Showing posts with label internet. Show all posts
Showing posts with label internet. Show all posts

Thursday, May 14, 2026

The Impact of AI-Generated Text on the Internet

Jonas Dolezal, Sawood Alam, Mark Graham, and Maty Bohacek, The Impact of AI-Generated Text on the Internet.

Abstract: The proliferation of AI-generated and AI-assisted text on the internet is feared to contribute to a degradation in semantic and stylistic diversity, factual accuracy, and other negative developments (sometimes subsumed under the “Dead Internet Theory”). What has hindered answering these questions is that it has not been understood just how much of the internet is actually AI-generated or AI-edited. To this end, we construct a representative sample of websites published on the internet between 2022 and 2025 using the Internet Archive, and apply a state-of-the-art AI text detector on them. We find that by mid-2025, roughly 35% of newly published websites were classified as AI-generated or AI-assisted, up from zero before ChatGPT’s launch in late 2022. We also find statistically significant evidence for some of the identified hypotheses; for example, that increases in AI-generated text on the internet correlate negatively with semantic diversity and positively with the prevalence of positive sentiment. We do not, however, find statistically significant evidence supporting the hypothesis that an increased rate of AI-generated text on the internet decreases factual accuracy or stylistic diversity. Notably, this diverges from public perception, which we measure in a user study, where the majority of US adults turned out to believe in all four of the above-mentioned hypotheses. Individuals who do not use AI or use it infrequently tend to believe in these negative impacts more than those who use it frequently; similarly, individuals who hold negative views of AI tend to believe in these hypotheses more than those with favorable views of the technology.

From the introduction:

Ever since ChatGPT first made large language models (LLMs) available to the wider public in 2022, which was followed by mass adoption, there have been concerns about the impact of AI-generated text (as well as AI-generated content in other modalities) on the internet and online discourse (Ferrara, 2026; Muzumdar et al., 2025). Specifically, many known limitations and failure modes of LLMs, including factual hallucinations (Huang et al., 2025), sycophancy (Malmqvist, 2025), verbosity (Saito et al., 2023), and more, have raised concerns that unchecked proliferation of such content could reduce the overall quality of internet content (Shumailov et al., 2024; Xing et al., 2025). These hypotheses are sometimes subsumed under the “Dead Internet Theory,” which they loosely expand, but which, on its own, predates the widespread use of LLMs (Muzumdar et al., 2025). These hypotheses have been difficult to verify, primarily because there is limited understand- ing of how much internet content is actually AI-generated (Santy et al., 2025; Spennemann, 2025). In this paper, we attempt to address these questions. We concern ourselves only with LLM- generated text,leaving other modalities for future work, and use LLM-generated and AI-generated interchangeably.

The authors have produced a less technical version of their research online HERE, where they have a shorter abstract of their findings:

The proliferation of AI-generated and AI-assisted text on the internet is feared to contribute to a degradation in semantic and stylistic diversity, factual accuracy, and other negative developments. We find that by mid-2025, roughly 35% of newly published websites were classified as AI-generated or AI-assisted, up from zero before ChatGPT's launch in late 2022. We also find evidence suggesting that increases in AI-generated text on the internet bring about a decrease in semantic diversity and an increase in positive sentiment. We do not, however, find statistically significant evidence supporting the hypothesis that an increased rate of AI-generated text on the internet decreases factual accuracy or stylistic diversity. Notably, our findings diverge from public perception of AI's impact on the internet.

Here's a statement of their methodology:

Answering this question is harder than it might seem. Constructing a statistically representative sample of the internet is difficult, as there is no central index, popular domains are vastly over-represented in most crawls, and archival coverage has shifted considerably over time. To work around this, we draw on the Internet Archive's Wayback Machine and apply a multi-dimensional stratified sampling approach, approximating a uniform random draw from publicly accessible web pages published between 2022 and 2025 (see Section 3.1 in our paper).

On top of this sample, we need a reliable way to tell AI-generated and AI-assisted text apart from human-written text. AI-generated text detection is itself an open problem, so rather than committing to a single detector, we experiment with four prominent methods selected based on their performance on the RAID benchmark: Binoculars, Desklib, DivEye, and Pangram v3. We then run our own robustness checks across text length, HTML versus plain text, model family, model version, and language, and choose the detector that comes out the strongest overall — Pangram v3 (see Appendix A in our paper).

AI-Generated Text on the Internet from Mid-2022 to Mid-2025. The proportion of websites classified as fully AI-generated (red) and AI-generated or AI-assisted (purple) based on Pangram v3 detection applied to representative samples obtained from the Internet Archive. The dashed line marks ChatGPT's public launch in November 2022.

