation itself—is the deep structure behind our era. Steam, electricity, computing, the internet: each was a phase transition within a longer arc of intensification. Cities accelerate such processes, as Braudel showed, concentrating capital and labor until new forms of economic organization emerge—Venice, Antwerp, Amsterdam, London, each becoming sites at which the future arrived first. Such conditions are not opposed to la longue durée; they are the moments when intensification crosses a threshold.

The continued pace of change this year underscores that there has been no return to equilibrium. But this has been accompanied by unprecedented resistance to technology, appearing as simultaneous terror at its apocalyptic nature (in jobs, if nothing else) and dismissal as useless, especially in Gen Z. A January 2026 Civiqs survey found that 57 percent of Americans aged 18–34 view AI negatively—more than any other age group. Curiously, the seniors category, which now includes most boomers, was the least resistant to AI, followed by Gen X and older millennials, all groups that grew up seeing radical societal and technological changes.8. It seems paradoxical that the smartphone generation recoils from the tools of the future. To understand this resistance means understanding the mentalité that shaped it—what Braudel’s successors in the Annales school called the collective psychology formed through lived experience.9. For Gen Z, that formative experience was network culture—both a successor to postmodernism and a form of collective psychology I did not fully understand at the time. Writing on network culture in 2008, it seemed to me that social media promised connection; instead, it brought division.10. The networked self was indeed constituted through networks, not merely isolated in postmodern fragmentation, but the fragmentation was now collective. Networked publics built barriers against one another, creating what Robert Putnam called cyberbalkanization: retreat into a comfortable niche among people just like oneself, views merely reinforcing views.11. Identity wars and mimetic conflict flared across filter bubbles that amplified outrage and tribal scapegoating as both MAGA and wokism built toxic online cultures. QAnon and a thousand other conspiracy theories propagated through Facebook groups and YouTube recommendations. Young men drifted into incel communities where loneliness became ideology and livestreaming mass shootings was celebrated. Influencers built their empires on hatred—Hasan Piker framed Hamas’s October 7 massacre as anticolonial resistance while Nick Fuentes celebrated mass shooters as vanguards of race war and civilizational collapse.

Nor did this just fragment culture—it exacted a massive psychic toll, as social contagion spread new forms of self-harm and mental illness. During the pandemic, teenage girls began presenting tic-like behaviors—not Tourette’s syndrome, but something researchers termed “mass social media-induced illness,”12. spread by TikTok videos about Tourette’s rather than any actual disease. The pattern was unprecedented but not unique. Eating disorders spread through thinspiration hashtags. Self-harm tutorials circulated on Instagram. The platforms that were supposed to bring us together instead spread desires, disorders, and identities through pure social contagion—and with them, violence and polarization. A generation that grew up inside this experiment—that watched it reshape their peers’ bodies, minds, and identities—is right to be skeptical of the next technological promise.

In 2010, it seemed like network culture had a good chance of becoming understood as the successor to postmodernism. Bruce Sterling and I were engaged in a kind of dialogue about it online. He predicted that network culture would last “about a decade before something else comes along.”13. And he was right, as I acknowledged in my 2020 Year in Review. By then, network culture was exhausted, and with the Covidean break, it seemed time for something new. In 2023, I taught a course at the New Centre for Research & Practice to try to broadly sketch the emerging era. It’s still early and hard to fathom, like trying to understand postmodernism in 1971 or network culture in 1998, but it’s clear that if postmodernism was underwritten by the explosion of mass media, network culture by the Internet, social media, and the smartphone, then the current era is shaped by AI.

But if Gen Z, scarred by the effects of social media, has been reacting with deep fear and anxiety, Sterling how epitmozes the other reaction, dismissal. In the most recent State of the World, for example, he derides AI-generated content as “desiccated bullshit that can’t even bother to lie.” He compares the vibe-coding atmosphere to an acid trip, mocking the professionals who utter “mindblown stuff” like “we may be solving all of software” and “I have godlike powers now.” For Sterling, AI can produce nothing but slop. Now Bruce has always had a healthy skepticism toward tech claims, but I can’t help but think of Johannes Trithemius, the fifteenth-century abbot who wrote De Laude Scriptorum just as Gutenberg’s press was spreading across Europe—defending the scriptorium against a technology he could not see would remake the world.

There are even deeper, more existential fears, and I’ve spent the past year addressing them on my blog, in the process laying the foundation for a book on the topic: AI as plagiarism machine; AI as hallucination engine; AI as stochastic parrot, mindlessly repeating what it has ingested (Sterling’s critique); and AI as uncanny double, too close to us for comfort. As I explain, the discomfort arises not from the machine’s otherness but from its likeness: a mirror held up to processes we preferred to believe were uniquely ours.

It’s no accident that I published these essays on my blog. As far as my personal year in review goes, this was very much the year of the blog. I have no plans to ever publish in an academic journal again. Why would I? Who would read it? Why would I want to publish something paywalled, reinforcing the walled gardens of inequality that academia is so desperate to maintain—even as it proclaims itself the champion of open inquiry and democratized knowledge? Academia has become the realm of what Peter Sloterdijk called cynical reason: rehearsing the tropes of ideology critique while knowing the game is empty and playing it anyway. This revolts me.

But for almost ten years now, since the shutting down of the labs at Columbia’s architecture school, I have been content to write from the position of the outsider, something I reflected on in “On the Golden Age of Blogging”. That essay was prompted by a strange comment from Scott Alexander, who lamented on Dwarkesh Patel’s podcast that he had personally made a strategic error in not blogging during what he called the “golden age,” imagining that “the people from that era all founded news organizations or something.” The golden age he remembers is a fiction, as golden ages often are—and he gets the stakes entirely wrong. Evan Williams founded Blogger in 1999, sold it to Google, co-founded Twitter, then created Medium, which convinced hapless readers pay to read slop long before AI slop was ever a thing. The early bloggers who sought professionalization found themselves absorbed into the worst of the worst, writing for BuzzFeed, peddling nostalgia listicles that rotted psyches.

There was, however, a golden age for me, and I miss it: the architecture blogging community circa 2007—Owen Hatherley, Geoff Manaugh, Enrique Ramirez, Fred Scharmen, Sam Jacob, Mimi Zeiger (whose Loud Paper was less a blog and more a zine, but a key part of the culture), and others. We inherited from zine culture an informal, conversational tone and the will to stand outside architectural spectacle. But ArchDaily and Dezeen commercialized the form, shifting from independent critique to marketing and product. Startup culture absorbed architectural talent.

Blogging was powerful precisely because we had no stakes in it—we owned and controlled our means of intellectual production. The golden age of blogging is not in the past; it is now. After years of proclaiming I would blog more, in 2025, I really did. I wrote over 83,700 words on varnelis.net and the Florilegium—essay-length pieces on landscape, native plants, AI and art, architecture, infrastructure, politics, and tourism. My only regret is that my presidency at the Native Plant Society of New Jersey consumes so much of my thinking about native plants that little remains for writing. But the time will come, and if nothing else, my investigation of the Japanese garden aesthetic should point in the future direction for my writing on landscape.

I also continued to make AI art, or to be more precise, what I called stochastic histories. A major project was a substantial reworking of The Lost Canals of Vilnius, a counterfactual history in which, after the Great Fire of 1610, Voivode Mikalojus Radvila Našlaitėlis rebuilt the city with Venetian-style canals, complete with gondoliers, water processions, and a hybrid “Vilnius Venetian” architecture. As research, I used Gemini to translate Radvila’s sixteenth-century Latin pilgrimage narrative. AI, like photography or film, is what you make of it. Film is perhaps the better analogy—anyone can make a video. Making something worthwhile is another matter entirely. In December, I also completed East Coast/West Coast: After Bob and Nancy, a generative restaging of Nancy Holt and Robert Smithson’s 1969 video dialogue using two AI speakers.

There were other substantial essays, too. In “Oversaturation: On Tourism and the Image”, I finally put down on paper something I had wanted the Netlab to address while at Columbia, but that proved too dangerous for the school to support. Universities cannot critique the very systems of overproduction they depend upon for survival. Publish or perish and endless symposia nobody is interested in are the academic versions of overproduction, but more than that, any architecture school claiming global currency cannot afford to offend either other institutions, like museums, that give it legitimacy, or, for that matter, the trustees that fund both. As I point out, tourism has always been mediated by imagery; take Piranesi’s vedute or the Claude Glass. Grand Tourists always had representations at hand to interpret their direct experience—but a new crisis point has been reached with both overtourism and the overproduction of images. Algorithmic logic now reorganizes cultural geography around “most Instagrammable spots,” making historical significance secondary to content potential. The Fushimi Inari shrine in Kyoto is the case in point—a 1,300-year-old shrine that Instagram made famous and that has now ceased to serve as a religious site due to the influx of visitors. The Japanese have a term for this: kankō kōgai, tourism pollution. Tourism has become the paradigm of contemporary experience—the production of imagery without cultural meaning; everything feeds the same algorithmic mill. Even strategies of resistance get metabolized—slow travel becomes a hashtag, psychogeography becomes an Instagram guide.

The Bilbao effect, which was a major driver of oversaturation, was itself a product of globalization. Hans Ibelings coined “supermodernism” in 1998 to refer to the architectural expression of Marc Augé’s “non-places,” an architecture optimized for the perpetual circulation of bodies and capital. It was the architecture of network culture, of the Concorde and the Internet. Koolhaas diagnosed its endgame in his 2002 “Junkspace“—”Regurgitation is the new creativity”—and then, tellingly, stopped writing. Today, network culture is long gone; nationalism is on the rise. The Internet is a dark forest now14. while the disconnected life is on the rise.15 The most exclusive resorts now advertise no Wi-Fi, no cell service, no addresses—only coordinates. Disconnection has become the ultimate luxury, sold back to the same people who built the infrastructure of connection. More cities are alarmed by the effects of overtourism than desire to attract tourists. In the US, new architectural proposals appeal to a retardataire aesthetic—Trump displaying models of a triumphal arch inspired by Albert Speer and marking a triumph of nothing in particular in models in three sizes (“I happen to think the large looks the best“), a four-hundred-million-dollar ballroom modeled on Mar-a-Lago, an executive order mandating classical architecture for federal buildings that Stephen Miller explicitly framed as culture war.

Yet both Bilbao and MAGA are spectacle, architecture-as-branding. But the Bilbao effect is imploding. No city believes anymore that a signature building by a starchitect will transform its fortunes. The parametricists have nothing left to say. Parametric design promised formal liberation—responsive, site-specific, computationally derived—but what it delivered was the most efficient, ugliest box. If the promise was the blob, the reality is the “5-over-1”: wood-frame residential floors stacked on a concrete podium with ground-floor retail, wrapped in a pastiche of brick veneer, fiber cement panels, and that obligatory conical turret element meant to signal “we thought about this corner.” As for AI-generated architecture, it is merely boring—giant sequoias hollowed out as apartment buildings, white concrete towers with impossible cantilevers, and lush vegetation sprouting from every surface—the same utopian fantasy rendered a thousand times over. These are renders of renders: AI trained on architectural visualization produces visualizations that are utterly disconnected from any tectonic reality. A new generation may emerge in response to new needs, but for now, the discipline has lost its cultural purchase. Architecture, for us, is a thing of the past.

The art world, too, has slowed. Museums are putting on fewer shows, shifting from aggressive schedules to longer, more deliberate exhibitions—or simply cutting programming as budgets tighten.16. The frantic pace of the Biennale circuit has exhausted dealers and collectors alike; smaller fairs are folding, and even the major ones feel like obligations rather than events. Galleries that survived the pandemic are now closing quietly, without the drama of a market crash—just a slow bleed of foot traffic, sales, and cultural attention. There is no new movement, no emergent critical framework, no sense of direction. The market churns on—auction prices for blue-chip artists remain high, collectors still speculate, art advisors still advise—but the sense of cultural mission has dissipated. What remains is commerce without conviction, a field that has forgotten why it exists beyond the perpetuation of its own economy. The institutions that trained artists for this field are collapsing alongside it.

As enrollment dwindles, design schools are collapsing—not merely contracting, but ceasing to exist. Most recently, the California College of the Arts announced in January 2026 that it would close after the 2026–27 academic year17., the last remaining independent art and design school in the Bay Area. It follows a grim procession: the San Francisco Art Institute (2020), Mills College (2022), the Pennsylvania Academy of the Fine Arts (2023), and Woodbury University’s acquisition by Redlands and subsequent adjunctification—a fate that has methodically undone so many schools as faculty become contingent labor and institutions into hollow administrative structures run by well-paid, cost-optimizing consultants.

There is personal resonance for me in this. Simon’s Rock College of Bard, which shuttered its Great Barrington campus in 2025, was where I studied for my first two years before transferring to Cornell—a pioneer of early college education that offered a radical pedagogical experiment in what learning could be beyond conventional schooling. I arrived there straight from high school, as did my good friend and colleague Ed Keller; clearly, something interesting was in the water back then. Simon’s Rock made the development of young minds its central mission rather than an incidental focus of brand management or endowment growth, and its alumni list is impressive for such a small school. It has an afterlife at Bard, but it’s an echo at best.

The difference between these institutional deaths and simple market failure is this: they are not being replaced. When a retail business fails, another may open elsewhere. When a school closes, there is no succession. The market offers no alternative. Instead, what remains are the corporate university satellites—for-profit programs nested within larger institutions (like Woodbury’s absorption into Redlands), stripped of autonomy, their faculty reduced to precariat, their curricula bent toward what can be measured and marketed. The art schools that survive do so by transforming into something else: luxury finishing schools for wealthy families or research appendages to larger universities, where “design thinking” becomes another management consultant’s tool. The pedagogical mission—to create conditions where students might develop serious aesthetic judgment, where they might encounter genuine problems and be forced to think through them—is not merely challenged but impossible. The closure of these schools does not signal a failure of art education; it signals that the very idea of art education as something valuable in itself has been liquidated.

This hollowing out of cultural institutions is not incidental to the political moment—it is one of its hallmarks. Politically, most people have checked out. This is not 2017, when each provocation demanded a response; the outrage cycle has given way to numbness. In “National Populism as a Transitional Mode of Regulation”, I argued that Trump, Orbán, Meloni, and their ilk represent not a return to fascism but something new: the authoritarian management of declining expectations. National Populism correctly identifies that neoliberalism’s promise of shared prosperity has failed, but it channels legitimate grievances toward scapegoats rather than addressing the technological displacement actually causing them. This is its tragic irony: the National Populist base—workers made obsolete by neoliberalism and unable to participate in AI Capitalism—finds its legitimate anger directed into a movement that accelerates the very forces rendering them superfluous. Their value to capital lies in political disruption rather than economic production; they are consumers and voters, but no longer needed as workers. National Populist leaders offer psychological compensation—dignity, recognition, transgressive identity politics—rather than material improvement. The apocalyptic tenor of populist culture, its end-times thinking and conspiracy theories, provides a framework for populations sensing their own economic redundancy.

The alliance between tech billionaires and populist leaders is unstable. AI Capitalism requires borderless computation and global talent flows; nationalist protectionism contradicts these at every turn. Musk, Thiel, and Andreessen have aligned with the movement to dismantle the regulatory state, not because they share its vision but because populism serves as a useful battering ram against institutional constraints. Once those barriers fall, the movement and its human-centric concerns can be discarded. National Populism, as I conclude, is not the future—it is a political interlude, a transitional mode that will not survive contact with the economic forces it has helped unleash.

If National Populism is transitional, is there a positive vision that can replace it? In “After the Infrastructural City”, I responded to Ezra Klein and Derek Thompson’s book Abundance, perhaps the most influential book of 2025, which argues that America’s inability to build is a political choice, not a technical constraint. Their solution: streamline regulation, invest boldly, build more. It’s a compelling vision—and a necessary corrective to decades of paralysis. But Abundance shares a curious blindspot with Muskian pronatalism: both assume we need more people. Musk preaches that declining birthrates spell civilizational collapse; Klein and Thompson build their vision on populations that will mysteriously arrive to fill what’s built, perhaps by immigration. Neither accounts for the possibility that AI changes the equation entirely—that a smaller population, augmented by intelligent systems, might not be a crisis at all. Populations are already shrinking across much of the developed world. What I call “actually-existing degrowth”—not the voluntary eco-leftist kind, but the unplanned demographic contraction now underway in Japan, Korea, and much of Europe—is coming for the United States too. Declining birth rates, aging populations, and regional depopulation: these are not future scenarios but present facts.

This doesn’t invalidate the Abundance agenda; it redefines it. Abundance cannot mean building more for populations that will not arrive. It must mean building better, adaptive, intelligent infrastructure for smaller, older societies. AI, rather than merely destroying jobs, can help navigate this transition: smart grids, autonomous transit, predictive healthcare. The opportunity is real. Managed shrinkage, done well, can mean more livable cities, restored ecosystems, higher quality of life. The question is whether political leaders can articulate a vision of flourishing within limits—or whether nostalgia for growth will leave us building for a future that never comes.

Against the exhaustion of institutions, against the hollowing out of architecture and art, against the closure of the schools that trained people to imagine, the blog remains. It may not be much, but it is one independent voice outside the collapsing structures around me. I wrote over 83,000 words this year. I made art. I thought through problems that matter to me with the help of AI, which provided me with tools I could only have dreamt of merely a year ago. Today, I uploaded hundreds of thousands of words from my essays to a directory in Obsidian so that Claude could draw connections between them (see here for just how one can set this up).

The future is already here—it just isn’t evenly distributed. Some are afraid or are still pretending AI isn’t happening. Phase transitions are uncomfortable. They are also where the interesting work gets done. One makes of one’s time what one makes.

1. William Gibson, quoted in Scott Rosenberg, “Virtual Reality Check Digital Daydreams, Cyberspace Nightmares,” San Francisco Examiner, April 19, 1992, Style section, C1. This is the earliest verified print citation, unearthed by Fred Shapiro, editor of the Yale Book of Quotations.

2. Fernand Braudel, The Mediterranean and the Mediterranean World in the Age of Philip II, trans. Siân Reynolds (New York: Harper & Row, 1972), 21.

3. Braudel, The Mediterranean, 901.

4. Fernand Braudel, “Personal Testimony,” Journal of Modern History 44, no. 4 (December 1972): 448–67.

5. Paule Braudel, “Les origines intellectuelles de Fernand Braudel: un témoignage,” Annales: Histoire, Sciences Sociales 47, no. 1 (1992): 237–44.

6. Howard Caygill, “Braudel’s Prison Notebooks,” History Workshop Journal 57 (Spring 2004): 151–60.

7. Vannevar Bush, “As We May Think,” The Atlantic Monthly 176, no. 1 (July 1945): 101–8.

8. Civiqs, “Do you think that the increasing use of artificial intelligence, or AI, is a good thing or a bad thing?,” January 2026, https://civiqs.com/results/ai_good_or_bad.

9. The concept of mentalités emerged from studies of phenomena like the witch trials, where beliefs and fears spread through communities in ways that could not be reduced to individual irrationality. For an overview of mentalités as a historiographical concept, see Jacques Le Goff, “Mentalities: A History of Ambiguities,” in Constructing the Past: Essays in Historical Methodology, ed. Jacques Le Goff and Pierre Nora (Cambridge: Cambridge University Press, 1985), 166–180.

10. Kazys Varnelis, “The Rise of Network Culture,” in Networked Publics (Cambridge: MIT Press, 2008), 145–160.

11. Robert Putnam, “The Other Pin Drops,” Inc., May 16, 2000.

12. Kirsten R. Müller-Vahl et al., “Stop That! It’s Not Tourette’s but a New Type of Mass Sociogenic Illness,” Brain 145, no. 2 (August 2021): 476–480, https://pubmed.ncbi.nlm.nih.gov/34424292/.

13. Bruce Sterling, “Atemporality for the Creative Artist,” keynote address, Transmediale 10, Berlin, February 6, 2010.

14. Yancey Strickler, “The Dark Forest Theory of the Internet,” 2019, https://www.ystrickler.com/the-dark-forest-theory-of-the-internet/. See also The Dark Forest Anthology of the Internet (Metalabel, 2024).

15. “Trend: Not Just Digital Detox, But Analog Travel,” Global Wellness Summit, 2025, https://www.globalwellnesssummit.com/blog/trend-not-just-digital-detox-but-analog-travel/.

16. “The Big Slowdown: Why Museums and Galleries Are Putting on Fewer Shows,” The Art Newspaper, March 10, 2025, https://www.theartnewspaper.com/2025/03/10/the-big-slowdown-why-museums-and-galleries-are-putting-on-fewer-shows.

17. California College of the Arts, the last remaining private art and design school in the Bay Area, announced in January 2026 that it would close after the 2026–27 academic year. See “‘Nowhere Left to Go’: As California College of the Arts Closes, So Does a Pathway for Bay Area Artists,” KQED, January 13, 2026, https://www.kqed.org/news/12070453/nowhere-left-to-go-as-california-college-of-the-arts-closes-so-does-a-pathway-for-bay-area-artists.

On the Golden Age of Blogging

In the April 3 episode of the Dwarkesh podcast, Scott Alexander (of SlateStarCodex fame) lamented: “I am so mad at myself for missing most of the golden age of blogging. I feel like if I had started a blog in 2000 or something, then—I don’t know, I’ve done well for myself, I can’t complain—but the people from that era all founded news organizations or something. I mean, God save me from that fate. I would have liked to have been there. I would have liked to see what I could have done in that area.”