H/t Tyler Cowen.

Wednesday, May 7, 2025

Grrl Power on the Web

Jessica Testa, Is This the ‘Manosphere’ for Women? NYTimes, May 7, 2025.

Men dominate the top of the podcasting charts. As listeners, they slightly outnumber women, too. “Brocasters” and the “manosphere” have even become a media obsession, and for good reason. During the election, conservatives successfully tapped into a world of dude-driven content to reach disengaged voters.

But there has been a surge of podcasts made by women, for women, too. And a company called Dear Media is at the center of much of it.

Based in Austin, Texas, Dear Media operates the largest network of podcasts for women. Its nearly 100 shows are as freewheeling and chummy as those in the “manosphere,” similarly hosted by comedians and content creators. Except here, Joe Rogan’s alpha masculinity and Logan Paul’s unabashed idiocy are swapped for girlboss confidence and therapy speak.

Gone, too, is the overt conservatism that now blankets the manosphere — but not all of its ideas. Dear Media emphasizes health and wellness in its programming, at times dipping into the same kind of contrarian thinking that powers Make America Healthy Again, the agenda of Health Secretary Robert F. Kennedy Jr.

Wellness:

If feminist news was the nucleus of “lady blogs” a decade ago, wellness takes its place today. Edison Research recently identified the two topics most interesting to female podcast listeners: self-care and mental health.

In an interview, Michael Bosstick, the chief executive of Dear Media, described his editorial sensibility as “nonjudgmental” and oriented around free speech. He has described himself as a political independent, with a “Don’t Tread on Me” tattoo, and publicly supports Mr. Kennedy, once writing on X that Covid-19 vaccines were “experimental shots that caused many harm.”

Politics:

Mr. Bosstick is firm that he has no interest in policing or censoring his hosts. On his own show, he said, whether he agrees or disagrees with a guest, “I want to listen openly, and I try to kind of culturally push that to the rest of the company.”

His thinking aligns with that of the manosphere: Cancel culture is bad, unchecked conversation is good. People are too “hypersensitive,” Mr. Bosstick said.

During the election, that attitude attracted President Trump’s campaign to new media, as he racked up appearances with podcasters including Theo Von, Lex Fridman and Andrew Schulz. In January, two prominent Trumpworld women were similarly drawn to Dear Media.

A week before the inauguration, the Bossticks released a rare interview with the first daughter Ivanka Trump, followed by one with Cheryl Hines, the actress and wife of Mr. Kennedy, three days later. (Last summer, the couple also interviewed Calley Means, a top adviser to Mr. Kennedy.)

Ms. Trump hardly talked politics, which she called a “very dark, negative business.” Over nearly two hours, she talked more about her life, burning French toast and practicing jujitsu.

Money:

While declining to provide specific figures, Mr. Bosstick said Dear Media’s revenue fell between $51 million and $100 million, largely through advertising. But he does not see podcasts as the primary business. The goal is to “help creators monetize” through various channels: events, merchandise, television, publishing. Video podcasts may be booming, but they are just “one sliver of media,” Mr. Bosstick said.

The company, which has raised about $12 million in investments, has in turn invested in about 10 consumer brands attached to its talent roster, such as Spritz Society canned cocktails and Arrae supplements.

There's more at the link.

Tuesday, May 28, 2024

The Metaverse won't work because it will be too slow for interpersonal synch

It has to do with timing lags over the internet. The auditory system is very sensitive. Adam Neely starts explaining at about 5:18, where he's talking about the difficulty of jamming with other musicians over the internet. It turns out that the speed of light is a limiting factor. Thus, he explains (5:33) that it takes almost 20 ms to travel from New York to Paris, which is ten times the 2 ms audio threshold for simultaneity. Thus musicians in New York and Paris cannot jam with one another over the internet. To do that, the musicians have to be physically close to one another to cut down the latency. He gets to the Metaverse at 6:33.

Saturday, July 15, 2023

Who owns the gates to culture and knowledge?

Julia Angwin, The Gatekeepers of Knowledge Don’t Want Us to See What They Know, NYTimes, July 14, 2023. From the article:

We are living through an information revolution. The traditional gatekeepers of knowledge — librarians, journalists and government officials — have largely been replaced by technological gatekeepers — search engines, artificial intelligence chatbots and social media feeds.

Whatever their flaws, the old gatekeepers were, at least on paper, beholden to the public. The new gatekeepers are fundamentally beholden only to profit and to their shareholders.

That is about to change, thanks to a bold experiment by the European Union.

With key provisions going into effect on Aug. 25, an ambitious package of E.U. rules, the Digital Services Act and Digital Markets Act, is the most extensive effort toward checking the power of Big Tech (beyond the outright bans in places like China and India). For the first time, tech platforms will have to be responsive to the public in myriad ways, including giving users the right to appeal when their content is removed, providing a choice of algorithms and banning the microtargeting of children and of adults based upon sensitive data such as religion, ethnicity and sexual orientation. The reforms also require large tech platforms to audit their algorithms to determine how they affect democracy, human rights and the physical and mental health of minors and other users.

This will be the first time that companies will be required to identify and address the harms that their platforms enable. To hold them accountable, the law also requires large tech platforms like Facebook and Twitter to provide researchers with access to real-time data from their platforms. But there is a crucial element that has yet to be decided by the European Union: whether journalists will get access to any of that data.

Sunday, April 16, 2023

Competition between Google and Microsoft in search

I’ve got doubts about competition between Google and Microsoft on artificial intelligence. Setting aside my doubts about competition as a universal economic virtue – after all, one can win by cheating – I fear it will lead to the premature release of products, as it already has in the case of Google’s Bard. On the other hand, competition in the search space, that’s different. Google has had it too easy for too long.

The problem, of course, is that competition in search is intimately bound up with competition in AI. I fear we’re in for a bumpy ride.

The NYTimes has an article about competition in search: Google Devising Radical Search Changes to Beat Back A.I. Rivals, by Nico Grant, April 16, 2023. The article opens with Google in a panic over the possibility that “Samsung was considering replacing Google with Microsoft’s Bing as the default search engine on its devices.” Apple has a similar contract that ends this year.

A.I. competitors like the new Bing are quickly becoming the most serious threat to Google’s search business in 25 years, and in response, Google is racing to build an all-new search engine powered by the technology. It is also upgrading the existing one with A.I. features, according to internal documents reviewed by The Times.

The new features, under the project name Magi, are being created by designers, engineers and executives working in so-called sprint rooms to tweak and test the latest versions. The new search engine would offer users a far more personalized experience than the company’s current service, attempting to anticipate users’ needs.

Lara Levin, a Google spokeswoman, said in a statement that “not every brainstorm deck or product idea leads to a launch, but as we’ve said before, we’re excited about bringing new A.I.-powered features to search, and will share more details soon.”

Billions of people use Google’s search engine every day for everything from finding restaurants and directions to understanding a medical diagnosis, and that simple white page with the company logo and an empty bar in the middle is one of the most widely used web pages in the world. Changes to it would have a significant impact on the lives of ordinary people, and until recently it was hard to imagine anything challenging it.

There’s much more in the article.

Monday, February 6, 2023

Howard Rheingold reflects on governance in 6 online communities he's been a part of

Howard Rheingold, Online governance: Six Case Studies, Feb 10, 2022.

How should online communities be governed? I’m not concerned here with Facebook and Reddit, which have special problems because of their size and volume of interaction. Despite all the attention that the gigantic social media platforms attract, the Web is full of smaller communities — communities of interest and practice, mutual aid communities, support communities, learning communities, activist communities, science communities, businesses and their customers. They meet through message boards, chat rooms, Slack, mailing lists, wikis. Eventually (or ideally before they launch), the issue of how the community should be governed arises — usually because of some transgression that stirs community ire. Today, a wealth of online community governance toolkits has emerged. I won’t attempt to be comprehensive here, but will mention my experiences with governance in the WELL, The River, Electric Minds, Brainstorms, Slashdot, and the Omidyar Network. At the end of this essay, I include links to resources and toolkits. Please let me know of any resources that ought to be added.

I've appended two of Howard's write-ups, the one about Brainstorms since I am a member of that community, and the one about the Omidyar Network since it collapsed in on itself and Howard has interesting observations to make about it.

About Brainstorms

Brainstorms is private in the sense that you have to be invited to join. But getting invited is not difficult nor are their any hoops to jump once invited. The privacy exists, well, for privacy, but also so that members can be dismissed for cause. There is a procedure for this, but I've never witnessed it.

I've been a member of Brainstorms since before 9/11. Just how much before, I can't really say, though it couldn't have been long since it didn't exist until 1998. I remember 9/11 because I spent a significant part of that day online at Brainstorms talking about it.

I spent quite a bit of online time there for a few years, participating actively in a number of threads. But then along came the blogosphere, where I joined a group blog called The Valve (now defunct), then set up New Savanna in 2010, then Facebook and Twitter. These days my participation at Brainstorms is limited to posting photographs a couple of times a week, checking in on the life-story of Hadar Aviram, who has a rich and complicated life, where I occasionally post a comment, and occasionally posting a comment to my own life-story thread. I suppose what I find most interesting about Hadar's thread is that she posts often and has a small community of people who comment. That community seems like her personal Greek chorus.