As someone who participated in it directly, the reality of early blogging bears little resemblance to Alexander’s mythologized version. Still, prompted by Alexander’s romantic musing, I thought it worthwhile to spend a few days engaged in a kind of digital archaeology, excavating three years of my early blog posts (2000–2003) from an old hard drive, then converting and uploading these posts to WordPress. So this is not a single post, it’s 377 posts. All of these—save, of course, this one, another 8,000-word long-form monster that took three weeks to write, the sort of thing a 57-year-old produces looking back at days that, in retrospect, seem a whole lot better than they were—are brief, a link with a sentence or two of commentary, nothing more. This sort of blog—and indeed all blogs—evolved from “What’s New” pages, which were standard features on early websites. NCSA (National Center for Supercomputing Applications)’s site, maintained by Mosaic (the first real web browser) developer (and present-day Venture Capitalist and right-wing extremist) Marc Andreessen included such a “What’s New” page, tracking new sites appearing online from June 1993 to June 1996, cataloging everything from fan pages on the TV show The Prisoner hosted at the University of Limerick to the South Bay Ski Club.

These early proto-blogs were remarkably minimal: brief updates, link collections, and fragmentary observations rather than the developed essays that define contemporary blogging. Justin Hall’s online daily diary, launched in December 1994 when Hall was still a high school student, pioneered personal disclosure online, creating an intimate window into his life that differed significantly from the curatorial approach of other early web pioneers. Justin’s diary was something of a real-life digital Dawson’s Creek, captivating for many readers.

In contrast, sites like Dave Winer’s Scripting News and John Barger’s Robot Wisdom (both in 1997) functioned more as a form of Internet DJing: brief observations, often snarky, sometimes just a link with a one-line comment—a far cry from the long-form posts that characterize contemporary blogging. Robot Wisdom, running on Winer’s complicated Frontier’s NewsPage software and where Barger coined the term “weblog,” exemplified this DJ mentality with its eclectic collection of links and brief commentary. Just as I would do, these early bloggers functioned primarily as curators rather than producing sustained arguments. The blog post as we know it today—a self-contained essay with developed arguments and sustained reflections—was still years away from emerging as the dominant form.

What united these diverse early web pioneers—from Hall’s confessional diary to Winer and Barger’s curatorial platforms—was a distinctive rhetorical approach that diverged sharply from institutional communication. They (We) employed an informal, conversational tone that established a direct relationship with readers—creating the sense of accessing an insider’s perspective. “Snark”—that knowing, ironic distance that signals the writer’s superior understanding of a situation—became a hallmark of this emerging style. This rhetorical strategy—common to Zines and to the Punk aesthetic that ruled the counterculture at the time—positioned the writer as a savvy guide cutting through official narratives to reveal a more authentic, personal truth. Unlike traditional media’s affected neutrality, these early web writers embraced subjectivity, making their positions and preferences explicit, often with cutting wit and cultural references that their intended audience would have understood.

Blogging was, above all, personal. So, with that in mind, permit me a digression into my personal experience and how it paralleled the experience of blogging.  As a high school student in the early 1980s, I sold articles about programming the Commodore VIC-20 to magazines like Compute! and Creative Computing, and though these pieces never saw print, they provided crucial income for a teenage programmer. I was involved in BBS culture in the early 1980s, but only as an occasional user: my parents weren’t about to let me have a modem on their precious phone lines. During my early college years at Simon’s Rock (1984-1986), I ran a one-man software company, writing an expansion to the BASIC programming language that gave the VIC-20 the ability to draw rudimentary graphics such as circles and lines. I also wrote code for a friend’s software company that used UNIX machines. At the same time, I took a course in journalism with Andrew Pincus, journalist and music reviewer at the famous Berkshire Eagle. I was fascinated by the New Journalists Andrew introduced us to—Truman Capote to Tom Wolfe to Hunter S. Thompson—and the idea of saying utterly unacceptable things in print. It got me called into the Provost’s Office more than once. Unfortunately, when I joined the History of Architecture program at Cornell, I had less time for either tech or journalism. Compounding this, Cornell Information Technology standardized on the Macintosh campus-wide and while the Mac had a revolutionary interface, part of that magic was that it was deliberately difficult to program; the first Mac was intended as an appliance for professionals, not a hobbyist’s computer and was priced accordingly, costing $2,495 ($7,600 in 2024 dollars). This began a long hiatus from coding that only ended in the 2010s when I started to integrate tech and art. Nevertheless, appliance or not, I was fascinated by the emerging technologies of the day. At Cornell, I was able to use the Internet to regularly send e-mail to friends at other campuses and browse USENET for the first time, a natural evolution of my interests. I remember using Fetch, an FTP program to download software from the Funet software repository in Finland in 1989, marveling that my keystrokes were making a hard drive head halfway across the world seek some obscure sector I had selected.

In the summer of 1993, I witnessed firsthand the potential of the World Wide Web through an early version of the NSCA-Mosaic browser running on a SUN workstation. The Honolulu Community College’s “Dinosaurs in Hawaii” exhibit—one of the first fifty websites ever created—left me stunned with its (now primitive) integration of imagery and text on a page—more like a book or magazine as opposed to a command line interface or the interface of the competing Gopher system, something of a hybrid between a Hypercard stack and a set of file directories. My initial explorations were using the primitive MacWWW browser both at Cornell’s public computer labs and via the 14.4k US Robotics “Sportster” modem I had in my apartment, but once NCSA released the Mac beta of Mosaic in September 1993, I adopted it immediately, even earning a mention in the first Mac press coverage of the browser’s release.

First mention of the Web in the Mac press, Macweek, October 11, 1993

By spring 1995, I had established my own hand-coded HTML website, a painstaking endeavor in the dial-up era where each update or correction required a lengthy time uploading to the FTP site, followed by checking on not one, but many, browsers (Microsoft Internet Explorer lagged behind web standards and was notorious for always failing to render a web site correctly). It also hadn’t been easy to find a web host. Hosting Internet sites was not easy and until I subscribed to the local Internet service provider—Homer Wilson Smith’s Lightlink, I had no way to put up anything on the Internet.

My first site was not much more than a curated collection of online architecture and humanities resources, a specialized directory or landing page for links that I used and expected others would want to use as well, plus “The Lair of the Chrome Peacock,” a page dedicated to the use and maintenance of the La Pavoni line of espresso machines. Like most early web users, I visited “What’s New” aggregators such as MacInTouch, which gathered scattered news from across the Internet and provided brief editorial commentary daily starting in 1994.  This was, once again, the site proprietor as Internet curator/DJ, laying down a pattern of links and commenting on them daily. In the spring of 1996, I developed an idea for a web site titled “ARC * Wire,” “both a central clearinghouse for news on architecture and … a forum for new design work, theory, and criticism.”1. Not only would this site aggregate all new architecture links, I hoped it would also serve as a journal, even a successor to the 90s theory journal Assemblage, which was rumored to be finishing its run soon. I talked with some friends like David J. Lewis, then working at Princeton Architectural Press, and C. Greig Crysler, then editor of the Canadian architecture journal A/R/C about various ideas—including for a SPY Magazine style parody site called Cranked or 13th Floor, but again, the technical challenges were high, seemingly insurmountable.

Kazys.net, c. 1999

I brought that idea with me after joining the Southern California Institute of Architecture (SCI_Arc) faculty in 1996. In March 1997, I was invited to a meeting at SCI_Arc with Paul Petrunia, then an undergraduate architecture student at the school and a nascent web developer, together with the school’s publications team to discuss replacing SCI_Arc’s static website—just a single image with a few footer links, not even utilizing the basic imagemap functionality common at the time. I proposed a revamped version of ARC * Wire called SCI * Arc * Wire (I uncovered the original proposal, time-stamped March 22, 1996, and posted it here) to transform the site into what would have not only been the first architecture blog but also a dynamic news platform that would have positioned SCI_Arc as not just a physical institution but as a global digital nexus for architectural discourse. My concept extended beyond mere updates to include a comprehensive archive of faculty works as well as invited essays in a journal format, effectively creating an authoritative online architectural publication establishing the school as an intellectual leader in digital space, rivaling Columbia’s GSAPP and the Architectural Association’s Design Research Laboratory. The school dismissed the idea and Paul agreed that it was likely technically unfeasible, a reasonable assessment given the era’s technology—it hadn’t worked out for me the previous year with ARC*Wire. Yet the concept seems to have resonated with him; that summer, he launched Archinect, which became architecture’s preeminent online hub, albeit without the academic focus I had intended for SCI * Arc * Wire. It was for the better, if the school had done it, the high degree of polarization there would have likely led to friction, controversy and an early dismissal from my job.2.

This pattern of dismissed ideas finding validation soon after became a recurring theme throughout the early web era, where technical limitations often temporarily delayed inevitable developments. The term “weblog” entered common usage in 1997 through Barger’s Robot Wisdom before being abbreviated to “blog”—a linguistic compression that, despite its inelegance, captured the medium’s emphasis on immediacy and informality. The true democratization of blogging arrived with Pyra Labs’ launch of Blogspot (now Blogger) in August 1999, which automated much of the technical drudgery that had previously limited participation. I launched this site in April 1998 as kazys.net. It was still a collection of links, but some of those links were now links to my essays. In May 2000, however, I successfully integrated Blogspot’s functionality into my site, marking my formal entry into what journalists were now calling the “blogosphere.”

In retrospect, I’m not entirely sure what my goals were in starting to blog, and this lack of clarity may have been a strength. Blogging represented the future, and like many others caught up in the era’s technological optimism, I wanted to participate in whatever that future might bring. Unlike many contemporaries gaining attention through confessional posting, I maintained boundaries between my intellectual work and personal life. I conceptualized my blog not as a diary but as a living archive of intellectual inquiry, a public workshop for developing ideas without academic publishing’s formal constraints. Instead, I conceived of the blog as a rolling record of intellectual preoccupations, a space for working through ideas in public without the pressure of formal publication.

Los Angeles became the central subject of many of these early posts,  functioning as my laboratory for understanding emerging network structures. These entries—composed during brief intervals at home, at SCI_Arc’s computer lab, or via my PowerBook’s WiFi PCMCIA card between teaching obligations—formed an analytical portrait of a metropolis in transition. Los Angeles was evolving beyond its traditional identity defined by automobiles and entertainment into something more complex: a critical node within expanding global networks of capital, information, and culture. Through blogging, I could document this transformation in real time, without waiting for the slow machinery of academic publishing to validate these observations.

By 2003, my relationship with blogging had changed. The birth of Viltis, our first child, coincided with growing disillusionment with SCI_Arc, which under Director Eric Owen Moss had begun embodying the very architectural culture it once critiqued—becoming increasingly hierarchical, spectacle-obsessed, and hostile to critical reflection. Watching the institution drift from its radical origins toward conventional celebration of architectural celebrity and the ethical impropriety of working for the proto-Trumpian school director ultimately prompted my resignation in spring 2004. At the same time, blogging had become a bit of a chore. Who and what was I doing it for? I needed to take a break and I suspended the blog without notice—perhaps without knowing—at the end of July. Robert Sumrell and I worked on our growing radical architecture project AUDC, I enjoyed working more with the Center for Land Use Interpretation, became the President of the Los Angeles Forum for Architecture and Urban Design, a group that had the energy increasingly lacking at SCI_Arc, and in what free time I had, I plotted my next steps.

In fall 2004, I started Simultaneous Environments, a project mapping Los Angeles’s transformation from 1990 to 2005—charting its evolution from a city primarily defined by automobiles and Hollywood to one increasingly shaped by networked systems. This research eventually evolved into The Infrastructural City, which reframed Los Angeles through the lens of its exhausted infrastructures and stalled futures, while exploring potential responses to these conditions.

The next spring, Detlef Mertins invited me to spend the semester at the University of Pennsylvania, where my final attempt at securing a traditional history of architecture position collapsed under the weight of academic politics internal to the school. I redirected my interests toward more productive terrain. I spent the following year at USC’s Annenberg Center for Communications directing the Networked Publics project, running a team of scholars to investigate a topic that seemed very much of the moment. Recognizing our need for a collaborative platform, I taught myself the Drupal content management system in order to develop networkedpublics.org as a group blog. Knowing that this would be a challenge, I relaunched my personal site on the platform in May 2005.

By the time I returned to blogging, the architectural blogosphere had begun to take shape. Blogging offered a fundamentally new model of architectural discourse: informal, distributed, and networked writing operating outside established media channels. This emerging form functioned at a different cadence and register than traditional publishing—more immediate, conversational, and liberated from institutional gatekeepers. As disillusionment with conventional architectural criticism grew, the democratized access to publication allowed independent, critical voices to flourish with unprecedented speed.

Hyperlinking, the very core of the hypertextuality in HTML served as the structural foundation of this architectural blogosphere. My site, alongside platforms like Archinect and things magazine, utilized linking not merely as citation but as a fundamental architecture—creating pathways connecting architectural theory with digital media, technology, urbanism, and cultural critique. These hyperlinks functioned simultaneously as source documentation and conversation, weaving disparate topics into coherent networks. This interconnected structure mirrored the networked urban conditions many of us were analyzing—a digital form following intellectual function. RSS (Really Simple Syndication) had emerged (Aaron Swartz—whose death will always haunt MIT and JSTOR—was a co-developer of RSS), allowing individuals—or sites—to subscribe to feeds from multiple sites. With the release of the elegant NetNewsWire reader for the Mac in 2002, it became possible to have an aggregator that one could use to browse the news across multiple sites on one’s own desktop. Released in October 2005, Google Reader dominated the RSS market—even being used to provide aggregated feeds for desktop apps like Reeder, which became popular after NetNewsWire seemed to grow stale.

This networked structure of thought mirrored the networked urban conditions many of us were discussing—a form following intellectual function. By 2007 a vibrant architectural blogosphere had formed: Owen Hatherley’s Sit Down Man, You’re a Bloody Tragedy, John Hill’s A Daily Dose of Architecture (which actually originated in 1999 and is thus one of the first architecture blogs), Mimi Zeiger’s online extension of her legendary late-90s zine Loud Paper (the blog launched July 2007), Dan Hill’s City of Sound, Bryan Finoki’s Subtopia, Geoff Manaugh’s BLDGBLOG, Alexander Trevi’s Pruned, Enrique Ramirez’s aggregät 4/5/6, Adam Greenfield’s Speedbird, and Molly Steenson’s girlwonder all offered distinctive voices united by a skepticism toward the dominant architectural critics such as Nicolai Ouroussoff, Christopher Hawthorne, and Paul Goldberger, whose work primarily celebrated the “Bilbao Effect” and its proliferation of iconic buildings by celebrity architects.

Our resistance to architectural spectacle was collective, even if our tactics diverged. Owen pressed for a return to the egalitarian, even Communist, ethos of Britain’s post-war housing—what he later codified as Militant Modernism. Dan, Adam and I argued that real urban change would flow through broadband pipes, mobile networks, and APIs the soft circuitry governing streets more decisively than steel or concrete. Mimi channeled her Loud Paper zine into an online pamphlet that celebrated guerrilla exhibitions, pop detours, and what would later be called “tactical” interventions. Bryan compiled a field guide to military urbanism, tracking how border walls and carceral logics seep into everyday space. Geoff spun speculative tales of sinkholes, weather warfare, and underground worlds, recasting design as imaginative fiction. Enrique crafted deeply researched cine-spatial histories of oceans, flight, and media. Together we dismantled the Bilbao-era starchitect narrative from a dozen directions—left politics, network urbanism, zine culture, data infrastructure, martial geographies, landscape futures, and cinematic atmospheres. Critique thrived in polyphony.

In 2007, Lebbeus Woods launched his blog, an acknowledgment that the scene had become socially acceptable, even for established figures in architectural discourse. Lebbeus’s blog instantly rose to the top and he kindly engaged with the many students and young architects who were his fans until shortly before his death in 2010. And, unlike most avant-garde architects of his generation, he shared the anti-establishment ethos of the younger blogosphere. In writing of Jean Nouvel winning the 2008 Pritzker Prize, for example, he wrote (this is the April 1, 2008 blog post “Sic Transit Gloria,” in full):

Jean Nouvel has won the 2008 Pritzker Prize for Architecture. No surprise. The lack of surprise makes it is easy to view the Pritzker as establishment laurels for those who are already well-established. Like the Nobel Prizes, it is conferred on safe, already certified choices. Nouvel’s buildings are certainly of a high quality of design. At his best, he designs beautiful buildings. Who can quarrel? The Pritzker tries to remind us that the design of beautiful buildings within, or maybe—on occasion—closer to the limits of the accepted canon of beauty is the ultimate goal of architecture. People need enclosed spaces, and it is up to architects to design them in ways that satisfy the needs of body and soul. Or, at least, in ways that reassure us about what we already know. Nouvel works masterfully within the limits of what we already know.

So, is there a problem with any of this? Not at all. I say, let the rich bestow upon the famous whatever they like. Let the rituals of power play themselves out as they always have. It is quite a seductive spectacle. We all become part of it, say, by posting comments like this on blogs.

But I have to question how relevant the Pritzker Prize is for the expanding world of architecture. Whatever its claim to reward innovation and expand discourse, I would say not very. Its focus on buildings, and often expensive buildings, leaves out much of the most innovative work going on in the field today, by younger architects making smaller-scale projects, or experimental ideas that never get off the boards or out of the computer—ideas that get published and change our ways of thinking about what architecture is and can be. The Pritzker sends the message that unless one builds, and in a spectacular way, one will never qualify for “architecture’s top honor.” The catch-22 here is that to build you need clients, and to build spectacularly, like Nouvel, very rich clients, and they are seldom willing to risk sponsoring the genuinely new. So, the subliminal message is, don’t push the envelope too far.

The existence of the Pritzker reminds us that the powerful are not as self-assured as they like to appear. They need to engage continually in demonstrations of their power, such as getting on—in an upbeat way—the front page of the New York Times, as well as other major newspapers and magazines around the world. Oddly enough, the Times perennially rails against the Nobel Prizes, not least because its founder was an armaments manufacturer who bought respectability in posterity by creating prizes in his name for intellectual achievement. And it works. When we think of Alfred Nobel, his name becomes synonymous with Albert Einstein, Samuel Beckett, Toni Morrison, Martin Luther King, Jr. The Pritzker is sponsored by the Hyatt Foundation, not exactly “merchants of death.” Still, the formula works. By associating themselves with successful architects and the world of creative thought, they make a significant step up in cultural, and historical, terms. Their domain, and their power to affect the world, is extended and consolidated.

For the recipients of the Pritzker, it’s easy enough to understand why they would accept it, often with speeches of praise for the Prize and its sponsors. The money may be relatively paltry ($100,000, compared with the Nobel’s $1,500,000), but every bit helps and, after all, why not? What’s the harm? Only two have ever declined the Nobel, on principle: Sartre and Le Duc. So far, no architect has declined the Pritzker. If that were to happen, it really would ‘expand the discourse.’

All this we already know. So, why bother to write about it? Perhaps only to step back from the spectacle long enough to see its contour, and its limits. Only then is it possible to see its true place in the order of things, and the wider world that lies beyond.

LW

Unfortunately, at about this time, big, commercially-oriented blogs were established—notably ArchDaily and Dezeen. Part blog, part traditional media platform, these were professionalized, polished, and fluff-heavy, frequently aligned with marketing, project announcements, product placement, and sponsored content, devoid of independent voices, but perfect for the market. Their emergence marked the beginning of the transformation of architectural blogging from a space of critique to one of promotion.

The first discussion about blogs in architecture appears to be an April 2009 MIT panel I shared with Javier Arbona and Mark Jarzombek, titled “Blogitecture, or Architecture on the Internet.” I began my talk with a moment of reflection that holds up 16 years later: “I’m going to go back before the origins of architectural blogging … to the origins of the web and point out that I was there at the right time at many of these places and managed to miss every opportunity (almost).” At that moment, blogging had reached a crossroads—caught between its early promise of democratized critique and its emerging entanglement with the very spectacle it sought to evade. I noted that RSS feeds, comments, and trackbacks (in which one blog would notify another when it referenced it) broke the boundary between author and audience and transformed blogging into “a switching machine for data flows”—the Internet DJ now had the capacity for near-real-time feedback. This was a genuinely new form of writing and created the possibility of real back-and-forth dialogue among writers. Later that year I would participate in Helen Thornigton and Jo-Anne Green’s Networked: A (Networked) Book about (Networked Art) project, contributing a chapter to a book running on Commentpress, a fork of WordPress created by the Institute for the Future of the Book. The goal of this project being to create an open, interactive book in which readers could act as co-authors, leaving comments and creating a community-driven discourse while challenging the idea of the book as a static object.

Even as I celebrated the new technological transformation, however, I lamented that architectural discourse itself had slid from history to theory to criticism with both history and theory nearly extinguished while newspaper critics served mostly as publicists for what urban sociologist Harvey Molotch termed the “Growth Machine.” In an influential 1976 paper, Molotch described how local elites, developers, media, and institutions form coalitions to promote development regardless of actual public benefit—essentially viewing cities as machines for generating wealth rather than habitable environments.3. Architectural criticism had largely become subservient to this apparatus, celebrating spectacular buildings that served as capital magnets while ignoring broader urban conditions. Blogs, I felt, arrested that slide by giving dispersed voices a platform to challenge this hegemony, creating spaces for more critical, historically informed discourse outside the growth machine’s influence.

Javier, an architect and geographer, turned to statistics showing that bloggers were overwhelmingly affluent professionals; thus he explained, legal blogs flourished because lawyers possessed both the disposable time and the appetite for arguments about privacy and intellectual-property rights. Architecture blogs mirrored that class bias, he argued. and within architecture, authority followed a pyramid: sites like BLDGBLOGInhabitat, and Dezeen occupied the apex, their posts ricocheting through Archinect, Twitter, and Tumblr until even student blogs adopted the day’s talking points—such as open sourcegreeninnovation— which had passed uninspected from Silicon Valley into design studios. Javier suggested that a discourse on labor rights was missing in the architectural blogosphere (oddly enough, such a discourse was about to take off, but wound up as a movement centered around Yale University, a questionable place to start a workers’ revolution). Arbona displayed A Daily Dose of Architecture‘s sidebar of abandoned blogs—a digital potter’s field—and rhetorically asked why enthusiasm so often curdled into silence.