Here's what Howard says about Brainstorms:

After Electric Minds went away, I found that I missed having an online braintrust and community. Instead of going back to the WELL, I started Brainstorms. My friend and online pioneer, the late Lisa Kimball, introduced me to Caucus, a web-based forum. So I set up a number of conferences — a general commons, and conferences on mind, science, arts & culture, phun, etc. — and started a few conversation threads in each one. By this time I had learned that it is best to start out with a few threads — about this conference, ask questions, introductions, and what should we talk about — and slowly grow the collection of threads. Then I emailed the self-registration URL to a few dozen people I knew around the world. When I got up the next day, there was a blooming, buzzing community in formation. That was 1998. I thought of it as “Howard’s Bar & Grill.” Participants agree to not attack each other, and the proprietor reserves the right to throw out miscreants. The way I put it at the time was: “an online community where it is possible to throw out assholes.”

For the most part, Brainstorms — which still exists! — was what I meant it to be: adults having serious and fun conversations without flaming or trolling. At Brainstorms’ height in the years before Facebook, Tumblr, Twitter offered so many alternatives to a general-purpose message board, we had about 500 active members. We had face to face get-togethers of dozens in San Francisco, Amsterdam, Melbourne, and Memphis. By now, the approximately 50–70 of us who are left are a couple decades older. We maintained a mutual aid fund that gave or loaned hundreds and sometimes thousands to BSers in need. We probably dispersed tens of thousands of dollars over the years. Governance long ago ceased to be an issue…after an initial period of sporadic turbulence. That doesn’t mean that every gets along all the time. It does mean that we settled into norms regarding how to about disagreeing.

I thought a benevolent dictatorship would take a lot less of my time than consensus. I was wrong about that. I made the decisions, after polite backchannel attempts to settle disputes or educate rude newcomers, but that didn’t kill meta. There were always different sides. Besides public arguments, I was bombarded with backchannel email lobbying. And a couple of the people who I tossed out took it personally in scary ways — for a while I had security on speed-dial at Berkeley and Stanford because a local troll threatened to disrupt my classes. Eventually, after four or five years, I tired of being Dad. I took a break from Brainstorms for a few months, and appointed a committee of 5 old-time BSers to take care of conflicts. They adjudicated a couple disputes, and for the last 15 years or so, entire years have gone by without needing their decisions. Every couple years, one of the committee retires and names a replacement.

And Omidyar:

Omidyar Network Online: Why Effective Online Governance Requires a Decision-Making Boot-Sector

Omidyar Network is the name of the philanthropic organization created by Ebay founder Pierre Omidyar. It was also the name of an online community from the early 2000s in which Omidyar experimented with a governance schema that grew out of his experience with Ebay. Online auctions among strangers pose what is known as “the prisoner’s dilemma” — a social dilemma in which each party is forced to accept a less than optimal deal out of a lack of trust for the other party. Omidyar instituted seller ratings that enabled buyers to rate sellers on a scale of one to five. Sellers with higher ratings make more sales, and sellers with a sufficiently high rating become “power sellers.”

The online network that Pierre and his wife set up was a resource for non-profits. For no charge, members could set up forums, mailing lists, wikis for their group. In addition to the groups, there was a commons in which members could talk about community issues. Because Pierre and Pam Omidyar had stated that they intended to contribute their approximately $15 billion of ebay stock to worthy causes, the online network grew rapidly — about 4000 members in the first few months. Because the seller reputation system worked so well at ebay, Omidyar set up a more complex reputation system: Members could gain points by logging in, by commenting, by creating conversation threads; members could also gift points to others and to take points away from others. I can’t remember what extra privileges positive points granted, but I remember that if any member’s points dropped to a negative level (I think I recall that it was -50), their posts would no longer be visible in public, and people would have to go to the sanctioned member’s home page to see what they have to say.

I was interested in this governance experiment, so I joined the network shortly before the Omidyars announced their first experiment in networked philanthropy. They pledged to contribute an initial $25,000 to a group or groups chosen by the community. Thus began a massive meta thrash. All 4000+ members of the community were made aware of the community conversation about the grant. Probably around 1000 people checked out the conversation thread, around 200 participated in the conversation, and around 25 people contributed the majority of the posts, which ended up numbering in the thousands. Attempts at consensus continued to fail on minority objections. My contributions to the conversation thread consisted largely of unheeded warnings that without a clear decision-making procedure, this conversation was doomed to be an infinite meta rathole.