But of course, if we both engaged in the all-too-predictable discussion about privilege in blogging, we were speaking from within the academy, a place of institutional security, if not comfort, that gave us both a voice and an audience. Our criticisms of spectacle, technology, and capital unfolded within the medium of an academic conference, a ritualistic site we naturally didn’t dare question, since being in the belly of the beast makes one reluctant to prod it. Blogging had promised to disrupt such privileged circuits—indeed, for a time it appeared it might. Take the potter’s field of blogs; is there anything wrong with giving something a try and then giving up on it? Or the argument that blogs are only for those with sufficient leisure time to engage in writing and the wealth to access the Internet? Reading, bowling, knitting, and Marxist activism are all leisure time activities only those relatively well off, on a global scale, can engage in. Why not, then, blog if that gives you joy and community? When it was fun, blogging was part of everyday life, not a form of labor. That sense of camaraderie, of participating in a distributed conversation that felt vital and immediate, was its own reward.

We took steps to mature as a critical movement. Mimi Zeiger organized an underground discussion group called LGNLGN which included virtually all the names mentioned so far, save Lebbeus Woods, and we sought to find ways to professionalize and expand our reach. In the end, the most substantive outcome was the New City Reader, a broadsheet newspaper co-published by Joseph Grima and myself, edited by Alan Rapp leading a small editorial team, in 2010 as part of the Last Newspaper exhibition at the New Museum. This project represented an attempt to bridge the gap between online discourse and physical media, producing critical content that could circulate beyond the screen while maintaining the networked sensibility of blog culture.

Joseph Grima and Kazys Varnelis, editors, New City Reader, 2010.

But by the time the New City Reader was published, things had started to change. Just as the blogosphere reached its zenith of cultural influence around 2007-2008, social media began its ascent. Initially launched exclusively for Harvard College students in 2004, The Facebook opened up slowly, first to Stanford, Columbia, and Yale, then the whole Ivy League and downwards through the .edu hierarchy until it was accessible to the wider public in 2006. Dropping the definite article, Facebook reimagined online interaction through user profiles and explicit “friend” connections. Its defining innovation was the News Feed, an algorithmically curated stream introduced in 2006,  aggregating updates from a user’s network, eliminating the need to proactively visit individual websites.

The launch of Twitter in 2006—co-founded by Blogger creator Evan Williams—marked another pivotal moment. Twitter pioneered “microblogging” with its 140-character (later 280) updates designed for real-time sharing. While Williams conceptualized it more as an “information network” than a social platform, its impact on attention flows was profound. The short, snarky posts that characterized early blogging found a natural home on these platforms, particularly Twitter, which seemed tailor-made for the Internet DJ mentality. But critics like Alexandra Lange observed that most architects—which isn’t the same as most bloggers—predominantly used these platforms for self-promotion rather than substantive discourse, missing opportunities for deeper engagement with the cultural implications of their work.4. Tumblr (2007) offered yet another variation, blending short posts, images, and links in a format closer to traditional blogging but still emphasizing brevity and multimedia content. 

Facebook’s 2012 acquisition of Instagram for $1 billion marked a pivotal shift in online discourse, particularly for visually-oriented fields like architecture. Instagram’s image-first format fundamentally changed how architectural work was presented and consumed online. The platform’s emphasis on striking visuals, deemphasis on discussion, and its highly efficient distribution system accelerated what critics called “render culture”—where photogenic qualities often trumped practical considerations in architectural evaluation. For many architects and firms, Instagram became the primary platform for sharing work, with some projects seemingly designed more for their “Instagrammability” than spatial experience. This visual dominance particularly impacted architectural discourse, which had traditionally balanced textual critique with visual documentation. The “Like” button, introduced in 2009 further streamlined interaction, offering frictionless engagement but discouraging the substantive conversations that flourished in blog comment sections. These features created a compelling “walled garden” designed to capture and retain user attention within Facebook’s ecosystem. Facebook, Instagram, and Twitter replaced the blog for many. Indeed, with the need to keep up with all three, and perhaps Tumblr, there was less time for blogging. 

Discovery shifted from blogrolls and search engines to algorithmic feeds controlled by platform companies. Consumption habits changed dramatically, influenced by smartphones and mobile internet access, which favored bite-sized content consumed on the go. The Pew Internet & American Life Project documented a notable decline in blogging among younger demographics after 2006, coinciding directly with their adoption of social networking platforms.5.

Simultaneously, traditional architectural criticism started to collapse. The sycophancy of design criticism with regards to starchitecture, the speed and accessibility of digital platforms, a continuing decline in print readership, and a decline in newspaper finances caused by the rise of social media and Craigslist’s decimation of the advertising market contributed to less and less lengthy, rigorous critiques of the sort once found in print journals. Architectural analysis became increasingly fragmented, as the glossy, PR-driven sites Dezeen and ArchDaily took over. Unlike the independent blogs of the 2000s that frequently challenged dominant architectural narratives, these platforms largely reinforced them, prioritizing photogenic projects aligned with market trends, emphasizing aesthetic novelty and technological innovation disconnected from broader cultural or political contexts. This would soon be dubbed “Instagrammable architecture,” to refer to spaces or renderings conceived chiefly for shareable impact on an iPhone screen. Representation, especially in the schools, shifted to a pastel-toned isometric axonometry—a “post-digital” mode that fusing Archigram-style  cartoons with Monument Valley-like pixel art. A 30-degree grid, uniform line weight, and flat vector fills equalize vending carts, solar panels, and structural bays, casting buildings as candy-colored storyboards. Dramatic stairs, chromatic backdrops, and razor-sharp geometries proliferated less for spatial experience than for social media potential. What resulted was an architectural discourse increasingly dominated by visual consumption rather than critical reflection—a pattern that has only intensified with subsequent platform evolutions.

Architecture itself reached an impasse by the 2010s. The problems that animated the energetic neo-avant-garde architecture of the 1990s and early 2000s—the return to modernism, the blob vs. the box, the pursuit of iconic buildings to trigger the ‘Bilbao Effect’, an architecture for the creative city, and the exploration of computational design’s potential for generating new material expressions—had been mined to death and exhausted. When these problems were seemingly ‘solved,’ architecture as cultural inquiry didn’t advance to new territory but instead floundered in a conceptual vacuum. Worse yet, from a philosophical perspective, architecture’s traditional proper object—the shaping of collective space—was now subsumed by computation itself. If computation had promised to revolutionize architectural form, it now redirected collective experience toward virtual spaces. Where young people interested in both art and technology once went into architecture, now they were lured in not only by the profits but also by the transformational promise of the Internet startup. The cure turned out to be the poison. The new commons ceased to be the plaza or the museum atrium and was now the space of the network, mediated through screens rather than physical structures. People, it seemed, could now only conceive of architectural space by taking selfies of themselves on Instagram. It wasn’t merely that architecture reduced itself to Instagrammable imagery; it was that architecture conceptually disappeared on Instagram, becoming merely another category of consumable visual content rather than a spatial art with social agency. Unlike previous moments of architectural crisis that generated theoretical breakthroughs, this exhaustion produced no corresponding illumination—only a sense of architecture’s increasing irrelevance to contemporary experience.”

Of course, as Hal Foster argues in his reading of the Freudian concept of Nachträglichkeit (deferred action) in The Return of the Real, cultural phenomena often operate through delayed recognition. The trauma or innovation registers initially without full comprehension, only to resurface later with renewed significance. Freud’s term suggests that meaning is assigned retroactively, as previously latent connections become visible. By this logic, the seemingly exhausted questions of architecture—how to shape collective experience, how to reconcile digital and physical realms, how to create meaningful intervention in increasingly privatized space—will likely one day return with renewed urgency under different conditions. But for now, architecture has entered a period in which it is for us, once again what Hegel would call “a thing of the past.” There is much to the idea of philosophy traveling through disciplines and to the recurrence of the death of architecture, but this digression is a matter for another essay.

Nor did starchitecture survive this period. According to Davide Ponzini, author of Starchitecture: Scenes, Actors, and Spectacles in Contemporary Cities, the symbolic significance of star architects declined markedly in the late 2010s. The death of iconic figures such as Zaha Hadid in 2016 further underscored this diminishing era, bringing to a close the dominance of a few celebrated architects like Frank Gehry, Norman Foster, and Rem Koolhaas. The cultural authority once held by these figures has significantly eroded, as critics increasingly challenged starchitecture for its extravagance, egocentrism, and inadequacy in addressing urgent environmental and social concerns. Ponzini suggests that these dramatic, spectacular buildings, often disconnected from their urban contexts, came to embody the contradictions and failures of urban development strategies that prioritized visual and symbolic impact over practical functionality, sustainability, and social inclusivity. Furthermore, as architectural discourse became fragmented across numerous digital platforms and diverse critical voices emerged, the unified cultural narrative and consensus that originally fueled starchitecture’s global prominence disintegrated.6.

But the blogosphere would not replace architectural criticism. On the contrary, the rise of social media was at the expense of the blog. Engagement on Facebook, Twitter, or Tumblr took far less effort and anyone could participate. In the words of Throbbing Gristle, “Move a fin and the world turns / Sit in a chair and pictures change”—no navigating to separate sites, no logging in, just dashing off quick replies with the potential for wider visibility. It was the greatest thing since television. And if social platforms initially drove significant referral traffic to blogs, they gradually optimized to keep users within their own ecosystems, shadowbanning content with external links, leading to measurable declines in external referrals to blogs over time.

In retrospect, the first sign of stress in the blogosphere was the relentless rise of comment spam. What had begun as a trickle in the early 2000s became a devastating flood by 2010-2013, with automated bots filling blog comment sections with word salad along with links to questionable pharmaceuticals and gambling sites. Comment sections, once the vibrant heart of blogging communities where authors and readers engaged in substantive dialogue, became battlegrounds between legitimate conversation and algorithmic pollution. Blog owners found themselves spending hours each week moderating comments or implementing increasingly complex CAPTCHA and authentication systems that created friction for genuine commenters. Services like Akismet emerged to filter spam but required constant updating to keep pace with spammers’ evolving techniques and were annoying to use. Many influential bloggers eventually made the painful decision to disable comments entirely, severing a core interactive element that had distinguished blogs from traditional publishing. This degradation of community interaction further pushed conversations toward centralized social platforms where identity verification and spam filtering were built into the infrastructure, accelerating the migration of both content creators and their audiences away from the independent blogosphere.

A severe blow to the blogosphere came in 2013 when Google shuttered Google Reader, its popular RSS aggregator. RSS (Really Simple Syndication) had been the technological backbone allowing readers to efficiently follow multiple blogs without visiting each site individually. Google’s decision, citing declining usage, eliminated the most widely-used tool for RSS consumption overnight. Bloggers reported massive subscriber losses—Andrew Chen famously lost nearly 100,000 subscribers instantly. The infrastructure supporting independent blog discovery and consumption had been significantly weakened, pushing more creators and readers toward platform-centric ecosystems. The RSS icon, once ubiquitous across the web, began its slow fade from prominence.

As social media platforms captured ever-larger shares of attention,  Medium, launched in 2012 by Evan Williams, represented a particularly insidious development—the co-founder of Blogger and Twitter—now building a platform that would ultimately absorb and commodify the independent publishing ecosystem his first creation had helped establish. Presenting itself as a savior of quality writing with its clean design and superior reading experience, Medium encouraged writers and entire publications to migrate their content, positioning itself as an alternative to maintaining independent blogs. But Medium was less a defender of blogging, more its corporate undertaker. After amassing sufficient content and attention, in 2017 Medium abruptly erected paywalls around content that had previously been freely accessible. Writers who had built audiences on the promise of open distribution suddenly found their work locked behind a $5/month subscription fee, with compensation tied to engagement metrics like “claps” and later reading time—a direct extension of the attention economy logic pioneered by social media platforms. This bait-and-switch tactic converted independent blogs into undifferentiated “content” feeding Medium’s subscription machine. Like Facebook and Twitter, Medium demonstrated that centralized platforms ultimately serve their own interests, not those of the creators who supply their content. The Medium experiment demonstrated the eagerness of Silicon Valley billionaires to capitalize subcultures without regard for the damage it would do. Rather than supporting a diverse ecosystem of voices with distinctive identities and direct audience relationships, Medium homogenized content under a single brand and aesthetic, reducing creators to interchangeable suppliers for its subscription business. Medium represented not evolution but extraction—a corporate attempt to capture the value of blogging while eliminating its independence and diversity.

Compounding all this was a political rift. By the time of the 2016 election, political polarization online had grown—not merely the right vs the left, but also an increasingly shrill progressive woke left that attacked liberals and centrists who were concerned, above all, that Trump not be elected. Social media, which of course played a huge role in this process, became the focus of attention and the vibrant conversations once hosted in the blogosphere degraded into bickering on sites, where interaction—however unpleasant—required less friction. The blogosphere splintered.  

But even amidst this narrative of decline, blogging continued, only less so within architecture now. While microblogging and social feeds captured mass attention, a distinct cohort of interconnected blogs—largely British Marxists and lovers of post-punk music—maintained and deepened the intellectual promise of the medium through rigorous, extended explorations of complex topics and longer form pieces. Mark Fisher’s k-punk, launched in 2003 was what Simon Reynolds described as “the central hub of a ‘constellation of blogs’ in which popular culture, music, film, politics, and critical theory were discussed in tandem by journalists, academics, and colleagues.”7. This loose collective included writers like Reynolds himself, Benjamin Noys with his “No Useless Leniency” blog, Matthew Ingram who ran Woebot, aforementioned architecture blogger Owen Hatherley as well as others engaged in sustained cross-blog conversations. Blogging, for this group, became a central site for developing theoretical concepts. Noys notably coined the term “accelerationism” as an object of critique, while Fisher explored the possibilities of “acid communism”—demonstrating how blogging could sustain serious intellectual work outside academic institutions, which they increasingly saw as precarious. Fisher’s blog, Ryan Meehan wrote, was a “as a release valve from the pressures of academic writing. It became an impassioned and authoritative node in the hyperactive mid-aughts blogosphere—a network that, viewed from the social media quagmire of 2018, seems romantically free.”7. Fisher’s explorations of concepts like “capitalist realism,” “hauntology,” and the “slow cancellation of the future” began as blog posts before evolving into influential books published on the Zero Books imprint. His suicide in 2017 left a void in online Leftist thought, but cemented his legacy as one of the most important cultural theorists of his generation.

Returning to Scott Alexander, his Slate Star Codex (later a substack called Astral Codex Ten), written anonymously until his identity was controversially revealed by the New York Times, became a central hub for the rationalist community and attracted a wider readership interested in science, medicine, philosophy, politics, and futurism.The blog’s long-form essays delved into often controversial subjects, employing detailed arguments and statistical reasoning in ways that some saw as intellectually rigorous and others viewed as a gateway for legitimizing fringe or socially harmful ideas. Alexander’s practice of noting the “epistemic status” of his posts and the blog’s highly engaged comment sections fostered a culture of rigorous discussion that stood in stark contrast to social media’s superficiality. Similarly, Gwern.net, the website of pseudonymous independent researcher Gwern Branwen, represented another facet of the long-form resurgence. Known for exceptionally deep, data-rich research on topics ranging from AI and machine learning to psychology and self-experimentation, Gwern’s distinctive style features extensive annotations, footnotes, and sidenotes, reflecting a meticulous research process and a commitment to comprehensive evidence. The site’s unique design prioritizes content accessibility while deliberately rejecting web bloat. Gwern, incidentally, claims to live off of about $12,000 a year in order to devote himself full time to his passion of blogging. Longform podcasts like Dwarkesh (Patel’s), which features in-depth conversations with thinkers on AI and related matters have helped introduce the idea of long-form thought with broader audiences, undoing the dopamine-heavy obsession with the social media post. In providing platforms for extended, nuanced discussions, these podcasts serve as complementary mediums to long-form writing, allowing ideas to develop and cross-pollinate across formats.

Starting in 2017, Substack began providing a simple, integrated solution for such writers to launch newsletters and charge subscriptions. The platform experienced rapid growth, significantly accelerated by the COVID-19 pandemic, reaching over one million paying subscribers by late 2021. This model built on the same core values that drove blogging’s initial success—the desire for independent expression, direct audience connection, control over one’s platform, and space to explore ideas in depth.

For me personally, this renaissance of long-form, thoughtful content and the ease of uploading long form posts to WordPress and Substack presents a new opportunity. Though the architectural blogosphere I was once part of has largely disappeared, many of the problems I tried to address then have reemerged with even greater urgency—the interplay of space and technology, the role of technology in our lives, and society’s relationship to the built and natural environment, topics that demand the kind of sustained, thoughtful engagement that quick social media posts can’t provide. As yet, I don’t see a longer-form architecture community taking shape, and I wonder if I would be part of it—although John Hill’s A Daily Dose of Architecture, now A Weekly Dose of Architecture Books is on Substack as is Enrique Ramirez, with his “A New Literary History of Architecture.” For his part Adam Greenfield continues to write regularly at Patreon.   

To conclude, let’s turn back to Scott Alexander’s musing about blogging’s “golden age” and his missed opportunity. While some early web pioneers like Marc Andreessen did indeed parlay their online presence into tremendous wealth and influence, others experienced a far different trajectory. Consider Robot Wisdom’s Jorn Barger, the coiner of “weblog.” Despite his pioneering status, Barger later found himself “homeless and broke” in San Francisco, “living on less than a dollar a day,” as reported by Wired magazine in 2005.8. The contrast is stark: the very man who named the medium couldn’t convert his innovation into financial stability, while others built empires. Given Andreessen’s recent behavior, it’s hardly a matter of mental stability; it’s the usual matter of getting lucky.

This disparity highlights how blogs were never primarily vehicles for wealth creation, institution-building, or an expected source of income for most of us. We did it for the hell of it. Alexander gets it all wrong by focusing on remuneration or professionalization. Blogging had—and once again has—a “golden age” precisely because we had no stakes in it. Blogs were significant not as business ventures but as independent voices outside traditional media channels. And that endures today. The true power of blogging is in ownership and control over one’s means of intellectual production. Those bloggers who sought professionalization or institutional validation—including in the academy— found themselves reabsorbed into existing systems. Pace Alexander, the most enduring legacy of blogging isn’t found in those who “graduated” to mainstream success but in those who maintained their independence, creating an alternative intellectual commons that operated according to different values than the market or the academy.

Blogging taught us that intellectual communities can form and flourish outside traditional institutions. My own journey from architecture blogging to explorations of native plant design and AI Art illustrates how these distributed conversations allow for intellectual evolution that might be constrained within more rigid disciplinary boundaries. The networked thinking that characterized the architectural blogosphere—its ability to connect seemingly disparate domains through hyperlinks and cross-disciplinary dialogue—offers a model for approaching the complex challenges we now face, from climate adaptation to reimagining collective space in a post-pandemic time.

As Mark Zuckerberg’s recent acknowledgment that “social media is over” suggests, the pendulum is swiftly swinging away from the algorithmic feeds and dopamine-driven engagement mechanisms that dominated the 2010s.9. Whether this is because individuals are occupying themselves in AI chatbots, just watching more streaming TV to escape, or are interested in longer form writing is, as yet unclear. Likely, it is all of these. Blogging, in some form, will endure for some time to come. Rather than a discrete historical moment, blogging is an enduring mode of independent intellectual engagement that persists despite evolving technological contexts. While architectural bloggers may have failed to transform architectural labor practices or fully democratize discourse, we nevertheless established alternative pathways for developing and sharing critical perspectives, if only temporarily. That achievement, modest though it may seem, represents a meaningful contribution to culture and maybe that’s enough.

1. Kazys Varnelis, Proposal for SCI * Arc * Wire, https://varnelis.net/sci-arc-wire/

2. Kazys Varnelis, Proposal for SCI * Arc * Wire, https://varnelis.net/sci-arc-wire/

3. Harvey Molotch, “The City as a Growth Machine: Toward a Political Economy of Place,” American Journal of Sociology, Volume 8, Number 2 (September 1976) , 309-332, https://www.jstor.org/stable/2777096.

4. Alexandra Lange, “Opinion, It’s easy to make fun of Bjarke Ingels on Instagram,” Dezeen, January 7, 2014. https://www.dezeen.com/2014/01/07/opinion-alexandra-lange-on-how-architects-should-use-social-media/

5. Amanda Lenhart, Kristen Purcell, Aaron Smith, and Kathryn Zickuhr, Social Media and Young Adults, Pew Research Center, Internet & American Life Project, February 3, 2010, https://www.pewresearch.org/internet/2010/02/03/social-media-and-young-adults-3/.

6. Davide Ponzini and Michele Nastasi, Starchitecture: Scenes, Actors, and Spectacles in Contemporary Cities (New York: Monacelli Press, 2016), 190-192.

7. Simon Reynolds, “Music & Theory,” Frieze, September 18, 2009, https://web.archive.org/web/20160304073857/http://blog.frieze.com/music_theory/.

8. Paul Boutin, “Robot Wisdom on the Street,” WIRED, July 1, 2005, https://www.wired.com/2005/07/robot-wisdom-on-the-street/.

9. Kyle Chayka, “Mark Zuckerberg Says Social Media is Over,” The New Yorker, April 23, 2025, https://www.newyorker.com/culture/infinite-scroll/mark-zuckerberg-says-social-media-is-over

National Populism as a Transitional Mode of Regulation

Like many of you, I have been trying to make sense of the current political landscape in the US and elsewhere, but I have also have been working on updating my argument about Network Culture—a historical phase marked by digital networks and neoliberal economics—and exploring what comes next. The result is this new essay on National Populism, not merely as a political ideology but as a transitional form emerging during capitalism’s shift from neoliberalism to what I term “AI Capitalism,” in which artificial intelligence becomes central to economic activity. There aren’t references to art or architecture here—those will come later, once the cultural outlines of this moment become clearer.