The group did agree that it would be necessary to vote, in the absence of consensus. But how would the franchise be defined? Could someone who joined yesterday have a vote? Would a minimum number of karma points be required? How would the vote be carried out? So another round of infinite meta revolved around how to decide how to decide. The network as it was originally set up did not specify a decision mechanism, nor did the Omidyars intervene by declaring mechanisms by fiat. Eventually, giving up on voting, the exhausted networkers divided the grant among three groups. The process appeared to drain rather than boost the community’s ability to explore the networked philanthropy the Omidyars had hoped to initiate.

And than the whole issue of community decision-making blew up when it turned out that one of the most vocal debaters in the grant-making thrash had created dozens of sock-puppet accounts and was awarding points to friends and subtracting points from those who disagreed with her. Not too long after that, the Omidyars shut down the online community. That’s when I became convinced that virtual communities must begin by defining minimum decision-making procedures. I recalled that when personal computers were still young, they had to be booted up with a floppy disk. The first information to be read from that disk was a “boot sector” that instructed the mechanism on how to load the operating system. My experience with governance thrashes in the WELL, the River, Electric Minds, Brainstorms, and the Omidyar Network led me to believe that it is nearly impossible for a group to formulate a decision-making procedure from scratch — they need a boot sector with a minimum description of who can vote, and how the vote should be carried out. With that, the community can change the rules if they want — but at least they won’t have to argue about how to decide how to decide.

Online Governance Resources

Howard begins this section with a link to a paper by Amy X. Zhang, Grant Hugh, and Michael Bernstein, PolicyKit: Building Governance in Online Communities. Here's the abstract:

The software behind online community platforms encodes a governance model that represents a strikingly narrow set of governance possibilities focused on moderators and administrators. When online communities desire other forms of government, such as ones that take many members’ opinions into account or that distribute power in non-trivial ways, communities must resort to laborious manual effort. In this paper, we present PolicyKit, a software infrastructure that empowers online community members to concisely author a wide range of governance procedures and automatically carry out those procedures on their home platforms. We draw on political science theory to encode community governance into policies, or short imperative functions that specify a procedure for determining whether a user-initiated action can execute. Actions that can be governed by policies encompass everyday activities such as posting or moderating a message, but actions can also encompass changes to the policies themselves, enabling the evolution of governance over time. We demonstrate the expressivity of PolicyKit through implementations of governance models such as a random jury deliberation, a multi-stage caucus, a reputation system, and a promotion procedure inspired by Wikipedia’s Request for Adminship (RfA) process.

Wednesday, December 22, 2021

Did "The Matrix" depict a more optimistic vision of the future than we are now living?

Samuel Earle, "The Timeline We’re on Is Even Darker Than ‘The Matrix’ Envisioned," NYTimes, December 22, 2021.

In original Matrix films had an undercurrent of optimism:

Yet alongside this dark tableau, “The Matrix” also contained dreams of a better internet than our own. The eponymous computer simulation is a sinister mechanism of control, imposed upon humans to harness their energy. But after the simulation is seen as a construction (enabled by swallowing the “red pill”), people have the power to plug back in and traverse it as a truer version of themselves.

Such possibilities seem beyond the current web:

Despite the pseudonyms, trolls and alter egos that still dwell in some corners of the internet, its main byways now prize consistency and transparency over the risks of anonymity and reinvention. The idea of the internet as a place to cultivate an identity outside the slots other people put you in has been eclipsed by a social media-driven focus on creating an aspirational personal brand. Self-realization is now measured in likes, shares and follower counts.

“Our digital presentations are slicker, influencer-influenced,” Ms. Turkle, a professor of the social studies of science and technology at MIT, told me. “Everyone wants to present themselves in their best light, but now we have a corporate filter of what ‘pleases.’”

The cultural shift toward holding one narrowly defined identity — online and offline, across platforms — aligns neatly with Silicon Valley’s interests. The aim of many tech companies is to know us more intimately than we know ourselves, to predict our desires and anxieties — all the better to sell us stuff. The presumption that we each hold a single “authentic” identity simplifies the task, suggesting to advertisers that we are consistent, predictable consumers.

Sunday, August 8, 2021

Is Academia Eating Its Young?

Bumped to the top of the queue from a decade ago on general principle. Nothing's changed on this except that other social media have displaced blogging. The pandemic has killed off some of the weaker schools and hastened the transition to forms of distant learning. Permatemping is still a scandal.