My interest is in how National Populism serves as a stopgap amid profound economic transformation. Populist movements promise restored manufacturing, national sovereignty, and cultural revival, yet in practice their policies accelerate financialization and deregulation, ironically undermining these promises. The tragic irony is that the National Populist base itself suffers the most: their legitimate grievances about economic dispossession are channeled into a movement that intensifies rather than resolves their obsolescence.

For two decades, beginning at SCI_Arc and concluding at Columbia’s Graduate School of Architecture, Preservation, and Planning, I taught a course called Network City. In that course, I employed the Regulation School’s framework of regimes of accumulation and modes of regulation to explore urban transformations accompanying the shift from Fordist mass production to post-Fordist flexible specialization. Through case studies such as the Park Avenue business district in 1950s New York, the shopping mall, the suburban office park, the so-called “creative city” movement, the Bilbao-effect, and the dot.com workplace we examined the interplay of economic structures, governance models, and spatial organization underlying contemporary networked urbanism. Since the 2024 election, several former students and colleagues have reached out, expressing frustration as they try to make sense of the changes underway today. This prompted me to reflect at length on the phenomenon of Trumpism—better known as MAGA, or more appropriately, National Populism.

Although my focus here is on the American manifestation of National Populism, which makes global news daily, this phenomenon extends beyond the United States, appearing in countries like Brazil, Argentina, Hungary, and India. Between 1990 and 2018, the number of populist leaders in power worldwide jumped fivefold (from 4 to 20), reflecting a global shift in political dynamics.1 Globally, 2024 was a bad year for most incumbents, but right-wing populism maintained its strength overall, gaining ground in countries like Germany and France. No longer an outlier, such parties are now firmly part of the political landscape.2 While China’s authoritarian populism differs substantially in structure and goals, it too reflects the global turn toward reactionary politics, with the Chinese Communist Party making more appeals to national identity and cultural restoration similar to other populist movements. Unlike my Network City project, I am not going to use this essay to write about the architectural and urbanistic manifestations. There is, as yet, no clarity to these—and I am still working on interpreting its cultural logic—but the general contours of ongoing societal shifts have become clearer. I should add that while I am increasingly wary of academic narratives in which capitalism is presented as the chief animating force of history, it only makes sense to talk about capitalism when we are talking about socioeconomic developments.

Contemporary progressivism struggled to offer a compelling vision capable of addressing the economic and social anxieties fueling populist resentment. Indeed, its efforts seem to have backfired, with the progressive politics of the Biden administration—elected to be a moderate caretaker administration—driving the Democrats further and further away from their core constituency while alienating many among the minorities Democrats sought to court with their vision of identity politics. This ideological failure created an opening that National Populism eagerly exploited.

While analyses of National Populism in America commonly emphasize its political dimensions in America—nationalism, isolationism, and identity politics—its economic role as a mode of regulation has generally been overlooked. I argue that National Populism should be understood not merely as a political ideology but as a transitional regulatory mechanism emerging during capitalism’s shift from neoliberalism toward a new economic order dominated by artificial intelligence—what I term “AI Capitalism.” Even if this transformation remains largely unconscious—it is hard to believe that Trump himself has any grasp of its full dimensions and certainly his rank-and-file supporters do not—and poorly theorized, it is well underway and will inevitably provoke profound conflict in the years ahead, not just between Republicans and Democrats, but within the GOP as well.

Before proceeding further, I should clarify what I mean by regimes of accumulation and modes of regulation, two concepts central to the French Regulation School’s approach to political economy, made popular in the United States by David Harvey’s Condition of Postmodernity.3 A regime of accumulation refers to the way capitalism organizes production and consumption to enable profit and growth over an extended period. It encompasses specific technologies, labor processes, consumption patterns, and capital-labor relations that allow value creation and capture. The Fordist regime of accumulation, for example, featured mass production, unionized labor, rising wages, and mass consumption, while the flexible accumulation regime that followed emphasized global supply chains, precarious employment, and debt-fueled consumption.

A mode of regulation, by contrast, comprises the institutional forms, networks, and norms that stabilize a given regime of accumulation. This includes everything from formal laws and financial systems to cultural practices and modes of state intervention. The Keynesian welfare state was the mode of regulation that supported Fordism, while neoliberalism and the attendant monetary policy ultimately emerged as the regulatory framework for flexible accumulation. As Harvey concludes, broader cultural logics correspond to these modes, characterized by modernism and postmodernism respectively. In each case, the mode of regulation attempts to temporarily manage capitalism’s inherent contradictions and crisis tendencies through specific institutional arrangements.

In “Almost Anything,” my essay on the work of architect Kevin Roche that modernism had three main cultural phases—its nascent pre-World War 2 state, high modernism of the postwar era, and the late modernism of the 1960s and 1970s when the ideology was exhausted. In turn, postmodernism was succeeded by network culture, a concept I explored as neoliberalism’s final cultural and economic logic, a period characterized by decentralized networks, fragmented identities, and pervasive commodification of digital interaction (it is arguable that a pop art-rock culture preceded postmodernism as a logic of flexible accumulation even as it overlapped with late modernism.4 However, network culture represented neoliberalism’s final cultural phase, inadvertently amplifying its inherent contradictions rather than stabilizing it. Through the deregulation and amplification of information flows, fragmentation of the polity into “networked publics” and the collapse of stable identities, network culture eroded political consensus, generating fertile ground for populist narratives.

Like network culture, National Populism is transitional in nature. Modes of regulation don’t last forever. They emerge in response to the contradictions of prior systems, stabilize a regime of accumulation, and eventually break down. Its primary features—economic nationalism, identity politics, and anti-intellectualism—suggest a complex effort to manage the transition between economic regimes, even if it seems unlikely that its proponents have any idea of (or plan for) what it might evolve into.

Political scientist Cas Mudde defines populism as a “thin-centered ideology” that splits society into two antagonistic camps: “the pure people” versus “the corrupt elite.” Populists claim to uniquely represent the true people’s will against self-serving establishment elites. Importantly, populism is “thin” because it does not on its own offer a detailed economic or policy program – it attaches to a “host” ideology. In the case of National Populism, the host ideology is nationalism or nativism. National-populist movements thus frame the elite as globalist traitors and champion the people as the native citizens of the nation-state, often scapegoating immigrants or foreign influences for domestic woes.5

According to John Judis, populism is fundamentally driven by questions of economic distribution—who gets what— even if the anger often expresses itself in cultural terms. In Western nations, decades of neoliberal globalization allowed corporations to outsource jobs and chase lower wages abroad, while domestic workers felt left behind. By the 2010s, this translated into voter frustration with mainstream parties and openness to outsiders promising to upend the status quo. Specifically, Judis underscores that both left and right populists (from Bernie Sanders to Donald Trump) challenged at least one pillar of post-1970s neoliberalism: the free movement of corporations and capital across borders. In practice, right-wing national populists tend to combine cultural backlash with economic nationalism—opposing free trade agreements, advocating tariffs or protection for local industry, and pledging to bring back manufacturing jobs. They often pair this with welfare chauvinism (maintaining or expanding welfare benefits for native citizens only) and skepticism of immigration. This agenda is offered as a corrective to the dislocations and inequities that globalization (and now technological change) have created.6

At the same time, many scholars emphasize that National Populism is a symptom of deeper structural changes. Economic insecurity, regional inequalities, and the decline of traditional industries set the stage for populist resentment. Cas Mudde and others observe that while populists loudly critique elites, most still operate within a capitalist market framework, often proposing few concrete changes to the economic system. In fact, once in power, populist leaders frequently implement a mix of contradictory policies—tax cuts and deregulation benefiting businesses and investors, alongside trade protection or fiscal stimulus to appease working-class supporters. This inconsistency hints that National Populism may be less an enduring model and more an interim balancing act.

The current wave of technological innovation—centered on artificial intelligence, automation, and data—is ushering in what I term “AI Capitalism.” This is a rather natural term, but it seems that it was introduced by Nick Dyer-Witheford, Atle Mikkola Kjøsen, and James Steinhoff in their book Inhuman Power: Artificial Intelligence and the Future of Capitalism as “actually-existing AI Capitalism.” Of course, a book written over six years ago is ancient history in AI terms and their reference is to the earlier use of narrow AI or machine learning algorithms. Just as “actually-existing socialism” was a reference to the deeply flawed nature of Communist regimes, the authors use the phrase to refer to the difference between machine learning and the generalist AI of the sort that ChatGPT and other LLMs have offered in the years since. Still, the authors get a lot right, emphasizing how major tech corporations “see the cognitive and biological limits of the human as a barrier to accumulation” and aim to bypass those constraints using machine learning, automation, and predictive analytics. In their framework, AI is not just another industry but a means of cognition—an infrastructure that is fast becoming a general condition of capitalist production, much like electricity or global supply chains in earlier eras. They also share my view that AI Capitalism will have a profound impact on labor, making vast numbers of workers obsolete, critiquing both right-wing and left-wing perspectives that downplay the threat of automation, arguing instead that AI-driven job displacement will be deep and systemic. They document how AI does not merely assist workers but actively records human labor in order to replace it—Amazon’s robotic warehouses, AI-driven call center automation, and algorithmic surveillance are all part of this trend. Indeed, they also understand that the result of this automation will be an ever-growing surplus population that capital no longer requires, making permanent unemployment and precarity central features of AI Capitalism. More than that, they propose that even the narrow concept of machine learning, makes AI an essential form of infrastructure, such as water or energy:

“If AI becomes the new electricity, it will be applied not only as an intensified form of workplace automation, but also as a basis for a deep and extensive infrastructural reorganization of the capitalist economy as such. This ubiquity of AI would mean that it would not take the form of particular tools deployed by individual capitalists, but, like electricity and telecommunications are today, it would be infrastructure—the means of cognition—presupposed by the production processes of any and all capitalist enterprises. As such, it would be a general condition of production.”7

In both their analysis and mine, “AI as infrastructure” will be controlled by oligopolistic firms—tech giants who are investing billions to advance AI and capture its benefits. My own view of AI Capitalism—which also draws upon general consensus in business and tech journalism and commentary on AI—is that it represents an emergent stage of capitalism in which data and AI algorithms become core components of economic production, augmenting and replacing intellectual and creative laborers the way that factory labor was replaced under post-Fordism. While new jobs will also be created (and optimistic scenarios suggest overall employment can remain stable in the long run), the transition will be tumultuous. Entire sectors – from manufacturing to clerical office work – are being reshaped.

AI Capitalism, then, is characterized by the extensive automation of labor, including not just manual tasks but also cognitive and decision-making processes, the centrality of data as a commodity, with companies collecting and monetizing vast troves of information to train algorithms, winner-take-most markets due to network effects and high R&D costs, leading to dominant tech conglomerates and increasing ties between big tech and the state, especially as governments seek AI advantages for national security and economic growth.

Note that my view of AI Capitalism does not require significant advances or breakthroughs over present-day technologies. It does not require sentient AI, but rather simply extrapolates current trends on a relatively predictable curve. It is entirely possible that the growth curve could be much faster—although increasing challenges with training new models, continuing problems with hallucinations, and the cost of compute and energy for that compute—suggest that is unlikely—just as it is possible that there will be barriers that will be extremely difficult to surmount. The slow implementation of full-self-driving vehicles is an example of the latter.

Financialization and inequality are another hallmark of the current era. Enormous wealth is being generated at the top of the economic pyramid, especially in finance and tech, but it’s not trickling down. Corporate profits and stock valuations have hit record highs in recent years, even as populist anger at “elites” grows. Such inequality and financialization fuel the very grievances populists leverage. Workers see the “Wall Street and Silicon Valley elite” amassing fortunes while their own jobs feel precarious. AI Capitalism promises greater productivity and new innovations, but it is also disruptive and disorienting. It is automating away livelihoods, rewarding a transnational investor class, and concentrating economic power in a few tech firms. Moreover, its key resources —data, code, capital—flow easily across borders, making it a fundamentally global system. This new paradigm has yet to fully mature or be guided by updated regulations and social contracts. National Populism can be seen as a reactive adjustment: a political attempt to grapple with these economic tremors using the familiar tools of nationalism and statism, even if, at present a tenuous alliance between the National Populist administration and AI Capitalism exists. It is unlikely to hold for much longer. AI is not merely an extension of neoliberal capitalism but a break from it. The rise of AI Capitalism represents a structural shift that National Populism cannot fully contain—one that will provoke deeper economic and political crises as AI advances.

History offers parallels that help us understand National Populism’s role as a transitional mode of regulation. The interwar period of the 1930s provides perhaps the most instructive comparison. With the Great Depression, the earlier liberal capitalism faced a profound crisis of legitimacy. That economic order, which had dominated since the late 19th century, could no longer deliver stability or growth. Into this vacuum stepped reactionary movements that mobilized workers disenfranchised by economic collapse and technological change.

These movements, particularly fascism in Europe, presented themselves as defenders of national workers against both international finance and communist revolution. Like National Populism—which directly references them—they promised to restore national greatness through economic autarky, rearmament, militaristic foreign policy, and the protection of traditional industries. Yet despite their frequent anti-capitalist rhetoric, their actual economic function was quite different. Rather than reversing industrial capitalism, they accelerated industrial concentration, technological modernization, and the development of new production methods while miring their respective economies into deeper crisis. The historical irony is that fascism, particularly in Germany and Japan, served as a bridge between liberal capitalism’s collapse and the emergence of postwar Fordism-Keynesianism. By forcibly reorganizing economic institutions and centralizing industrial capacity, it inadvertently prepared the ground for the corporatist, mass-production economy that would follow.

Political economist Karl Polanyi later analyzed this period as a kind of “double movement.” Free-market capitalism in the early 20th century had disembedded itself from society, leading to rampant inequality and instability. Society then pushed back with a countermovement demanding protection from the market’s ravages. This counter-reaction took both progressive forms (social democracy and the New Deal’s welfare capitalism) and regressive forms (fascist or ultra-nationalist regimes). In Polanyi’s view, fascism was a path where democracy was sacrificed to safeguard capitalism through authoritarian means, once liberal democracy seemed unable to cope.8

Of course, the New Deal in the United States offers a democratic counterexample to fascism’s transitional role. Facing the same global Depression, the Roosevelt administration established new regulatory frameworks, labor protections, and social welfare systems that facilitated the shift to Fordist mass production without abandoning democratic institutions. Similarly, Scandinavian countries navigated the post-WWII transition to coordinated market economies through negotiated compromises between capital, labor, and the state. These examples remind us that while economic transitions inevitably create disruptive pressures, societies retain agency in determining how those pressures are managed.

Today’s national populism can be seen as a similar search for a stopgap solution amid economic upheaval. As in the 1930s, we see widespread disillusionment with liberal elites and international cooperation. Populist leaders invoke economic nationalism—tariffs, border walls, strongman negotiation against foreign competitors—to shield people from global market shocks. In the United States, for example, Donald Trump’s administration has taken a distinctly 1930s-style approach on trade, imposing sweeping tariffs to protect domestic industries. However, these measures can at best provide only fleeting relief before they backfire. Despite tariffs on China and other nations, the overall U.S. trade deficit soared to its highest level in over a decade during Trump’s first term (the combined trade gap hit $679 billion in 2020, up from $481 billion in 2016.9 While the bilateral deficit with China shrank, U.S. companies simply shifted imports to other countries. This underscores a key point: Nationalist economic policies struggle against the structural realities of globalized production.

Today’s transition differs significantly from the 1930s, though the pattern of reactionary politics during economic transformation remains instructive, with “America First” directly appealing to quasi-fascist nativism of that era. Our moment is unique due to the fundamental break in the relationship between technology and human labor. Previous technological revolutions transformed labor without eliminating its centrality to production. This isn’t simply another step in automation but a complete restructuring of capital-labor relations, potentially rendering human labor largely superfluous to capital accumulation. This explains why conventional economic responses—neoliberal flexibility or neo-Keynesian stimulus—appear increasingly inadequate. It also explains also why National Populism will not be able to maintain its tenuous alliance with AI Capitalism and why National Populism is doomed to policy failure. 

With neoliberalism unravelling, National Populism functions as a holding pattern that obscures deeper transitions to an AI-driven, increasingly post-labor capitalism. While its rhetoric centers on jobs, sovereignty, and cultural restoration, the economic forces it unleashes point in a different direction altogether—one that will ultimately make its core constituencies obsolete.

This transition explains many of National Populism’s contradictions. On the one hand, it promises to restore manufacturing jobs and industrial prosperity; on the other, it accelerates financialization and deregulation in ways that undermine those very promises. Protectionism and economic nationalism are at odds with the borderless nature of AI Capitalism, yet numerous tech leaders such as Elon Musk, Peter Thiel, and Palmer Luckey have aligned themselves with the movement while others, notably Sam Altman, Mark Zuckerberg, and Tim Cook have taken pains to pay tribute to Trump.

Several key tensions highlight this contradiction:

  1. Global Supply Chains vs. Tariffs: As noted earlier, populists can slap tariffs on imports, but multinational companies often reroute rather than truly reshore production. Modern manufacturing relies on components sourced worldwide. Complex industries like AI and electronics depend on rare earth minerals from Africa, semiconductors from East Asia, and engineering talent from everywhere. Efforts to localize entire supply chains face high costs and resistance from industry.

  2. Labor Automation vs. Job Promises: National populists frequently campaign on restoring lost industrial jobs (coal mining, steel, factory work), but the harsh reality is that many of those jobs have been eliminated more by automation than by trade. For instance, U.S. manufacturing output is near historic highs, but it employs far fewer workers than decades ago, because robots and software allow far higher productivity per worker. The push for AI and productivity improvements directly undercuts the promise of traditional jobs.

  3. Skilled Immigration and Tech Talent: AI Capitalism is fueled by human talent – engineers, scientists, and entrepreneurs often drawn from a global pool. Yet the nationalist thrust of populism includes harsher immigration restrictions and xenophobic rhetoric that can drive away needed talent. In particular, while many Indian Americans—who now dominated Silicon Valley leadership—supported Trump’s candidacy and he named Tulsi Gabbard as Director of National Intelligence and Kash Patel as Director of the FBI, anti-Indian rhetoric has increased among National Populists since the election. An uneasy tension masks a potential flare-up.

  4. Data and the Digital Economy: Data flows are intrinsically transnational – an AI system might train on user information from millions of people across dozens of countries. Nationalist policies that demand strict data localization or “internet sovereignty” can conflict with the way digital businesses operate. The nation-state finds it hard to assert economic control when information itself ignores borders.

National Populism emerges as a transitional form managing contradictions in a system under profound transformation. Its base consists primarily of people made obsolete by neoliberalism yet unable to participate in AI Capitalism—remnants of an industrial proletariat rather than knowledge workers or financial elites. This demographic differs from both Marx’s proletariat and the neoliberal precariat, presenting capitalism’s greatest challenge since the industrial revolution. With their economic function eliminated, this class’s cultural grievance becomes weaponized political force. Their value lies in political disruption rather than economic production. Though economically redundant, they remain consumers and political actors with legitimate grievances. Yet National Populism’s solutions cannot address the technological forces causing their obsolescence, and their political mobilization ironically accelerates the transformation threatening them.

This surplus position explains National Populist leadership’s unusual relationship with its base. Unlike traditional movements promising material improvements, it offers psychological compensation—dignity, recognition, and transgressive identity politics. Leaders provide cultural and political meaning to those capitalism has discarded rather than economic salvation. This impending abandonment explains populist culture’s increasingly apocalyptic tenor. End-times thinking, conspiracy theories, and decline narratives provide psychological framework for populations sensing their economic redundancy.

Prominent economic advisers to National Populism exemplify this contradiction. The Trump administration’s first term demonstrated this pattern: corporate tax cuts, financial deregulation, and trillion-dollar deficits expanded financialization while ostensibly supporting “forgotten Americans.” In the second term, massive cuts to government programs have been accompanied by promises for further tax cuts. Far from challenging Wall Street dominance, these policies intensified the dominance of financial capital over investment in infrastructure or manufacturing. Similarly, other National Populist regimes have expanded sovereign debt while reducing capital controls, allowing financial speculation to flourish under nationalist cover.

Expanded debt—both sovereign and personal—creates the financial infrastructure that AI Capital requires. AI-driven finance depends on massive data flows, algorithmic trading systems, and complex financial products—all enhanced by the deregulatory impulses of populist governance, as well as investment capital that will be marshalled for AI development. In the meantime, vast amounts of consumer and government debt will be generated to maintain consumption as human labor becomes less necessary.

National Populism’s cultural production deserves deeper examination as it prefigures AI Capitalism’s relationship to truth and reality. The movement thrives on an algorithmic curation of reality that mirrors the digital platforms its supporters often claim to distrust. While denouncing “fake news” and mainstream media, National Populism embraces synthetic realities and alternative fact structures that erode consensus-based truth regimes. Trump himself emerged from reality television, a medium endemic to network culture where authenticity is performative rather than substantive. This collapse of real/fake distinctions itself prefigured the AI-generated media landscape now emerging.

By destabilizing conventional epistemological frameworks while failing to offer coherent alternatives, National Populism prepares society for AI-based reality systems. When truth is already fractured and institutional authority delegitimized, algorithmic authority fills the vacuum. The populist assault on expertise and traditional knowledge production thus inadvertently paves the way for algorithmic governance of information—a core feature of AI Capitalism.

To fully understand the contradictions between National Populism and AI Capitalism, we must examine the ideological movement rapidly gaining traction among tech elites: Effective Accelerationism, or “e/acc.” This movement represents a radical departure from both traditional capitalism and neoliberalism. Unlike both National Populism and neoliberalism, which at least nominally center humans in their economic vision (whether as workers or consumers), like Futurism, e/acc explicitly rejects human-centric considerations in favor of maximum technological acceleration.

E/acc represents the ideological vanguard of AI Capitalism—not merely embracing technological change but actively working to remove all barriers to its maximization. Its philosophical roots extend beyond recent tech discourse to the darker corners of 1990s critical theory, particularly the work of Nick Land, who developed an influential framework of cybernetic Lovecraftianism—viewing technological acceleration as an unstoppable, inhuman force consuming human civilization. His concept of “hyperstition” posited that certain ideas function as self-fulfilling prophecies, bringing themselves into reality through their circulation. Today’s e/acc movement, championed by figures like venture capitalist Marc Andreessen, represents a corporate-friendly version of Land’s darker vision. Andreessen’s “Techno-Optimist Manifesto” repackages accelerationist ideas without Land’s gothic aesthetics or explicit anti-humanism, but retains the core principle: technological development must proceed regardless of social consequences.10

Peter Thiel’s position within this framework is particularly revealing. Despite funding National Populist politicians, Thiel’s philosophical outlook aligns closely with accelerationist principles. His dictum that “competition is for losers” reflects the e/acc view that market competition should be transcended by monopolistic control of technological development. Elon Musk occupies an even more contradictory position. While publicly expressing concerns about uncontrolled AI development, his business practices aggressively advanced automation across multiple sectors. Notably, Grok, the LLM he had developed at X is noted for having the weakest guardrails of any model developed in the US.

When we examine how these leading tech figures have embraced right-populist movements, we see a clear pattern of strategic alignment despite apparent ideological differences. Thiel was one of the few Silicon Valley luminaries to back Donald Trump in 2016. He spoke at Trump’s convention and later served on his transition team. Thiel’s motivations connect to his long-standing critique of what he sees as the complacency of the post-Cold War establishment. His support for Trump can be understood as a desire to disrupt the old regime that, in his view, had become mired in regulatory and bureaucratic inertia. Trump’s populism, with its contempt for “experts” and norms, was a blunt instrument to weaken the existing order. Musk initially kept politics at arm’s length, but by the early 2020s, feeling slighted by the Biden administration which did not invite him to an event celebrating Detroit’s investment in electric vehicles, he increasingly echoed right-populist talking points. Musk’s purchase of Twitter (now X) and his shift to reinstating banned right-wing accounts and attacking mainstream media narratives endeared him to populist conservatives. By aligning with populists, he gains a base of fervent supporters who see him as fighting the liberal establishment. This populist fandom can be leveraged to pressure policymakers to advance a techno-libertarian agenda that benefits his enterprises. At the same time, he has been aggressive in attacking other tech leaders who he feels threaten his business interests, notably Sam Altman, CEO of the current artificial intelligence leader, OpenAI. Andreessen, too, made a similar ideological journey. Once considered a moderate tech optimist, he swung to the right in recent years amid frustrations with the regulatory state. They, along with other tech figures like former PayPal executive David Sacks, share a belief that the political establishment is overly meddlesome, imposing burdensome regulations, antitrust actions, and taxes that impede Silicon Valley’s vision. 

Still, the conflict between e/acc and National Populism is inevitable even if temporarily concealed. E/acc views border controls, worker protections, and cultural conservatism as inefficiencies to be eliminated; National Populism depends on these very structures for its political identity. The current alliance exists because both movements oppose the administrative state and regulatory oversight, but for fundamentally different reasons—National Populism because it views these institutions as corrupted by globalist elites, e/acc because it sees them as impediments to technological acceleration. This temporary alignment explains why tech billionaires have become willing funders of populist movements despite their obvious ideological differences. For e/acc adherents, National Populism serves as a useful battering ram against the regulatory state—once those barriers are demolished, the movement and its human-centric concerns can be discarded. Ultimately, the National Populist base is fundamentally incompatible with AI Capitalism’s trajectory. Primarily low-skill and low-tech, this demographic faces increasing economic obsolescence within an AI-driven system. This incompatibility runs both ways. The base actively despises the very symbols of AI Capitalism, exhibiting a virulent hatred of electric vehicles and rejecting the cultural markers of technological elites, notably Tesla. After Musk’s alliance with Trump, the manufacturer has seen sales crater both in the US and worldwide.

As I pointed out earlier, the nationalist orientation of populist movements also directly conflicts with AI’s inherently global infrastructure. AI Capital demands borderless computation, global talent pools, and transnational flows of data. The economic vision of National Populism, with its emphasis on borders, national sovereignty, and protected markets, contradicts the fundamentally planetary scale that AI-driven capitalism requires to function efficiently.

As David Graham noted in The Atlantic, tech elites often don’t mind the “populist assaults” on establishment corporations or institutions, because they themselves operate somewhat outside the old corporate world. The venture-capitalist mindset of  Thiel, Musk, and Andreessen leads them to “like disruption. They don’t care if the old companies get turned upside down.”11 This startling insight reveals how even when populists attack Big Tech or Big Finance, they wind up helping rival tech entrepreneurs by hobbling their competitors or opening up new opportunities.

It seems clear that National Populism is not the future—it is a political interlude, a failed attempt to resist a transition that is already happening. Its economic promises are illusions, but its function is important: it provides a reactionary buffer that delays the recognition of AI-driven economic transformation. The real question is not whether National Populism can survive, but what system will replace it once it is no longer useful to those who are actually shaping the next economic order. The historical parallels I outlined earlier are instructive. Nazi Germany was a horrific regime, but economically it served as a transitional phase between liberal capitalism’s crisis and the postwar German economy. Similarly, Pinochet’s Chile served as a bridge from state-centered developmentalism to neoliberal market fundamentalism, using authoritarian political power to forcibly reshape economic institutions. National Populism similarly represents a reaction to the collapse of a prior order that inadvertently speeds up the transition to what comes next. Its anti-intellectualism and political disruption provide cover while AI consolidates its position and remakes the economy in ways that will ultimately make National Populism irrelevant. We can only hope it will be less violent than its predecessors. 

The transition from neoliberalism to AI Capitalism, with National Populism as its flawed mediator, represents one of the most significant economic and political reconfigurations of our time. What remains uncertain is not whether AI Capitalism will emerge—this transition is already underway—but what form it will take and how democratic institutions might channel its development. For the National Populist base, this presents a tragic irony: their legitimate grievances about economic dispossession are channeled into a political movement that accelerates rather than addresses the forces rendering them obsolete. National Populism correctly identifies that neoliberalism’s promise of shared prosperity has failed many communities. Its critique of unaccountable elites resonates because it contains elements of truth. The hollowing out of industrial regions, the concentration of opportunity in a few coastal hubs, and the growing chasm between the technological elite and everyone else are real phenomena that demand response. Their fate represents perhaps the central political question of our time: what happens to human populations that capital no longer requires?

What has been missing, however, is a vision that can address these issues rather than channel discontent toward convenient scapegoats. The challenge for forward-looking thinkers is to develop models that harness AI’s productive potential while ensuring its benefits are broadly shared. This requires reimagining both the state’s role in managing technological change and capital’s relationship to labor in an era when traditional employment may no longer serve as the primary mechanism for distributing income. The transition from neoliberalism to AI Capitalism will likely be as tumultuous as the shift from Fordism to flexible accumulation was in the late 20th century. Just as that earlier transition generated new modes of regulation (neoliberalism) and cultural logics (postmodernism, then network culture), our current moment demands fresh thinking about how societies can navigate technological transformation without sacrificing democratic values or human welfare.

The failure of Democrats, themselves dominated by a Progressive Populist base, in the 2024 presidential elections—alongside broader struggles facing neoliberal governments globally—highlights the inadequacy of existing political and economic models. Even progressive initiatives stumbled by entangling themselves in narrow, identity-based frameworks. For example, Alexandria Ocasio-Cortez and Ed Markey’s Green New Deal, and the climate provisions in Biden’s Inflation Reduction Act of 2022, addressed vital environmental challenges but undermined broad appeal by prioritizing specific communities or businesses based on identity criteria. Such approaches, however well-intentioned, offered rhetorical ammunition to National Populists, who portrayed them as divisive rather than inclusive.

Indeed, both Progressive and National Populists have shared an underlying assumption of scarcity, treating economic and social resources as fundamentally limited, to be divided among competing constituencies. This scarcity mindset shapes public discourse, turning debates over essential goods like housing, healthcare, energy, and education into zero-sum struggles. The National Populist alliance with accelerationist tech elites reveals the fundamental instability of this arrangement. What comes next will depend on our collective capacity to imagine and implement more robust, equitable, and democratic responses to the AI revolution—responses that acknowledge technological change while insisting human flourishing, not technological acceleration for its own sake, must remain central.

Yet even as I wrote this piece, a promising alternative emerged in Ezra Klein and Derek Thompson’s book Abundance. Klein and Thompson reject the scarcity mindset entirely, arguing instead for a return to a fundamentally American optimism, harnessing technological innovation to drive productivity growth and ensure broadly shared prosperity. They show how conflicts in housing, healthcare, energy, and education result less from genuine resource constraints than artificial scarcity imposed by outdated regulations, NIMBYism, and captured markets. Crucially, Klein and Thompson’s vision does not ignore equity or sustainability concerns. Instead, they argue we can achieve inclusive prosperity through ambitious, targeted reforms that prioritize growth, innovation, and democratic oversight. This abundance-oriented approach offers a path beyond the zero-sum thinking of populism or the unchecked accelerationism of tech elites. Such a pragmatic yet ambitious vision may provide the conceptual foundation for a new mode of regulation capable of managing AI Capitalism’s contradictions, guiding us toward a future where technology serves democratic values and human flourishing rather than supplanting them.

1. Tony Blair Institute for Global Change, “High Tide? Populism in Power, 1990-2020”
https://institute.global/insights/politics-and-governance/high-tide-populism-power-1990-2020.

2. Richard Wike, Moira Fagan, and Laura Clancy, “Global Elections in 2024: What We Learned in a Year of Political Disruption,” Pew Research Center, December 11, 2024, https://www.pewresearch.org/global/2024/12/11/global-elections-in-2024-what-we-learned-in-a-year-of-political-disruption/#the-staying-power-of-right-wing-populism.

3. David Harvey, The Condition of Postmodernity (New York: Blackwell, 1991). For an understanding of the regulation school, see Robert Boyer, The Regulation School: A Critical Introduction (New York: Columbia University Press, 1990) and Michel Aglietta, A Theory of Capitalist Regulation: The US Experience (London: Verso, 2015).

4. Kazys Varnelis, “The Meaning of Network Culture,” Eurozine January 14, 2010, https://www.eurozine.com/the-meaning-of-network-culture/?pdf, originally published as “Tinklo kultūros reikšmė,” Kulturos Barai, no 9. 2009, 66-77 and, in an earlier version as “the Meaning of Network Culture,” Networked Publics (Cambridge, MA: The MIT Press, 2008), 145-164.

5. Cas Mudde, “Populism in the Twenty-First Century: An Illiberal Democratic Response to Undemocratic Liberalism,” The Andrea Mitchell Center for the Study of Democracy, 2020. https://amc.sas.upenn.edu/cas-mudde-populism-twenty-first-century

6. German Marshall Fund and John Judis, “Three Questions With John Judis,” German Marshall Fund US, https://www.gmfus.org/news/three-questions-john-judis

7. Nick Dyer-Witheford, Atle Mikkola Kjøsen, and James Steinhoff, Inhuman Power: Artificial Intelligence and the Future of Capitalism (Pluto Press, 2019). For the discussion of a “trajectory towards a capitalism without human beings … a permanently unemployed section of the working class that consistently grows larger … [along with] the superlative growth of the surplus population…” see 140-141 in particular. On AI as infrastructural, see 30-31.

8. Karl Polanyi, The Great Transformation: The Political and Economic Origins of Our Time (Boston: Beacon Press, 2001), 136–140, 245-256.

9. Doug Palmer, “America’s Trade Gap Soared Under Trump, Final Figures Show,” Politico, February 5, 2021, https://www.politico.com/news/2021/02/05/2020-trade-figures-trump-failure-deficit-466116

10. Marc Andreessen, “The Techno-Optimist Manifesto,” Andreessen-Horowitz web site, https://a16z.com/the-techno-optimist-manifesto/

11. David A. Graham, “The Fakest Populism You Ever Saw,” The Atlantic, July 19, 2024, https://www.theatlantic.com/politics/archive/2024/07/trump-vance-fake-populism/679100/ 

2022 in review

I missed the year in review for 2021 entirely. The end of 2021 was stressful: it wasn’t a terrible year, the way the Trumpenjahren were, but it was bad. I ran out of steam and never pulled the post together. Not so this year. I’ve posted once a month on average, which is the most posts since 2016. Most of these were quite long opinion pieces and some, like the Critical AI Art projects took weeks of work to produce. Moreover, posting really began in earnest later in the year after I switched to my new server at Kinsta and my new theme powered by GeneratePress (see here). Not only is the site faster for you, dear reader, but it is also much faster for me to work on. For the first time in years, the future of this blog looks bright.

This post is comprised of four parts: The End of the Covidian, Geopolitical Transitions, Network Culture RIP, the Age of Desiring Machines

I. The End of the Covidian

2022 has been a good year and, although I know some of my readers will disagree, at least half of it felt like we left the pandemic behind. Goodbye to the Covidian era. As I write this, we are off skiing in northern Vermont and while some things still aren’t open and there are longer lines due to staffing shortages, it feels like COVID is over. Everyone in the family except me has had COVID during the last year and nobody got as seriously ill as our youngest did from the flu last month. Now, I’m far from an extremist on this: I have all my vaccines, all the boosters, and have taken reasonable precautions throughout the pandemic. But looking at the statistics, it’s clear that vaccines and herd immunity are here. Pretty much every article I see reposted to raise alarm about the new wave of COVID coming “any day now” declares that we need to watch out for a “troubling new variant,” but there is no troubling new variant, it’s just clickbait. Once there is a troubling new variant, then I’ll worry. In the meantime, this is the new normal. You’ll either be wearing a mask forever—which may be good if you are seriously immunocompromised—or not. Epidemiologists are pretty much always in a constant state of panic about diseases, it’s their training to do so; I’d probably be in a constant state of panic if I knew what they know. Instead, I’ll choose to live my life, which is what most people have done now.

It was always naïve or disingenuous of Dr. Fauci and others to claim that vaccines would utterly eliminate COVID the way Ebola was eliminated, COVID was already too widespread and contagious. But if COVID is, as claimed, a novel coronavirus, the odds are that once the massive and tragic initial impact is over, while it would never disappear, once we achieved a degree of immunity to it, it would be something we could live with in an endemic state, like existing coronaviruses, the new COVID normality. What about long COVID? Sure, it’s real, although many of the studies on long COVID seem quite poor, and instead of fretting about it, maybe we should pay attention to the long-term consequences of all viral infections? I have been struggling with IBS which began after a bad cold forty years ago, and Epstein-Barr, which causes mononucleosis, appears to cause multiple sclerosis. That’s pretty bad right there and while I have immense sympathy for anyone affected by any long viral disease, isolation, and constant masking have very real consequences on human life, particularly on child and adolescent development. We’re done with it and, unless and until something horrible appears, we’ll be living life in most ways as we did before March 2020. The COVID-induced supply chain crisis is largely over. New challenges are emerging, but the Covidian era is (likely) history.

II. Geopolitical Transitions

The biggest news of 2023 was, of course, the invasion of Ukraine. There has been huge suffering for Ukraine in the single largest violation of territorial sovereignty by a foreign power in Europe since World War II. But the Russian Bear stumbled and got badly bloodied. For centuries, Russia has been an awful neighbor, a bully, not a country that plays by the rule of law. Built on kleptocracy and theft at home, the state model for foreign relations is to invade, rape, kill, and exploit ethnic minorities and their sovereign lands to make up for the shortcomings of the kleptocratic Russian economy. As a result, leaders in the Baltics, Poland, and even Ukraine realized that post-cold War Russia was a threat and looked westward, where they might seek protection. Putin might have had a chance to counter this had he struck a decade ago, but for some inexplicable reason, his first excursion into Ukraine was halfhearted and he didn’t complete the task when he had a frightened lapdog as US President.

Corruption, incompetence, and an utter lack of strategic thinking undid the initial Russian thrust and, with help from the US and NATO, Ukraine is not only holding its own, it’s beating back Russian aggression. Russia is resorting to its usual tactics of massive bombardment of civilian positions from a safe distance, but with Ukraine, they’ve encountered a country not only fighting back on its own territory but also lobbing missiles back at Russian bases deep in their territory even as unknown saboteurs are destroying Russian infrastructure. It’s still unclear what the outcome will be or when: the result may simply be a question of who runs out of ammunition first, but Putin’s colossal miscalculation means there is a remarkably high chance it will be in the collapse of the criminal regime of Vladimir Putin. I am zero optimism that the result will be a new, more democratic, peace-loving regime in Russia. On the contrary, the collapse of the remaining empire will lead to a series of internal disputes and civil wars and a decline into a general ungovernability of the sort that has taken over much of the Middle East. Doubtless, China and other smaller powers will also make incursions into Russian territory, whittling away administrative regions for their own purposes. In Moscow and St. Petersburg, the next two decades may feel much more like cyberpunk dystopian versions of their 1990s selves: barely governable cities where the mafia and oligarchs take even more control while ordinary individuals resort to unprecedented measures to survive.

I’ve been to Europe a few times since the invasion and to Lithuania twice. Germany made a tremendous miscalculation under Merkel, allying itself with Russia and drinking deeply of its energy, but that route is going away forever and so is its role as leader of NATO and the EU. France and the UK have also been weakened by their complacency. These are economic empires in decline and—especially if Ukraine wins—a new center for Europe is going to emerge in the East, stretching from the Baltics down through Poland—which will be the most dominant force in this Europe—and into Ukraine. Turkey is already proving a powerhouse, but it is less likely to be a threat than an ally with this new democratic East bloc. This is where the energy in Europe is now: nations rejuvenated by existential threats frequently roar back as mighty powers, just the way Germany and Japan did after WWII. One last word about Russia: it re-introduced nuclear threats into East-West relations, but it did so poorly by repeatedly drawing lines that have been crossed. There has been no real escalation in readiness on the Russian side. While it certainly remains possible, it’s the silent bear you need to worry about, not the grunting one.

Although China hasn’t suffered the same humiliation that Russia has, it seems to be past its peak as well. The Zero COVID policy was an economic and social disaster that led to mass unrest and its end was utterly mismanaged. With Russia’s failure in Ukraine, Qi is forced to question his prospects for invading Taiwan while the West’s turn away from China has become even more urgent as its troubles with COVID cement the idea of China as an unreliable trading partner. Worse still, China has finally turned the corner to the other side of its demographic bubble and its population began contracting in 2022. It will be many generations before it is on the upswing again.

I don’t feel like I know enough about the global south, so I’ll skip that. But all this indicates that the 2020s are going to be very different than the 2010s. The Eastern European nations and Turkey will become increasingly important as Russia, Western Europe, and China are spent. It’s still unclear to me what countries outside of Europe will replace the BRICs, but no doubt there will be some surprising times afoot in this coming decade. Even if everyone may throw up their arms at this, the US—disregarding all its troubles—is likely to come out of the decade in a position of strength simply because of resources, population, a lack of real threats on its borders, and the existing geopolitical order. Much of this was foretold in geopolitical forecaster George Friedman’s 2011 book The Next 100 Years. Crucially, he repeatedly points out that no matter how violent disagreements between parties within the US really seem, the underlying policy doesn’t shift as much as it might appear it would, so notwithstanding Putin’s useful idiot in the White House, the US not only didn’t leave NATO, they left it stronger by forcing smaller countries to increase their defense spending; likewise, when Democrats took power in 2022, the US’s newly aggressive policy toward China didn’t really change. If you are interested in geopolitics, it’s worth a look.

III. Network Culture, RIP

Even as life is recovering and momentum is returning, there has been a renewed economic crisis throughout much of the world. Some of it is thanks to larger macroeconomic factors, e.g olb War, but much of it has to do with mistakes in economic policy—goosing of the market for far too long with loose monetary policy, quantitative easing, the misguided 2017 tax cuts, and too much pandemic relief. But the real cause is the end of a technological and economic cycle that began 20 or 30 years ago (depending on how we measure it) and had its heyday in the 2010s with the vaunted FAANG stocks (Facebook, Amazon, Apple, Netflix, Google) growing about ten times faster than the rest of the market and driving equity markets to new highs. Over 2022, Facebook/Meta is down roughly 65%, Amazon and Netflix are down over 51%, Google/Alphabet is down 40%, Apple is down 29%, and FAANG adjacent stock Tesla is down 68%. Bitcoin, itself, which isn’t a stock but rather a Ponzi scheme, is down 64%, the S&P Cryptocurrency Large Cap index is down a massive 69%, and we all know how things ended for Sam Bankman-Fried and SBF. Compare this to the Vanguard Consumer Staples Index Fund, which never ran up as high, but is down a mere 3.69%.

This terrible tech performance, particularly in cryptocurrency, is indicative of a speculative bubble deflating, but it also points to a generational shift in technology. I am not an absolute believer in Kondratieff waves—long economic waves based on technological development that writers from Carlota Perez to Fredric Jameson have embraced—they seem too deterministic to me, but there is also some macroeconomic sense to them. New technologies drive speculative investment, which results in returns that seek more investments of a similar kind. After a while, overinvestment leads to bloat, the bubble bursts, and the economic system declines precipitously. The sharing economy, Web 2.0, and that branding abomination, “Web3” are finished. And with the end of this system, so is its cultural logic, network culture.

I first wrote about network culture in the mid-2000s and my first piece on the topic came out in our book Networked Publics. You can read the original version here and a revised version here. This piece has been translated into numerous languages: Lithuanian, Hungarian, Spanish, Chinese, and others (I’ve lost track at this point). I started a book on the topic immediately thereafter but I wasn’t able to finish it due to external factors beyond my control and a debacle at the publisher. You can read various spin-offs in “Forced Exposure. Networks and the Poetics of Reality,” in Jo-Anne Greene, Networked. A Networked Book about Networked Art on turbulence.org, “History After the End. Network Culture and Atemporality,” Cornell Journal of Architecture 8, spring 2011, “Simultaneous Environments,” in Mark Shepard’s, Sentient City: Ubiquitous Computing, Architecture, and the Future of Urban Space, and in “Architecture of Financialization,” Perspecta 47, 2014 (in the coming days, I will post all these pieces to my site).