Saturn devouring young 400

The fate of bloggers has been a matter of some interest in the academic blogosphere. Not bloggers in general, but graduate students and junior faculty. The concern is not just that no professional credit accrues for blogging, but that blogging may actually hurt one’s career. It is a fact that many (older) faculty are, at best, ignorant of and indifferent to, online activity of all sorts, including blogging. Some – many? – are actively hostile to it.

This issue came to my attention most recently though a discussion at Arcade, a humanities site sponsored by Stanford. Arcade includes a group blog where a wide range of humanists have blogging rights. This particular issue was not the central focus of the discussion – which was about the anxieties attendant upon thinking in public and, by the way, why aren’t people posting as much anymore? – but it did come up.

In thinking about the issue I was struck by one thought: it’s pointless to try legitimating blogging in the current academic world. Of course I think blogging is a valid intellectual pursuit,; that’s not an issue. The current academic world is too far gone to be able to accommodate blogging in a robust way. Just why I think that would require some explaining, which I don’t want to at the moment. Let’s just say that I’ve been deeply skeptical about academic institutions for some time, and this is just another reason to despair of them.

And that leads me to this question: Is the academic world eating it’s young? And if so, how can it survive?

I believe the answer to the first question is “Yes.” And the answer to the second question must therefore be, “No, it cannot survive. Not in its present form.”

On the first point, it’s long been common wisdom that graduate students and junior faculty must curb their creativity and get with the program so they can obtain, first, a tenure-track appointment, and then tenure itself. But, if you’ve been holding back during what may be the most creative period in your intellectual life, will you be able to open it up once you’ve gotten tenure, or will you have forgotten how? And, it’s not as though having tenure brings you home free. There are still many lines to toe, many hoops to jump. I take the case of Mark Changizi as an index of the current state of affairs. He’s one of the most brilliant and creative psychologists of his generation, but has chosen to abandon the academic ship because it didn’t give him room to think deep thoughts.

Then there is the scandal of permatemping, using graduate students and poorly paid adjuncts to teach the bulk of undergraduate courses. Not only does this make research impossible – and research has never been of primary interest to more than a small number of academics – but it makes good teaching difficult to impossible. How can you do a good job if you’re teaching five courses at a time, spread over two campuses? You can’t. All you can do is hang on and hope your spouse has good health coverage.

No, the current system cannot survive, not when the structure of the institution forces older faculty to grind the younger faculty into hamburger. At the same time there’s the pressure from the on-line for-profit world. Those institutions have their own problems, but they’re here to stay. They’ll continue to siphon students away from more traditional institutions, and those traditional institutions will offer more and more online instruction, taught by graduate students and adjuncts.

Many schools will survive, in some form. But the system as it currently exists is dying. It cannot be saved. The best that can be hoped for is that some gifted people at some institutions will be able take hold of change and do something interesting. I see precious little of that on the horizon.

Tuesday, August 3, 2021

Ramble at the beginning of August 2021: We’re lost, looking for meaning, alas inequality [the Tractatus]

The inequality that sends billionaires into space is not an independent and isolated feature of the contemporary world. But let’s not start there. Let’s start with something ‘easy,’ Wittgenstein’s Tractatus, then we can think about the meaning of life, and then we can take a look at inequality – kith and kin to those billionaires. I end with a somewhat whimsical suggestion for a Twitter tax for the rich.

This is something of a grab bag and a parking lot. I’ve been thinking of these things and put them here so I know where they are. In time I’ll attempt to make sense of them. Maybe.

Wittgenstein’s Tractatus

I recently read a post by Rohit (The Architecture of Knowledge) that reminded me of Wittgenstein’s Tractatus Logico-Philosophicus. I asked him about it and he replied that, yes, he was familiar with it and that it was “an inspiration forever in its scope and ambition.”

I understand where he’s coming from. I read it early in my undergraduate years at Johns Hopkins and it had a strong effect on me. Yes, for its logical structure, but also for its last proposition: “7. Whereof one cannot speak, thereof one must be silent.” I’d almost say its appeal was evenly split by that proposition and by the rest of it.

Looking back, it is a peculiar work. I went on to study computational semantics in my graduate years at SUNY Buffalo where I studied computational semantics with David Hays (while working on a degree in English literature). There I certainly was interested in something one might as well call “the architecture of knowledge,” that is, the structure of human knowledge. For the most part we – members of Hays’s research group – were interested in general principle, on the one hand, and small scale structures on the other. Yet my first major piece of work, an analysis of Shakespeare’s Sonnet 129, certainly implied large scale structure, as did my dissertation, “Cognitive Science and Literary Theory” (1978). That work certainly looks very different from Wittgenstein’s Tractatus.