The basic idea of network culture came out of my frustration that academics were still using Fredric Jameson’s “Postmodernism, or the Logic of Late Capitalism” over twenty years after it had been published, long after that epoch was finished. Jameson was never able to see past this, but a number of us did. Some other words were thrown around like metamodernism and post-postmodernism, but “network culture” made sense to me, indicating that there was a new cultural logic that was now based on relationships and connections primarily mediated by the Internet. As I wrote then, “Increasingly, the immaterial production of information and its distribution through the network is the dominant organizational principle for the global economy.” As Manuel Castells concluded in The Rise of the Network Society networks now supplant hierarchies and the production of information and the transmission of that information on networks is the key organizing factor in the world economy today. On a territorial and even geopolitical scale, Saskia Sassen pointed out in The Global City, megolpolises dominated, linked together by high-speed telecommunications networks, producing the financial and media operations that made the network economy thrive.

Network society was a globalizing society, what Michael Hardt and Antonio Negri called “Empire” and network culture was a global culture: subcultures and local undergrounds began to decline. In an economy dominated by sharing, cultural mixing, and rapid wealth generation, the idea of the artist as a cultural elite was largely replaced by an interest in participation and remix. And yet, art could also be tremendously valuable as venture capital relentlessly sought new outlets. NFTs were the logical outcome of all this, removing artistic merit in favor of pure speculation—especially from people who didn’t know what they were doing with art or investment—led to the creation of an utterly bogus $11 billion market of which over $800 million is stupid looking apes that look like they are waiting to audition for a Gorillaz video game.

NFTs and the Boring Ape Club were, however, the last gasp of network culture, a decadent last spurt that only proved the system was spent. The signs of cultural change are around us. Network culture is dying. Social media is not coming back, not in its traditional form. Just 7% of teenagers say they use Facebook constantly. https://www.pewresearch.org/internet/2022/08/10/teens-social-media-and-technology-2022/ Nobody on Earth, besides Mark Zuckerberg, wants to wear a VR headset to have a meeting in a virtual office full of amputated refugees from a rip-off of a Pixar movie. TikTok is popular, but I would be surprised if there aren’t massive restrictions or even an outright ban put in place by governments by the end of 2023. Twitter is in freefall. The world’s richest man has proven to be the world’s biggest idiot by spending a staggering $44 billion dollars on a site that was already in trouble and cementing it’s demise by acting like an idiot. These sites are not coming back. The one site that seems to absorb the attention of youth today—TikTok is much more like Youtube—a platform for consumption—rather than a traditional social media site and is under constant threat by Western regulators. The beginning of the end really happened in 2016. If on the one hand, Trump rose to power due to network culture—heavily employing social media and viral memes to mobilize followers—he also embodied the discontent with globalization that had always been there, but that had achieved a new fever pitch as the system spent itself.

Back in 2010 Bruce Sterling (in “Atemporality for the Creative Artist”) and I (“The Decade Ahead“) predicted that this epoch he and I called network culture would last at least ten years. I wrote: “Toward the end of the decade, there will be signs of the end of network culture. It’ll have had a good run of 30 years: the length of one generation. It’s at that stage that everything solid will melt into air again, but just how, I have no idea.” COVID was the break, and, after those ten years became “the decade of shit,” nobody is going to miss network culture. In retrospect, the 2000s were the decade of excitement, of forging new connections across age-old boundaries while finding old friends, of a world that had promise and was still imbued with the utopic promises of the early Internet and open source culture. The 2010s showed us just how toxic network culture could get, as both right (and left!) sought to squash dissent and get their minions in line. Hitler and Stalin would be proud of their descendants. A medium designed for utopic levels of human connectivity hurtled us toward civilizational collapse. That a disease spread by globalization and exacerbated by lies on social media (e.g. the anti-vax movement) ended all this is not surprising. We are lucky it wasn’t something worse.

IV. The Age of Desiring Machines

What comes after? My writing on network culture came a good way into that cultural epoch. Writing about this early is guaranteed to fail, but there were interesting if still premature, signs in 2022. First, there is the rise of Mastodon and decentralized communication. I have outlined my thoughts on this topic earlier, but suffice it to say, something new is in the works, a form of social media that hues closer to the original intent of the Internet. It may be that Mastodon always remains a small player in the net, but smallness is its strength. We need to bring back undergrounds and subcultures, not giant corporate meeting places that spread toxicity.

2022 has also been marked by the rise of “Artificial Intelligences” capable of producing text and images. I have explored these extensively and continue to do so. Ignore the horrific kitsch you see produced by these things, or better yet, don’t: the world of Deviantart and Artstation is bad, a byproduct of network culture permeated by simplistic online fan culture, NFTs were always stupid, now anybody can make things like that and this stuff is valueless. Good!

But calling ChatGPT, Midjourney, or Dall-E “intelligent” is wrong. These platforms have no ability to comprehend what they are doing. But might they be desiring machines? In the Deleuzean sense, a “desiring machine” is formed out of connections: every machine (or entity) is connected to another machine and in turn to another. This desire is not just about wanting something, but also about the process of becoming and creating through connections and that Is exactly what these platforms do. Responses to our prompts are based on the machine’s prediction of what a correct response would be. In other words, these systems are characterized by their desire to fulfill our desires. This is all very far from artificial general intelligence—although a baby crying for food is also far from a scientist or even a toddler in its ability to reason—but it is something new. There are a lot of unknowns here: we may already be at the end of the rapidly rising part of the S curve for these systems, or we may only be at the beginning. Either way, there is a reckoning in store for cultural producers and mid-level professionals producing banal work that will cause massive disruption.

There has been a lot of useless noise about the ability of these platforms to create fakes and I’ve played with that in my art, but where did we go wrong as educators? What happened to the idea that we should think critically? Wasn’t art history, as codified by Wölfflin, literally a matter of finding out how to authenticate something? Isn’t that what we learned in high school? Who are these people who have forgotten that “critical thinking” doesn’t mean blindly accepting whatever you see but rather that it means taking a critical distance from a text or an image?

Disruption is the key for the next few years, during which the outlines of a new cultural logic will begin to become apparent. The future is likely to be in terms of exacerbating the dictum attributed to cyberpunk writer William Gibson: “The future is already here, it’s just unevenly distributed.” Don’t expect a utopic condition from new technologies and don’t expect the rise of socialism. Everyone is out to grab what they can. I am older and less optimistic than I was ten years ago, less prone to see the spectre of capital behind everything but also less prone to think anything can change that much. As the second season of White Lotus just emphasized, the upcoming generation is as confused, toxic, and prone to gaslighting and self-deceit as the previous ones. Colossal numbers of kids are being medicated, and while some small percentage need it, the amount of medication psychiatrists dispense needlessly is staggering. Russia’s barbaric invasion of Ukraine, the American Right’s refusal to condemn an insurrection that imperiled democracy, and the thoroughgoing denial of climate change (this essay was written during downtime from a ski vacation that is ending in a massive rainstorm) prove that too many humans are still bastards.

In the meantime, I’ll keep working on the design of my 1/2 acre (1/5 hectare) native plant garden, my art, and my writing—most especially on this blog, where the only thing that can impede my publishing is me. I’d love to get new commissions, but if not, there’ll be more to come on this site. Let’s hope 2023 is a better one. I think it will be, but I don’t expect miracles. Let’s hope the new cultural logic is at least as interesting, but less toxic than network culture. That would be quite an accomplishment right there.

On Mastodon

The demise of Twitter and Facebook together with the rise of Mastodon is one of the biggest changes in online culture in the last decade and, naturally, I am covering it in my upcoming 2022 year in review. But in drafting it, the section on Mastodon kept growing until I thought it best to break it out into a separate post. The first part, an overview of Mastodon and its rise, is likely to be most interesting to general readers. The second part is a set of observations about how Mastodon could be improved, intended as an offering to improve the platform on its own terms, not to replicate existing social media. While I am relatively new to Mastodon, I have also been on the Internet for over thirty years and I started on more decentralized platforms like USENET, email lists, and forums, not to mention a scholar of network culture for well over twenty years so I have seen a lot of things work and a lot of things break.

I. The Rise of Mastodon

After a decade and a half of corporate social media, the current system, dominated by Facebook and Twitter, is spent. Who would have thought that algorithms designed to reduce engagement with actual friends and to instead promote celebrity and brand worship as well as political polarization would drive away people of all political persuasions? Who could imagine a VR world of cartoon avatars that look like legless small children would not be an attractive alternative to the workplace, be it real or Zoom? Who would think promoting genocide and allowing shootings to be live-streamed would be a bad idea? Increasingly, Facebook is a brand for old people like America Online, Talbots, or Prevention. The broader public is finally sick of this mental diarrhea and the result has been a rout for social media stocks. Like Facebook, Twitter had been in decline for years and Elon Musk’s purchase of Twitter has been a massive unforced error, destroying both the network and any lingering shred of credibility he still possessed. As I wrote this last night, the site went down for hours. Both Twitter and Elon are shot, their glory days are over and the end is near. Facebook can hardly be far behind, as doomed as Mark Zuckerberg’s presidential ambitions.

Mastodon was founded in 2016 and while I signed up for it a while back, I didn’t find it a compelling place for things that interested me—partly because I hadn’t encountered any good guides on how to do so and partly because it had few active users. Even though Mastodon still has less active users than any major social media network, as anyone who has been involved in technology in the last forty years will tell you, it’s not the number of users that matters, it’s how steep the growth curve is and starting with the Twitter exodus in November the numbers of monthly active users has grown at least tenfold from 300,000 to 2.5 million. Moreover, the engagement I am getting on Mastodon is far greater than on Twitter: a ratio of about 2 posts for every new follower on Mastodon, as opposed to 48 to one on Twitter. Within a month, I have found myself with almost half as many followers on Mastodon as on my two-year-old Twitter account (I purposely burned my previous account out of frustration during the end of the trumpenjahren). Moreover, whereas many of my Twitter followers are random bots, confused political extremists who have made a colossal mistake in following me, or accounts that haven’t been used in a year, my Mastodon friends are chiefly interested in experimental technology, art, and music, are native plant enthusiasts, are people who dream of a less corporate network culture, e.g. exactly the kind of intelligent friends I’d like to engage with. What interests me is the community of individuals working with art and technology I have met. I have seen more interesting work in a couple of months on Mastodon than I have in years on Twitter. I think part of the reason for this is that while Mastodon is about as technically challenging as a toaster, that is still enough to scare off many people (like this reporter), so it has the advantage of being a relatively good place, at least for now.

Crucially, Activitypub, the system underlying Mastodon, is a protocol, not a platform owned by a corporation (see this article on protocols vs. platforms by Mike Masnick). Mastodon is a non-profit that rejects VC funding and does not insert advertising in its feeds. But Mastodon isn’t just an open-source alternative to Twitter, it is a decentralized system that uses individual servers, (“instances” in Mastodon parlance) that are linked (“federated”) together. It is possible, although challenging and relatively expensive (unclear how much, but I would budget about as much as to operate a website or $100-$300 a year), to create one’s own instance and while there are instances that have hundreds of thousands of users (such as the one I am on, mastodon.social), there are also quite a few instances with only a few dozen users.

Whereas in traditional social networks content moderation is the purview of a large corporation, at Mastodon it devolves to the administration of each instance. Terms of service are up to each instance and if the administrators of one instance feel that another instance is not moderating content appropriately, they can stop federating with it. Extremist social network Gab, for example, is an instance, but most other instances refuse to connect to it. If it turns out that mastodon.social—which is clearly progressive politically—is not federating with instances that have reasonable albeit right-wing views, I might leave it for an instance that is more broad in its thinking, the same goes if extremists from either side begin to take over. This is only a hypothetical situation, but it gives me an out that other social networks don’t. There are few easy solutions to content moderation and as law scholar Alex Rozenshtein argues in this piece, the debates are likely to be “messy and public.” I noted that one Mastodon server devoted to archivists, librarians, and museum workers also bans images of insects that aren’t marked as sensitive, requiring a direct click to see. This would be quite disheartening for a curator of entomology! I suspect instances with overly restrictive regulations will be less popular in the end and that will largely be a good thing. That said, for now, I personally avoid posting political positions and avoid following anyone who posts too many. I don’t need another shitshow like Twitter and I certainly don’t need to get my news from people posting on social media.

Again, joining Mastodon isn’t difficult for anyone with a minor degree of tech savviness: you start by choosing an instance at joinmastodon.org based on what you want to see, where you live, or maybe based on your profession (for example, journalist, mathematician, or infosec), then you start following people. You can follow me at @kazys@mastodon.social but, if you have a Twitter account, a good way to follow individuals is to use movetodon, which scrapes the profiles of people you follow on Twitter to identify their mastodon account. Your main feed on Mastodon is composed of the people and hashtags you follow, but there are also both local and federated feeds, which show the most recent posts for either condition. In this case, being on a server that matters to you might make more sense.

II. Observations and Comments about Mastodon

First, discovery is something that Mastodon needs to work on. Understandably, search works for one’s server as well as for hashtags, but broader search across instances isn’t possible. This is apparently by design, to avoid trolls who search for topics to drop into. But the problem remains. How does one find something obscure, say posts about Nakagin Capsule Tower that aren’t tagged #nakagin? This seems to be a stumbling block that requires some really innovative thought.

Second, while it is possible to repost (or “boost”) a post (or “toot”), it isn’t possible to quote a post as this is seen as similarly seen as creating an atmosphere that encourages trolling. A number of Black social media users have complained that this undoes the call-and-response culture in their community. I would add that the real harm on Twitter isn’t from retweets, it’s from subtweets, in which someone takes a screenshot of a tweet and then adds a derogatory comment that the original poster, or lolcow, isn’t even aware of.

Third, the instance model is designed to encourage people to interact with their local community, as most smaller servers are based on self-identification. This is a great idea. One of the most destructive aspects of network culture has been a loss of subcultures and underground movements. Instead, we have a boring global soup. But having to choose an instance forces some individuals into making tough choices: is one’s sexual identity more important than one’s profession, is one’s country more important than one’s sexual identity, is one’s profession more important than one’s race? As of now, while one can see a local feed or a feed of everyone on Mastodon (useless at best), it seems impossible to see another server’s feed without joining it or making the effort to visit the server’s web page. The current solution preferred—multiple accounts—would make some sense for someone whose sexual identity and ethnic identity are important but who also wants to join a professional server, but what about someone whose identity brings together history, music synthesis, technology and art (e.g. me)? These are hard choices to make and it seems that being able to read and interact with feeds from multiple instances seems important.

Fourth and most crucially, Mastodon should create a distinct way to follow big accounts such as news sources and (gulp), celebrities of all stripes. The Long Tail is a ruse. The Internet has been prone to the Pareto principle or 80/20 rule (80 percent of the traffic goes to 20 percent of the sites… usually those numbers are even worse). Now Pareto wasn’t out there to promote democracy, he thought those 20 percent should rule society, an idea that immediately appealed to Mussolini. Social media companies have sought to encourage this, implementing algorithms to promote posts that get more likes instead of ranking posts chronologically. Facebook did this in 2009, constantly tweaking the algorithm in ways that made it worse and worse each time. Twitter ceased using chronological sorting and implemented its algorithm in 2016. By some uncanny coincidence, an idiot was elected US President in 2016, largely on the back of his Alzheimer’s induced tweets and, coincidentally Twitter also ceased being an interesting place to use that year. Social media algorithms set out to reinforce a model based on influence (as well as power and wealth) accumulating to a small number of individuals. They are terrible. But stars and influencers are also responsible for encouraging people to come to them (for example, one individual on Mastodon, who is also a well-known actor and social media personality follows under a hundred individuals but has hundreds of thousands of followers and added “boost this toot” and to one of his posts earlier last month). I didn’t disagree with his political message, but this seems deeply unhealthy to me as it reduces discourse to one-way communication dominated by the few. Now there might be situations in which this might make sense. For example, I might want to follow updates from my town, NASA, the location of Elon Musk’s jet, or a news source, and it seems ludicrous to expect them to follow me, plus as it is possible to follow RSS feeds (such as this blog) on Mastodon. But following a good number of these clogs up one’s timeline, which ideally should be a social feed from a community of friends. The solution here seems simple enough: create a section that we might call “channels” or “news” for these one-way accounts. That way, one could follow whatever one-way accounts one wanted while preserving the timeline for genuine interaction. This can be done with two lists now (say “friends” and “channels”), but that requires the active addition of accounts to each, which is needlessly time-consuming and means the timeline itself becomes useless. There is already a “News” tab on the official Mastodon iOS app, although not on the web interface, although this is already pre-populated with news sources and there appears to be no way to change this and is still meant to drive users to follow those profiles.

Fifth, from up in Section I, thinking of instances as communities or subcultures is an incredible step forward in building real places online, but it’s important to accommodate the natural human tendency to identify with multiple communities. Being able to register with, say, up to five instances to read and write to their local timelines would be better than just allowing local and federated timelines or at least make it possible to follow more than one local timeline.

art and text in AI (on Chat.openai.com)

I’ve been working with art and Artificial Intelligence image generators recently. Untalented artists are scared of AI image generators since they are tools for making bad art, which is all that is being promoted in the press. All of the image generators—save perhaps for Dall-E 2—are tuned to produce such work in order to generate revenue. Ed Keller calls this “the black light black velvet painting of our time.” AI Art is likely to destroy ArtStation and Deviant Art. But we have a jobs crisis and need people to work in more important positions like food service. Who can argue that this re-allocation of human labor is bad?

Virtually all the work being done with this—even work that is lauded as exemplary—is bad. A few friends do interesting work. I will let them publicize it, although it’s safe to say Ed is interrogating these generators as a medium, pushing their boundaries in ways that are interesting to me. A recent article on the topic in the Architects’ Newspaper highlighted only bad work. Patrick Schumacher has convened a symposium on “futuristic” work indistinguishable from his own, which means that it is midbrow and pedestrian. Archdaily published a piece that gives the impression that AI interior designs are the stuff of nightmares. I could waste my time and yours coming up with more links to bad work.

This should not surprise anyone. Almost everything is bad because intelligent thought in architecture is now nearly extinct.

Making interesting work with AI image generators requires work. It doesn’t take as much time as it might to create a painting by hand, but an AI painting might take hours, a day, or more. One needs a plan and a mind for experimentation. AI art generators are incapable of originality. Neither are almost all artists working today. That should not stop anyone as until the modern era, originality was not valued. In the Renaissance, “genius” did not mean making an unprecedented work, rather it meant the ability to shape something pleasing out of existing sources.

AI image generators create forgeries. One appropriate response, then, is to use them to create explicit forgeries. I undertook this as a project starting this summer with the Lost Canals of Vilnius, the Drowning of Doggerland, and the Witching Cats of New Jersey.

Any good forgery begins a kernel of truth. Basing texts and images on existing sources, I created alternate histories. I used a feedback effect: provocative works created by the image generators inspired my alternate histories and ideas from the alternate histories guided the prompts for image generators. Most of the work was done with Dall-E 2 since it alone is tuned to produces semi-convincing paintings and photographs.

I unintentionally duped some readers. But AI art and text generators are already being used to deceive people for political purposes, to satisfy high school and college requirements, and for work purposes. Convincing people has never been easier. This project was intended to foreground this. I intend to write an essay for this site on the topic, but this one came out first. I am in Vilnius, Lithuania which is not boring like New York City. Visiting here often jars me out of complacency, spurs new ideas, and forces action.

A related route is to use AI art in a surrealist sense. This does not mean prompts for paintings in the style of Salvador Dali. That is kitsch and awful. Rather, this is a way of accessing what Jung called the collective unconscious to produce visual automatic writing: Pinot-Gallizio’s “Industrial Painting” updated for our day. In a sense, the “Witching Cats” project came out of a prompt for a black and white cat that came out gruesomely misshaped, became an amusing diversion for Instagram, led me to read about photography and the occult in the nineteenth century, and that, in turn led to the essay and project. Otherwise, I have not released this work. I may or may not.

Two days ago (30 November 2022), OpenAI released a freely-accessible interface to ChatGPT, a language model capable of conversation in the format of a dialogue. Although it is possible to trip it up, it is also remarkable how effective it is at generating content, even content that can be quite lengthy and complex in nature. I began by trying to produce critical essays.

In all of these, my prompt will come first.

>In 1,500 words, compare “Modernist Painting” by Clement Greenberg and “Art and Objecthood” by Michael Fried.

Modernist Painting by Clement Greenberg and Art and Objecthood by Michael Fried are two important essays that address the question of what constitutes modernist painting. While both essays address this question, they take very different approaches and come to somewhat different conclusions.

In Modernist Painting, Greenberg argues that the defining characteristic of modernist painting is its emphasis on the flatness of the canvas. He argues that the history of painting is one of the progressive flattening of the picture plane, and that modernist painting represents the culmination of this process. According to Greenberg, the flatness of the canvas allows the painter to focus on the formal elements of the painting, such as line, color, and composition, and to create a more intense and immediate visual experience for the viewer.

In contrast, Art and Objecthood takes a more philosophical approach to the question of modernist painting. Fried argues that the defining characteristic of modernist painting is its refusal to be reduced to a mere object in the world. He argues that modernist painting is defined by its ability to transcend the merely physical and to create a “presence” that is distinct from the physical object of the painting itself. According to Fried, this “presence” is what allows modernist painting to truly engage the viewer and to create a meaningful artistic experience.