Wittgenstein was much taken with symbolic logic, which was relatively new at the time. He took its propositions to represent, that is, to be capable of representing, facts about the world. The world itself? Is that what he was chasing? Or did he think of those propositions as representing what Noam Chomsky came to call mentalese, the conceptual language of the human mind? Perhaps he did, but if so, he later came to rather different views on such subjects. In any event it certainly doesn’t look the work I did with David Hays or, for that, matter any of the roughly similar work others were doing at the time (see, e.g., John Sowa’s comprehensive website on knowledge representation).

Of course representing how we think about the world is very different from representing the world as it really is. That’s something fraught with metaphysical difficulties. And yet that’s something I’ve given some attention to in recent years, I’m thinking particularly about my work on pluralism, but also my more recent work, What economic growth and statistical semantics tell us about the structure of the world. This work, however, is quite different from, has a very different texture than, the earlier work on conceptual structures.

That there is a clear difference between these two bodies of work suggests that we’ve learned something in the decades that have passed since Wittgenstein wrote his Tractatus. That work, it seems to me in retrospect, is neither about the world nor about our representations of the world. Or it is indifferently about both. It is an undifferentiated metaphysical ether.

To live a meaningful life

Meanwhile, our visions of the future are vapid, a theme I’ve been exploring in a recent series of posts on billionaires-in-space and in a post on our visions of the future. I think this is linked to various posts I’ve made about living a meaningful life:

Our visions of the future are not coupled with, do not emerge from, a belief in living a meaningful life. Absent such a belief, billionaires going into space collapses into self-regarding joy-riding. These seem to be people unconnected with the world, holding themselves above and outside the world. They are hollow men.

But how does the human mind, the human spirit, find itself at home in the world? For that’s what’s at stake. In a post about Mark Moffett’s The Human Swarm I made some observations about identity that are relevant to the question of meaning:

The argument that needs to be made is that our nervous system affords us open-ended awareness of the world. I suspect that’s a joint product of the active nature of the nervous system and the emergence of language. On that active nature, the nervous system doesn’t passively take the world in, but rather actively probes the world through continuously projecting expectations – think, for example, of the model William Powers developed almost a half century ago in Behavior: The Control of Perception (1973). Thus perception is a process of verifying those projections (or, to use a more current language, updating Baysian priors).

The emergence of language leads to an endless curiosity about everything: What’s that? How does it work? Where’d it come from? Living becomes thus becomes a dialog with the world. And the question, Where did WE come from? will arise in that process. The answer initially takes the form of myth, of stories about origins. And those stories, in effect, establish the link between a society and world. That too is a matter of identity.

Those same myths and stories direct our search for meaning in life. Where do stories of billionaires joy-riding in space direct that search for meaning?

Inequality, executive pay, and stagnation

Back in 2019 Tyler Cowen published a short article in Time, Why CEOs Actually Deserve Their Gazillion-Dollar Salaries, which was excerpted from his recent book, Big Business: A Love Letter to an American Anti-Hero. After acknowledging that that top CEOs make 300 times as much as the average worker, Cowen argues:

While individual cases of overpayment definitely exist, in general, the determinants of CEO pay are not so mysterious and not so mired in corruption. In fact, overall CEO compensation for the top companies rises pretty much in lockstep with the value of those companies on the stock market.

The best model for understanding the growth of CEO pay, though, is that of limited CEO talent in a world where business opportunities for the top firms are growing rapidly. The efforts of America’s highest-earning 1% have been one of the more dynamic elements of the global economy. It’s not popular to say, but one reason their pay has gone up so much is that CEOs really have upped their game relative to many other workers in the U.S. economy.

Today’s CEO, at least for major American firms, must have many more skills than simply being able to “run the company.” CEOs must have a good sense of financial markets and maybe even how the company should trade in them. They also need better public relations skills than their predecessors, as the costs of even a minor slipup can be significant.

And so forth and so on, “yada yada,” to quote various characters from Seinfeld. I’m willing to grant that these highly compensated executives get their jobs through honest labor, rather than some form of corruption, and that, on the whole, they are more competent than the next lower tier of executive talent.

What I question is that ratio between CEO pay and the pay of the average worker, 300-to-1. Is that necessary? Wouldn’t a 30-to-1 compensation ratio leave plenty of room for these (mostly) guys to play “mine is bigger than yours”? Where did that ratio come from?