One key difference between the two essays is their focus. While Greenberg is primarily concerned with the formal qualities of modernist painting, Fried is more interested in its philosophical and artistic implications. This difference is reflected in their respective conclusions. While Greenberg ultimately argues that the flatness of the canvas is the defining characteristic of modernist painting, Fried argues that it is the ability of the painting to create a “presence” that sets it apart from other forms of art.

Another important difference between the two essays is their treatment of the viewer. For Greenberg, the viewer is an essential part of the modernist painting experience. He argues that the flatness of the canvas allows the viewer to engage more directly with the formal elements of the painting, and that this creates a more intense and immediate visual experience. In contrast, Fried is less concerned with the viewer’s experience of the painting and more interested in the painting’s ability to create a “presence” that transcends its physical form.

Overall, Modernist Painting by Clement Greenberg and Art and Objecthood by Michael Fried are two important essays that address the question of what constitutes modernist painting. While both essays take different approaches and come to different conclusions, they both provide valuable insights into the nature of modernist painting and its place in the history of art.

>Write a 2,000 word analysis of the song “Rid of Me” by PJ Harvey

PJ Harvey’s 1992 hit “Rid of Me” is a raw and emotional exploration of the pain and anger that can arise in a toxic relationship. Through the use of aggressive and confrontational lyrics, coupled with her powerful and haunting vocals, Harvey exposes the desperation and helplessness that can arise when trying to leave a harmful relationship.

The song opens with a haunting and intense instrumental introduction, featuring distorted guitar riffs and driving drums that set the tone for the rest of the track. This instrumental introduction immediately establishes the song’s aggressive and confrontational atmosphere, setting the stage for the confrontational and emotional lyrics that follow.

The first verse begins with Harvey’s powerful and haunting vocals, as she sings: “I want you gone / But I want you still.” This opening line perfectly encapsulates the confusion and conflicting emotions that can arise when trying to leave a toxic relationship. On the one hand, the narrator wants to rid herself of her partner, but on the other hand, she still desires their presence and connection.

The chorus, which follows the first verse, features the repeated line: “Rid of me / Rid of me.” This chorus serves as a cry for help and a desire to be free from the suffocating and harmful relationship. The repetition of the line, coupled with Harvey’s powerful and emotive vocals, emphasizes the narrator’s desperation and longing for freedom.

The second verse continues with the narrator’s emotional turmoil, as she sings: “I’m an animal / I’m a dog / I’m a death / I’m a thief.” These lyrics paint a vivid and disturbing picture of the narrator’s state of mind, as she feels trapped and degraded in her relationship. The use of animalistic imagery also highlights the loss of humanity and dignity that can occur in a toxic relationship.

The bridge of the song features the line: “I’m a mess / I’m a wreck / I’m a fault / I’m a scar.” These lyrics further emphasize the narrator’s feelings of inadequacy and worthlessness in the relationship. She sees herself as a burden and a source of pain for her partner, and longs for the release of being free from the relationship.

The final verse of the song features the line: “I’m not a child / I’m not a pet / I’m not a toy / I’m not a doll.” This line serves as a powerful declaration of the narrator’s autonomy and independence. She rejects the idea of being treated as a possession or object, and asserts her own worth and dignity.

The song’s instrumental outro features a repetitive guitar riff that gradually builds in intensity, culminating in a ferocious and powerful finale. This instrumental finale perfectly captures the narrator’s emotional release and liberation as she breaks free from the toxic relationship.

In conclusion, “Rid of Me” is a raw and emotional exploration of the pain and anger that can arise in a toxic relationship. Through the use of confrontational and powerful lyrics, coupled with Harvey’s haunting and emotive vocals, the song exposes the desperation and helplessness that can arise when trying to leave a harmful relationship. The song’s intense and driving instrumentals perfectly capture the narrator’s emotional turmoil and eventual release, making it a powerful and impactful anthem for anyone who has experienced the pain of a toxic relationship.

Unfortunately, ChatGPT can make mistakes and there are mistakes in that essay. Later, I asked it to write about the relationship between Vytautas Landsbergis and George Maciunas and it stated that while they are both Lithuanians interested in Fluxus and music, they had no direct relationship, which is untrue. ChatGPT receives a failing grade for not meeting the required word count, although adding those phrases produces longer analyses.

Still, the above essays are better than what 95% of graduate students are capable of doing in an afternoon or, usually, a weekend. This should not be surprising as, while ChatGPT is incapable of original thought, so are 95% of graduate students and, indeed, 95% of faculty.

When I did my dissertation at Cornell, a dissertation was defined as a “contribution to knowledge” but the vast majority of such work in the humanities is formulaic: find an obscure topic related to an important topic (an unknown modern architect in Brazil in the 1960s who made middling buildings but met Lina Bo Bardi at a restaurant for dinner once or a failed, uninteresting project by someone who saw Le Corbusier speak) represented in an archive, go to said archive, find some obscure fact of no importance and blather on about how important said find is, add a conclusion in which Foucault is invoked to make the work “political,” and submit as a dissertation. Or at least that was a “good,” Ivy-level dissertation in the 2000s. Today one would have to add some random verbiage about marginalized communities, identity, and oppression to indicate that one is more “political.” Such work is not intended to be original, but rather to indicate that one has been properly disciplined in graduate school and is now capable of independently programming other students to also produce work of no importance whatsoever. If it is true we live in a simulation, most of these people are NPCs. Original thought is almost non-existent.

If an AI can equal or better such work today, albeit only in short pieces, imagine it in a few years.

ChatGPT can also be used to create instruction-based art. A series of these follow.

>produce a textual score by John Cage

0’00”: Silence
0’01”: A single piano note is struck
0’02”: Silence
0’03”: A second piano note is struck
0’04”: Silence
0’05”: A third piano note is struck
0’06”: Silence
0’07”: A fourth piano note is struck
0’08”: Silence
0’09”: A fifth piano note is struck
0’10”: Silence
0’11”: A sixth piano note is struck
0’12”: Silence
0’13”: A seventh piano note is struck
0’14”: Silence
0’15”: An eighth piano note is struck
0’16”: Silence
0’17”: A ninth piano note is struck
0’18”: Silence
0’19”: A tenth piano note is struck
0’20”: Silence
0’21”: An eleventh piano note is struck
0’22”: Silence
0’23”: A twelfth piano note is struck
0’24”: Silence
0’25”: A thirteenth piano note is struck
0’26”: Silence
0’27”: A fourteenth piano note is struck
0’28”: Silence
0’29”: A fifteenth piano note is struck
0’30”: Silence
0’31”: A sixteenth piano note is struck
0’32”: Silence
0’33”: A seventeenth piano note is struck
0’34”: Silence
0’35”: An eighteenth piano note is struck
0’36”: Silence
0’37”: A nineteenth piano note is struck
0’38”: Silence
0’39”: A twentieth piano note is struck
0’40”: Silence
0’41”: A twenty-first piano note is struck
0’42”: Silence
0’43”: A twenty-second piano note is struck
0’44”: Silence
0’45”: A twenty-third piano note is struck
0’46”: Silence
0’47”: A twenty-fourth piano note is struck
0’48”: Silence
0’49”: A twenty-fifth piano note is struck
0’50”: Silence
0’51”: A twenty-sixth piano note is struck
0’52”: Silence
0’53”: A twenty-seventh piano note is struck
0’54”: Silence
0’55”: A twenty-eighth piano note is struck
0’56”: Silence
0’57”: A twenty-ninth piano note is struck
0’58”: Silence

>produce a textual score by John Cage

  1. Open the piano lid.
  2. Place a single sheet of paper on the piano keys.
  3. Using a pencil, randomly draw lines and shapes on the paper, creating a visual score.
  4. Sit at the piano and interpret the visual score, allowing the lines and shapes to guide the movements of your hands on the keys.
  5. Explore different sounds and timbres by experimenting with various techniques, such as striking the keys with different parts of your hand or using the pedals.
  6. Allow the music to emerge naturally and organically, without predetermined melodies or harmonies.
  7. Continue to improvise and explore the piano’s soundscape for a set duration of time, determined by the performer.
  8. When the performance is complete, gently remove the sheet of paper from the piano keys and close the lid.
  9. Reflect on the experience and the unique sounds and melodies that emerged during the performance.

>write the instructions for a new Sol Lewitt pencil wall drawing

  1. Choose a wall in your space that has a minimum height of 8 feet and a minimum width of 10 feet.
  2. Measure the dimensions of the wall and divide it into equal horizontal and vertical sections, creating a grid pattern with a minimum of 10 horizontal and 10 vertical lines.
  3. Using a pencil, randomly select one of the grid squares and draw a line that extends from one edge of the square to the opposite edge, without crossing any other lines or touching the sides of the square.
  4. Repeat this process for each of the grid squares, creating a series of lines that fill the entire wall.
  5. Experiment with different line lengths, angles, and directions in order to create a varied and dynamic composition.
  6. Avoid erasing any lines or making any corrections, and allow the drawing to evolve organically.
  7. When the drawing is complete, step back and admire the final composition.
  8. Optional: photograph the drawing and share it with others.

>write the instructions for a new Sol Lewitt pencil wall drawing

  1. Choose a blank wall in a well-lit area to create your wall drawing.
  2. Using a ruler and a pencil, draw a horizontal line across the center of the wall.
  3. Draw a second horizontal line above and below the center line, each spaced at equal intervals.
  4. Draw a third set of horizontal lines, above and below the second set, at the same intervals.
  5. Continue adding sets of horizontal lines until the entire wall is covered with a grid of evenly spaced lines.
  6. Using a different color pencil, randomly draw diagonal lines connecting the horizontal lines.
  7. Experiment with different patterns and densities of diagonal lines, creating a unique and visually interesting composition.
  8. Step back and observe the overall composition, adjusting and refining the diagonal lines as needed.

>write an instruction piece in the style of a list by Yoko Ono

Choose an everyday object, such as a paperclip, a pencil, or a piece of string.
Place the object in a simple, unadorned setting, such as on a blank sheet of paper or a plain background.
Using a camera or smartphone, take a photograph of the object in its chosen setting.
Share the photograph on social media, along with a simple and enigmatic caption that invites viewers to interpret the object in their own way.
Consider how the chosen object and its placement in the photograph might evoke emotions, thoughts, or memories in the viewer.

>write an instruction piece in the style of a list by Yoko Ono

Find a quiet, peaceful space where you can be alone and undisturbed.
Close your eyes and take a few deep breaths, allowing your body and mind to relax.
Slowly open your eyes and look around your surroundings, taking in the details and textures of the space.
Imagine yourself floating in a vast, empty void, free from the constraints of time and space.
Allow your mind to drift and explore, without judgment or expectations.
Listen to the sounds around you, and allow yourself to be present in the moment.
As thoughts and ideas arise, allow them to flow freely, without attachment or resistance.
When you feel ready, write down any thoughts or ideas that emerge during your meditation.
Take your time, and trust in the process.
When you are finished, close your eyes again and take a few deep breaths, grounding yourself back in the present moment.
Reflect on your experience, and let go of any judgments or expectations.
Enjoy the feeling of freedom and openness that comes from exploring your inner landscape.

All of the artists I have chosen employ deliberate repetition, variations within a framework, and an element of chance. I produced these pieces in an hour and yet they are plausible variations on their work. Not all attempts were as successful. It is, however, evident that one could, for example, generate a large number of these to hand out to an audience at a performance so for participants to perform and evaluate the results.

AI has been rapidly developing this year and ChatGPT is only an early step in this decade, likely to be dominated by AI. Serious artists need to ignore the bad art and forge ahead. AI image and text generators are the tools of our time.

Regarding Twitter

I have another, much more interesting and important thing to talk about—e.g. the witching cats of New Jersey—but enough people have asked me for my hot take on the acquisition of Twitter by Elon Musk that I’ll take a stab at it. I will get back to my regularly scheduled programming in the very near future, but for now, I am going to celebrate the fact that moving this site to kinsta earlier this week has made a rapid response like this one much easier.

In the 2005-2006 academic year, I led a team of scholars at the Annenberg Center for Communication in researching the topic of “Networked Publics.” This project led to a book of the same name, published by the MIT Press in 2008 and an aborted project on the broader topic of network culture that faced too much opposition from entrenched interests in the academy to see the light of day. Our little group didn’t always see eye-to-eye, but if there was anything we could agree with, it was the grim conclusion of the brilliant politics chapter written by Merlyna Lim and the late Mark E. Kann: although the Internet is a powerful resource for mobilization, it is a poor venue for democratic deliberation.

To this day, we have absolutely no evidence otherwise. The dramatic rise of algorithmically-produced content feeds in the nearly twenty years since our research year has made matters much worse. As algorithms respond to user engagement and reaction, they deliver the content that users want to see, creating a spiral of ever-increasing polarization (for a series of links to studies on this matter, see this piece from the Brookings Institution). Making matters worse are the social media “cops” on both sides who thrive on attacking viewpoints divergent from accepted consensus. If you say that vaccines are important and that the 2020 election was legitimate but your social media “friends” are right-wingers, you’ll get called out by an army of Trumpenproles if you say so. If you say that magic monetary theory is insane or that defunding the police will hurt African Americans more than anybody, you’ll get publicly shamed by the Internet cops on the Left. Then there are things that we know very well not to talk about with our peers. To take one example, virtually no academic would publicly say that doctoral programs are in crisis and universities are producing vastly more PhD.s than there will ever be jobs for just so that doctoral chairs can gain status in the academy and professors can get free research assistants, even though most academics I have spoken to about this agree wholeheartedly off the record. I am outside the academy so I can freely say this. But nobody in the academy can. The cops there have the power to destroy unorthodox thinkers so nobody will say it. Even outside academia, there are positions so politically dangerous that I won’t dare utter them out loud even here, notwithstanding that the vast majority of my peers agree with them privately (you’ll just have to guess, so sorry). Internet cops gain symbolic capital by dismissing ideas that don’t conform to the orthodoxy and it’s been interesting to observe as a few cops have, coincidentally or not, attained positions of minor leadership in the academy (probably because no sane person wants them). Cops tend not to think for themselves, but rather they are guided by what Venkatesh Rao calls beef-only thinkers, who demand unqualified support, folks like Glenn Greenwald, Michael Tracey, or Marjorie Taylor-Greene.

If, on the one hand, we have a social media landscape actively patrolled by cops, we also have a steady-state of outrage from the woke and the trumpenproles: everything is burning, right now/there has never been a more consequent election/there has never been a greater threat to democracy/etc. The news media—particularly TV news, but also online personalities—amp up the rhetoric in order to profit from the clicks. Outrage makes money, even as it makes us stupider. But after years and years of outrage, people are exhausted. It’s very much like pornography, searching for ever-greater stimulus, things got more and more extreme and eventually it wasn’t that people got scandalized, it’s that they got bored. Only the most politically active and the most insane want to the outrage to continue. After the nightmarish calamity following the 2016 election, COVID, the January 6 insurrection, and then the invasion of Ukraine, high levels of cortisol and adrenaline have literally taken their toll on our bodies, creating inflammatory reactions that leave us with no way to absorb more bad news. It may have been funny when a demented old grifter demanded to see Obama’s birth certificate, it was hilarious to see “America’s Mayor” stand up in front of the garage of a landscaping firm, burbling nonsense, his false teeth nearly falling out of his head, but nobody wants to see it again. In an economy sustained by growth, Facebook’s aging user group is down by significant numbers, and analysts are talking about a “death spiral.” Well, thank goodness. Nor is the subject of today’s diatribe, Twitter, immune, its active users are also fleeing, just in time for Musk’s ill-timed purchase. The only solution people are finding to this anxiety is disconnection, leaving all this crap behind. It will likely cause a rout for the Democrats in the 2022 election (I am not, however, entirely convinced of this), but if it does, it will be the far Left’s relentless barrage of alarmist news that will be, in large part, to blame.

What are the options? Clearly not Twitter. Elon probably should fire most of Twitter’s staff. Around 5,000 employees work at Twitter and, judging from the evidence, none of them save the people who make sure the servers don’t go down, do anything. The site has had virtually no innovation since its launch, the code is a notorious mess, and, well what do they actually do all day long?

Journalists like to promote TikTok as an alternative, but TikTok’s growth is limited to a high school and college age demographic in search of diversion. What political content exists there—a Johns Hopkins study suggests— is hardly any better than what can be found at Facebook and, in any event, like YouTube, TikTok is oriented less toward people producing their own content and more toward passive consumption and commenting. Instagram is not dissimilar, although there, the content-production tends to be aspirational and imitative in nature, contributing to body dysphoria and leading young women to seek to surgically reshape themselves into the “Instagram face” look pioneered by Jocelyn Wildenstein. The one healthy (for now) antidote for this is BeReal, an anti-Instagram that values immediacy and promotes looking real, or at least, candid and terrible, it’s only drawback is that virtually nobody except for college-age kids uses it (I know, you can visit mine, I have one follower). TIkTok and Instagram are terrible platforms for political messaging—although some will try—and BeReal is virtually useless for it. This, of course, is their charm.

If there are genuine alternative social media spaces right now, they are Discord and Substack. Both are flawed, but it’s Substack—not Twitter—that holds the only potential for a future social media platform right now.

Discord is a set of “servers” (not really servers, but they are called that, they are virtual spaces for micropublics) dedicated to a given topic, e.g. Minecraft gaming, Roblox, GTA, (Discord started as a space for emergent gaming communities), online generative art generators, techwear, eurorack synthesizers, alt-space, AI generative art, Arduino programming, white supremacism (this has since been banned) or whatever excites you. Discord is growing rapidly, but it has two major limiting factors. The first is that because of the conversation-like organization, unless one is actively engaged with a server, one rapidly loses track of what is being talked about. Catching up is neither intuitive nor, well, interesting. The second is that there is no broader link between these micropublics. Each server is a walled silo and there is no communication between them.

Substack is a platform for content-creators—mainly writers but also some podcasters— in which subscribers pay subscriptions for content and content-creators are promised income directly from their subscribers. Unlike the similar Patreon, however, it is more oriented around discovery and community. Log into the Substack site and you will be shown a sidebar with recommendations for other Substacks that are, shades of Facebook and Twitter, algorithmically recommended for you. Subscribe to a right-wing Substack, you’ll definitely be offered more. So, basically, another Facebook or Twitter. The up side is that Substack allows comments (depending on settings), and also has recently allowed users to incorporate RSS feeds from outside of Substack into their feed. If—and it’s a big if, one that likely won’t take place since it’s only my idea, not Substack’s—the platform can find a way to create glue between Substacks and users—such as making it easier for users to follow each other and talk directly to each other— it has some hope. The reason I hold any hope for Substack is that unlike Facebook or Twitter, it promotes long-form writing. This post makes no sense on Facebook or Twitter, but it will easily work on Substack. The second hitch is that Substack is heavily tied to a subscription economy. Most newsletters, it seems, cost about $10 a month. That’s great but can rapidly become unaffordable in an era where we are already paying for one or two newspapers, a couple of magazines, Netflix, Spotify, Hulu, and you get the idea… People are getting tired of subscriptions, very tired.

This leaves us with with WordPress. WordPress is not only the dominant worldwide blogging platform, it has a reader that allows readers to follow blogs and participate in dialogues in the comments. Again, it’s a long shot, but users are also implicitly encouraged to become bloggers and, well, why not? In the 1970s and early 1980s mass media seemed to be all-powerful, but then we had the Zine revolution and the explosion of the Internet, back when it was still fun and potential seemed everywhere. Substack and WordPress may not be the future, but I still think a platform with some future in it (I am no longer comfortable thinking it will be any lasting solution, let alone a utopian space of deliberative democracy) is coming if we have any hope of talking to each other online.

A final reflection on all this is how deeply sad this is for Elon Musk. Although he has recently shifted from being an icon of the Left to being an icon of the Right, he is the Steve Jobs of this age, having made both electric cars a reality and creating the first successful reusable rocket system. Ten years ago, even five years ago, owning two electric cars by 2022 would have seemed entirely implausible to me and yet, we replaced both our cars with Teslas in 2020. These cars require less maintenance than any other vehicles we have ever owned (thus far, our total repairs involve a heat pump valve on the Model Y and some issues with the rear gate on our Model S, plus some cabin air filter replacements oh, and new tires), have excellent performance (I’m a car guy, but my wife loves driving hers as well), and it is delightful to skip the weekly trips to the gas station. SpaceX’s phenomenal success speaks for itself (not only am I car guy, I’m a space nut) and Starship promises to revolutionize the space industry, and even though I am concerned about Starlink’s impact on astronomy, it has the ability to deliver secure Internet communications at broadband speeds virtually everywhere and has made a big, positive impact in military operations in the Ukraine. That, for whatever reason, Musk is spending his time and money on Twitter is very sad. If he wanted to have fun, wouldn’t it make better sense to just book a flight on a crew Dragon? I would. It’s not like he doesn’t won the only company that ever regularly sent people into outer space; eat your own dog food, Elon. Running Twitter seems incredibly boring. Moreover, Elon has a long list of failures to go along with the successes—Tesla Solar (I just put a new roof on my house, why wasn’t it from Tesla Solar?), Tesla’s Full Self-Driving has been around for almost two years and hasn’t gone anywhere (remember when Elon said we’d have self-driving Tesla taxis by 2020), and Tesla not only hasn’t released a single new vehicle since 2020, it has raised prices on existing vehicles while cutting out features such as front-facing radar and rear-facing ultrasonic sensors but hasn’t found ways to compensate. I own Tesla stock but I think they are in trouble unless Elon spends some time there soon and Twitter is a distraction he hardly needs.

Finally, as far as the future of Twitter. Who cares? It is a dying platform. I have met some great friends on it—many more than on Facebook—but if it dies and Facebook dies, it will be better for everyone. If Captain Dementia joins back up over the weekend, or Monday, or whenever, what does it matter. It’s not like anybody listened to him on Truth Social anyway.

2020 in Review

Grayness in Maine
Grayness on Mount Desert Island, 2020.

According to conventional chronological schemas, 2020—not 2019—is the last year of the 2010s.* This is convenient since, as I pointed out in last year’s premature review of the last decade, the 2010s were “the decade of shit” and 2021 is a stinking pile of shit. The worst decade since World War II ended with the worst year since 1945.

My “year in review” posts are usually almost as late as my taxes and when I finished last year’s post on February 12, we were all well aware that COVID was out there. Now, no question that I missed the severity of the pandemic back then, but I was on the money about its psychic effects. For all of the horror of COVID, it isn’t horrible enough. COVID is banal. Instead of bleeding out through all of our orifices as with Ebola, COVID is “a bad case of the flu” that leaves people dead or with debilitating cardiovascular and neurological ailments. But how different is my diagnosis, really, from what happened?

Now sure, this year [2020] we’ve already had firestorms the size of Austria ravaging Australia, a rain of rockets in Baghdad, Ukrainian jetliners getting shot out of the sky, a deadly pandemic in China caused by people eating creatures that they really shouldn’t, and the failure of the Senate to uphold the rule of law, but the banality of it all is crushing. While the Dark Mountain set drinks wine around a campfire, gets henna tattoos, and sings along to songs about the end of nature, for the rest of us, it’s just an exhausting, daily slog through the unrelentingly alarming headlines.

COVID brought us yet more crushing banality. The Idiot Tyrant is gone, but we are trying his impeachment yet again. Everything changes, but nothing changes. We were all the Dark Mountain set this year, sitting around our campfires, singing songs about the End. It was another atemporal slog, one day bleeding into another, every day a Sunday in a country where everything is closed on Sundays and there is nothing to do, every day stranger and more disconnected than the last, something captured in comedienne Julie Nolke’s series of videos entitled “Explaining the Pandemic to My Past Self.”

Amidst the disconnection, the Jackpot—or William Gibson’s term for a slow-motion apocalypse—cranked up a couple of notches. Just surviving the year was an accomplishment. The balance of life has been thoroughly disrupted and that disruption isn’t going away any time soon. It’s not just COVID: we now feel certain that there will be more pandemics, more massive wildfires, and more superstorms in our future. The Earth isn’t dying (sorry, climate doomers), but there will be huge losses of species worldwide, human population decline is well underway in advanced societies (the US is finally on the bandwagon here), and massive deaths will take place across the planet until the population comes back to a sustainable level decades from now.

But the premise of the Jackpot is that it isn’t a final apocalypse: there will be another side. In his Twitter feed (@GreatDismal), even Gibson focuses on the horrific and unjust nature of the Jackpot, but there will be winners, selected on the basis of wealth and sheer dumb luck. What might this say about the US election and the fact that 46% of Americans voted for a cretin? Now, there is nothing particularly new about melding Tourette’s and dementia into a public speaking style, there are plenty of lunatics sitting on their porches screaming obscenities at their lawn ornaments. Everybody knows that Uncle Scam’s persona as a billionaire—or rather the King of Debt (his own term!)—is an act. The man with the golden toilet is not a successful businessman. He is weak, a loser who can’t stay married or stay out of bankruptcy court. Four years of misrule ended in abject failure: defeat in both electoral and popular votes, being banned from social media and, with his businesses failing, being forced out of office in shame to face an unprecedented second impeachment, an array of civil litigation as well as criminal indictments for fraud, tax evasion, incitement to riot, and rape. But this—not a misguided notion of him as a success—is the real point of his appeal. The short-fingered vulgarian is a life-long loser, a reverse Midas whose every touch turn gold to lead. But in the face of the Gibsonian Jackpot, his appeal was not as a stupid version of Homer Simpson, grabbing whatever scraps he can and, when that failed, LARPing as President, destabilizing society, and just blowing everything up.

LARPing was big in 2020, which saw the attempted kidnapping of Michigan Governor Gretchen Witmer by wingnut idiots, various insane protests by COVID deniers, the attempted coup of the Capitol Insurrection, and the riots developing after the Black Lives Matter protests. BLM was the standout among these, not only a good, just cause, but also because the majority of the protests themselves were peaceful—such as the one in our town of Montclair, New Jersey. None of that was LARPing, but the riots that accompanied it were. For the most part, this was less people with genuine greivances and more Proud Boys, Boogaloos, anarchists, and grifters who came in to loot and burn whatever they could down. Although there were kooky moments on the Left like the Capital Hill Automonous Zone, Antifa, for however much it exists, didn’t do much, certainly proving to be far less trouble than white supermacist-infiltrated police forces in paramilitary gear. Still, the widely-vaunted second Civil War never came about and the arteriosclerotic LARPers on the Right limped off the field in defeat after their they got a spanking at the January putsch.

A number of observers at both the Capitol Insurrection and CHAZ —including some of the idiots who took part in it—noted that these events felt much like a game, specifically an Alternate Reality Game (ARG). In a typical ARG, players look for clues both online—think of the QAnon drops, the Trumpentweets, or the disinformation dished out by the skells at 4chan, 8chan, and so on—as well as out in the world. Jon Lebkowsky, in a post at the Well’s State of the World and Clive Thompson over at Wired compare QAnon to an ARG. Indeed, gaming is taking the place of religion (whichever grifter figures out how to meld this with Jesus and his pet dinosaurs will get very rich indeed), with the false promise that playing the game and winning will deliver one to the other side of the Jackpot. Somewhere, I read that when asked what he would do differently if he had made Blade Runner a decade later, Ridley Scott replied that he would be able to skip the elaborate sets and just point the camera down the streets of 1990s Los Angeles. Today, the same could be said for the Hunger Games today.

But not everything was LARPing. If Cheeto Jesus is an icon for LARPing losers, Biden was elected on the premise of staving off the Jackpot by returning adults to the White House. This is not a bad thing, we might as well try. Still, from the perspective of Jackpot culture, the most interesting political development of the year was the candidacy of Andrew Yang whose cheery advocacy of Universal Basic Income (aka the Freedom Dividend) masked the dark, Jackpot-like nature of his predictions. Let’s quote Yang’s campaign site on this: “In the next 12 years, 1 out of 3 American workers are at risk of losing their jobs to new technologies—and unlike with previous waves of automation, this time new jobs will not appear quickly enough in large enough numbers to make up for it.” No matter how friendly Yang’s delivery, there is a grim realism to his politics, an acceptance that things will never be better for a massive sector of the population. Certainly some individuals will find ways to use their $1,000 a month freedom dividend as a subsidy to do something new and amazing, but 95% will not. Rather, they will form a new and permanent underclass as they fade into extinction. Again, the point of Yang’s candidacy isn’t the cheerleading for math and STEM, it’s the frank acknowledgement that the Jackpot is already here.

On the other hand, toward the end of the year, Tyler Cowen suggested that we might be nearing the end of the Great Stagnation (he is, of course, the author of an influential pamphlet on the topic) and you can find a good summary of the thinking, pro and con by Cowen’s student Eli Dourado here. In this view, advances such as the mRNA vaccine, the spread of electric, somewhat self-driving vehicles, the pandemic-induced rise of remote work, and huge drop in the cost of spaceflight are changing things radically and could lead to a real rise in Total Factor Productivity from the low level it has been stuck at since 2005. Is this a sign of the end of the Jackpot? Unlikely. That won’t come until a series of more massive technological breaks, probably (but not necessarily) involving breakthroughs in health (the end of cancer, heart disease, and dementia), the reversal of climate change, working nanotechnology, and artificial general intelligence. But still, there are signs that early inflection points are at hand.

Personally, we experienced one of these inflection points when we replaced our aging (and aged) BMWs with Teslas. I wound up getting a used Tesla Model S last January and then immediately turned around and ordered a brand new Model Y that we received in June. No more trip to the gas station, and while “Full-self driving” is both expensive and nowhere near fully self driving, it is a big change. Longer road trips—which under the pandemic have been to nurseries on either side of the Pennsylvania border to buy native plants—have become much easier, even if I still have to keep my hands on the wheel and fiddle with it constantly to prevent self-driving from disengaging. But harping on too much about the incomplete nature of self-driving is poor sport: in the last year, Tesla added stop light recognition to self-driving and a new update in beta promises to make city streets fully navigable. Less than a decade ago, self driving was only a theoretical project. Now I use it for 90% of my highway driving. That’s a sizable revolution right there. Also, the all-electric and connected nature of these cars makes getting takeout and sitting in climate-controlled comfort in my vehicle when on the road a delight. Electric vehicles were a big success this year and in our neighborhood which is a bellwether for the adoption of future technology (when I saw iPhones replace Blackberries on the bus and train into the city, I bought a bit of Apple stock and made a small fortune) and Teslas have replaced BMWs as the most common vehicle in driveways.

Back to the pandemic, which accelerated a sizable shift in habitation patterns. Throughout the summer, there was a lot of nonsense from neoliberal journalists and urban boosters about how cities are going to come back booming, but with more bike lanes, wider sidewalks, less traffic, and more awesome tactical urbanist projects to appeal to millennials. Lately, however, those voices have fallen silent and with good reason. In this suburb the commuter train platforms are still bare in the mornings and the bus into the city, once packed to standing room only levels every evening, hasn’t run in five months. A friend who works in commercial real estate says that occupancy in New York City offices is at 15% of pre-pandemic levels. Business air travel is still off a cliff. Remote work isn’t ideal for everyone and every job, but neither was going into the office. For sure, the dystopian open offices, co-working spaces, and offices as “fun” zones are done and finished. People are renovating their houses, or upsizing, to better live in a post-pandemic world of remote work. Another friend who works for a large ad agency told me that they did not renew their lease for office space and do not plan to ever go back to in person work, at least for the vast majority of the staff. When employees gain over two hours a day from not commuting and corporations save vast fortunes on rent, remote work seems a lot more appealing. Retail sales here and in the surrounding towns have gone through the roof, just as they have in many suburbs.

But it isn’t just suburbia that has prospered at the expense of the city, exurbia has returned too. Way back in 1955Auguste Comte Spectorsky identified a growing American cultural class that he dubbed “the exurbanites” made up of “symbol manipulators” such as advertisers, musicians, artists, and other members of what we today call the creative class. Spectorsky observed that many of these individuals eventually tended to drift back to the city. This time may be different. After two decades in the city, the creative class is turning to places outside the city with attractive older houses and midcentury modern properties, walkable neighborhoods (virtually all of Montclair, for example, has sidewalks), good schools (which generally mean high property taxes but are an indicator for a smarter, engaged populace), amenities like parks and places to hike, decent bandwidth, as well as independent restaurants, shops, and cultural attractions. There will always be variations in taste: some people really do want to eat at Cheesecake Factory and live in a Toll Brothers McMansion, but these will appeal to relatively few of the people fleeing cities at this point. Thus, the Hudson Valley—full of older, more interesting architecture, great natural resources and quirky towns—is booming. I predict some reversion to toward the mean after the pandemic ends and some of the people who fled to the country realize they aren’t suited to a place without Soulcycle, but this will be only a partial and temporary reversion.

I predict that even after the pandemic ends, there will be a greater interest in self-sufficiency among young people who move to suburbia and exurbia. Manicured laws will be less important than vegetable gardens. Homesteading, permaculture, and a drive back to the land not seen since the 1960s are under way. It would be a very good thing if the next generation was more in touch with their land and less prone to hiring “landscapers” who treat properties as sites subject to industrial interventions such as chemical fertilizer for lawns, a phalanx of gas-powered lawn mowers and leaf blowers to remove any stray biological matter.

As far as cities go, the pandemic is triggering a necessary contraction. The massive annihilation of real estate value it has caused should go a long way to undo the foolish notion that urban real estate is always a great investment. It’s not, just ask anyone who bought a house in Detroit in 1965. Real estate in first and second tier global cities has become wildly expensive, disconnected from the underlying fundamentals. When individuals are paying rents that absorb over 30% of their salaries to investor-owners who are not covering their mortgage with those rents, something is very wrong. This broken system has been able to function due to the perceived hedonic value of restaurants, bars, and cultural events, but these things too have been failing over recent years. Long prior to the pandemic, the cost of rent decimated independently-owned restaurants and retailers, with the latter also hurt by on-line shopping. The golden age of dining out (if it really was the golden age… I would say that better food could have been had in other, less copycat eras) was already declared over in 2019. “High-rent blight,” in which entire streets’ worth of storefronts were empty due to ludicrous rents, has been common for some time now. Tourists made up more and more of the street crowds while loss-leader flagship stores for chains like Nike and Victoria’s Secret replaced local businesses. With the hedonic argument for staying in the city rapidly disappearing, it was only a matter of time before individuals began departing and, in New York, population had begun to drop by 2018 (see more on all of this in Kevin Baker’s piece for the Atlantic, “Affluence Killed New York, Not the Pandemic”). Perversely, this is a good thing as it will likely lead to a bust in commercial real estate prices and a decline in unoccupied or AirBNB’d apartments, thus making global cities like New York places that have potential again. Moreover, many second tier cities such as St. Louis, Kansas City, and Cleveland are experiencing new growth as individuals able to work remotely are looking for places that are less expensive—and thus have more potential—than New York or San Francisco.

These shifts are huge and for the better. As I tried to tell my colleagues at the university, there is no housing crisis, at least not in the US and Europe, there is only an appearance of one because of the uneven distribution of housing: a glut in some areas, a shortfall in others. The pandemic has likely undone this a bit. Of course, places that are too politically Red, too full of chains, too full of copycat McMansions are unlikely to come back anytime soon, if ever. The Jackpot continues.

Still, I’m observing a perversely rosy future for the urban (and suburban and exurban) environment is the Biden administration’s interest in infrastructure. Back in 2008, I shocked design critics when I stated that there would be no progress in infrastructure for the foreseeable future. “But, Obama,” they complained. “But, Obama,” I clapped back, “just appointed Larry Summers as his chief economic advisor and Summers will bail out the banks, not fund infrastructure.” I expect the opposite from Biden who has adopted a “nothing left to lose” position as purportedly one-term President, is a devotee of train travel and is eager to make great progress on climate change. Appointing Pete Buttigieg, one of his two smartest opponents in the primary (the other being Andrew Yang, of course), to Secretary of Transportation is a key move. This will be Buttigieg’s opportunity to prove himself on the national stage and he will fight hard to do that, just as Biden expects. Expect more electrification across the board and, I suspect, more advances with self-driving vehicles. Although certain measures—such as, in the New York City area alone, the Gateway Tunnel between New Jersey and New York, now delayed over a decade thanks to Chris Christie and Donald Trump’s vindictiveness against commuter communities that would not vote for them and the reconstruction of Port Authority Bus Terminal—will help cities, again, I predict more emphasis on decentralization and activity outside the city.

All this may have salutary cultural implications. The global city is played out. Little of interest happens in New York, San Francisco, London, Paris, or Barcelona. These cities are too expensive for the sort of experimentation that made them great cultural centers and the diffusive nature of the Internet, capitalism, and overtourism have made them all the same. Residents of cities that have been victims of overtourism have seen this as an opportunity to reset, while the physical isolation of cities is going to increase reliance on local institutions. With some luck, all this leads to a new underground, with greater difference creating greater diversity and potential. Of fashion, Bruce Sterling writes, “Fashion will re-appear, and some new style will dominate the 2020s, but the longer it takes to emerge from its morgue-like shadow, the more radically different it will look.” The same could be true of all culture. Globalization was an incredibly powerful force but has been played out. I don’t agree with the protectionist instincts of the Trumpenproles but today culture’s hope is to thrive on the basis of the difference between places and cultures, not on greater sameness. Architecture has been very slow to react to all of this, in part because many intelligent young people have drifted into other fields, like startups, but I am optimistic that we might soon get past the ubiquitous white-painted brick walls and wood common table (the architecture of the least effort possible, to match fashion and food driven by the least effort possible), the tired old Bilbao-effect, and quirky development pseudo-modernism.

So much optimism on my part! Even I am shocked that I am so positive. But why not? The end to this exhausted first phase of network culture is overdue. Time for a new decade, at last.

*The reason for this is that there is no Year Zero. 31 December 1BC is followed by 1 January 1AD.

I am trying to break the Internet

I don’t see how we can remain enthusiastic about network culture. In the decade since the release of the iPhone, the Internet has gone from being a playpen for geeks and outsiders to the primary theater for politics and culture. Even three years ago the thought that a major global leader would use Twitter to announce major policy initiatives would have seemed futuristic and a little naïve. Now we have it and it’s the darkest time most America has collectively experienced in memory. We lurch headlong into a future, but it’s a new bad future.

And, as we seem to be drowning in information, we seem to have lost our ability to communicate and absorb knowledge. Almost nobody reads and writes blogs anymore (please don’t get me started about Medium and the final, thorough destruction of independent content its startup model is premised on). Magazines and journals are well and truly dead. Most books by theorists and academics are soundly ignored too, which is probably a good thing given that theory has become permeated by a neo-fascist identity politics. Outside of the telecocoon of our partners, children, closest friend or two, and immediate work associates, we no longer call each other, we no longer e-mail each other, and we ghost each other as much as we text each other. Earlier forms of Internet culture are also on the rocks: listservs have become replaced by Facebook groups that produce no thought or discussion of any substance whatsoever, and most online forums have died as well. The aforementioned Twitter should be dead, but is kept alive by our desire to see what the lunatic in the White House will say next, and otherwise serves as home to a few misfits and general oddballs. Facebook absorbs everything, reducing all human communication to nothing and algorithmically directs us to only see those posts that give us a fleeting satisfaction. It’s more imagistic spawn, Instagram, is the model of the new Internet, driving us to constantly one up each other with a lifestyle pornography. There are, of course, exceptions—I actually like Reddit and there are some niche online forums that have moments of productivity—but it’s bad out there. I have played my own part in migrating to these awful commercial platforms and regret it. Something must be done. Endless promises and little delivery.

In the case of this site, I’ve resolved to fix matters over and over again, but each time was undone by the content management system, Drupal. Drupal is awful. Way back in 2005, Drupal seemed like a good choice, with its module-based open architecture, and the promise of a content management system that could go far beyond a blog. At the time, I was fascinated with the idea of the networked book and Drupal seemed to offer such functionality built in. Unfortunately, Drupal long ago began to resemble a 1980s American car company, suffering from over-complexity, putting design and user interface last, and unable to got basic features working. A critical flaw is that the development team long ago decided that “the drop is always moving”: each major version of Drupal breaks all the existing plug-ins and themes without which a site is ugly and limited. And therein lies the rub. I should have updated my site eight years ago when Drupal 7 was released, but the horribly botched release of that version would have broken my site thoroughly if it had been possible to update it at all (I am now forgetting if it was Drupal 7 or 8 that did not have upgrade capabilities when first released and the upgrade cycle to Drupal 6 cost me over a month of work). Add to that constant trouble with excessive memory overhead, the obliteration of comments by spambots, together with a general feeling that the whole creaking mess was going to explode like a steam-powered Soviet tractor meant that I knew I’d never be upgrading Drupal. But where to go?

A New Year’s resolution for 2017 had been to redo my site in a new content management system and I upgraded the front end of my site to Kirby. Kirby is fantastic. It took me an afternoon to build the site (I am still running Kirby over at the network architecture lab. In Drupal it would have taken a week. Still, although now I had a decent portfolio together, the blog languished and I made a handful in 2017 and only one post in 2018. In part, this was due to life: I am still cleaning up after the decades of neglect that our house suffered before we bought it, I am spend more time with my family, and I am working on my conceptual art and sound practices. Still, doing anything in Drupal was a nightmare and I stayed away from any substantive blogging.

So it happens that we were away skiing at Smuggler’s Notch in Vermont this week and, as so often happens, we had freezing rain all day. So I decided to finally upgrade this blog and move it to WordPress. I had used WordPress before in Jo-Anne Green’s Networked book project and that had pushed the platform beyond what it was capable of at that time, leaving me with a bad opinion of the platform, but during the intervening years, WordPress has matured into a capable platform (even with the recent growing pains caused by the Gutenberg blogging interface).

Frédéric Gilles’s amazing Drupal to WordPress plugin imported all of the data from Drupal—even comments!—better than I had ever dreamed was possible, better than I would have expected from an upgrade within Drupal. I’ve long loved Indexhibit and use it on AUDC’s site so I was glad to base the new site on Leanda Ryan’s Inxhibit theme (much as I love Indexhibit, it simply isn’t designed for blogging). A day of work later and, although there are some bugs here and there, I am confident enough to replace Varnelis.net with WordPress.

One of the major impediments to blogging that I faced with Drupal is that inputting text on the Web is a nightmare (I can’t tell you how many posts I’ve lost by accidentally hitting the wrong key on Drupal) and uploading text inevitably seemed to introduce formatting issues, no matter how hard I tried. In the case of WordPress, I knew that there were many more tools available at my disposal so I decided to try iAWriter and found it worked flawlessly. I started writing this post on my Mac, seamlessly picked it to my iPad Pro and posted this entry. This is simple, the way Content Management Systems were supposed to be, without losing the independence that blogs make possible.

I’ve also put Feedly front and center for daily reading on my iPad. There are plenty of great blogs out there are acting as a resistance to the managed content of Facebook, Twitter, and Medium and I am setting out to rediscover them. With these changes to my blog, maybe, just maybe, I may once again join them as well.

 

Network Histories at Michigan, 3/8/18

I am delighted to be delivering the keynote address at the P+ARG Conference at the University of Michigan on March 8, 2018, at 6pm. More details here

My talk is titled "Network Histories: Baran and Milgram in Perspective" and the abstract reads roughly … 

The foundational work done by social psychologist Stanley Milgram and telecommunications researcher Paul Baran on networks in the 1960s remains profoundly influential today, establishing the basis of network theory. But both projects are more complicated than they seem: Milgram’s famous “Six Degrees of Separation” appears to have been largely fabricated while Baran’s plan for a “Distributed Network” is inevitably read within a retrospective mythography. This talk sets out to uncover not so much a theory of networks as an ideology of networks, seeking not a celebration but rather an understanding.