We know roughly when it arose, after the 1960s. In August of 2020 the Economic Policy Institute reported:

In 2019, a CEO at one of the top 350 firms in the U.S. was paid $21.3 million on average (using a “realized” measure of CEO pay that counts stock awards when vested and stock options when cashed in rather than when granted). This 14% increase from 2018 occurred because of rapid growth in vested stock awards and exercised stock options tied to stock market growth. Using a different “granted” measure of CEO pay, average top CEO compensation was $14.5 million in 2019. In 2019, the ratio of CEO-to-typical-worker compensation was 320-to-1 under the realized measure of CEO pay; that is up from 293-to-1 in 2018 and a big increase from 21-to-1 in 1965 and 61-to-1 in 1989. CEOs are even making a lot more—about six times as much—as other very high earners (wage earners in the top 0.1%). From 1978 to 2019, CEO pay based on realized compensation grew by 1,167%, far outstripping S&P stock market growth (741%) and top 0.1% earnings growth (which was 337% between 1978 and 2018, the latest data year available). In contrast, compensation of the typical worker grew by just 13.7% from 1978 to 2019.

Cowen has also written about something he calls The Great Stagnation (2011), when America’s productivity started slumping. As I recall he sees stagnation as beginning in the 1970s. So CEO pay starting galloping upward at roughly the same time economic productivity started slowing down. Is there a connection there?

I’m not suggesting that one caused the other, though I must admit the correlation is tempting. However, the CEO compensation system is not a system that is isolated from and thus independent of the overall economy. They are two aspects of the same economic system.

Saturday, July 31, 2021

Is Facebook’s Metaverse announcement a distraction from PR problems in its core social media domain?

Enquiring minds want to know.

No doubt you’ve heard the Facebook is going all-in on research toward the Metaverse. What is the Metaverse? It’s a supercalifragilisticexpialidocious high-tech whosiewhatsit where we can all work together in virtual harmony. It’ll be way cool. Super.

However, Tim de Chant at Ars Technica suggests “Facebook’s metaverse gambit is a distraction from its deep-seated problems,” July 28, 2021. It’s the timing:

It’s clear that Zuckerberg has been thinking about this metaverse idea for a while. But the timing of Facebook's announcement is interesting, to say the least. Facebook has “a history of doing these kinds of technical projects that look like they might be revolutionary at times when they’re being criticized for their lack of social responsibility,” Jen Goldbeck, a computer scientist and professor at the University of Maryland, told Ars. [...]

It's probably an overstatement to say that the metaverse news was released to serve as an intentional distraction from the company’s current problems. But the thought undoubtedly crossed someone’s mind at the company. There’s a “70 percent” chance that Facebook’s metaverse project is a “distraction from all the bad things that are going on,” Goldbeck said. “The last thing they want is more discussion of their algorithms and Q-Anon and extremist groups.”

As a distraction, the metaverse is almost too perfect. It’s a flashy news item that attempts to position Facebook and Zuckerberg as visionaries inventing the future. It’s also literally a way to escape reality. If Zuckerberg ever wanted a metaverse to escape to, where Facebook exerted influence over everything from physics to religion, that time is probably now.

Not meat and potatoes:

But in many ways, the metaverse project is a fundamental shift away from what made Facebook successful. First, Facebook is a company that tends to succeed when it iterates on existing ideas or solves genuine problems. Zuckerberg’s initial version of Facebook, called "Facemash," was a 2003 Harvard knockoff of another site called “Am I Hot or Not” that launched in 2000. Later Facebook iterations aped features from Friendster, MySpace, and others.

Even the company’s technological crown jewel, the News Feed, wasn’t created due to pie-in-the-sky dreaming about what could take social media to the next level. It was a solution to a real problem—as people added more friends on Facebook, they were getting inundated with updates. News Feed and its algorithms helped cut down on the chaos, prioritizing content that a user was more likely to engage with. Another Facebook hallmark—ads in the News Feed, the company’s current cash cow—was born out of its struggles to monetize users on mobile phones.

The metaverse also cuts against a key part of Facebook’s brand—allowing people to share their lives with friends and family. For many, the platform’s appeal comes from users' ability to like, comment on, and share content related to their real lives—not their virtual ones.

It feels a bit like what ethologists call vacuum activity. Wikipedia:

Vacuum activities (or vacuum behaviours) are innate, fixed action patterns (FAPs) of animal behaviour that are performed in the absence of a sign stimulus (releaser) that normally elicit them. This type of abnormal behaviour shows that a key stimulus is not always needed to produce an activity. Vacuum activities often take place when an animal is placed in captivity and is subjected to a lack of stimuli that would normally cause a FAP.

See also displacement activity:

Displacement activities occur when an animal experiences high motivation for two or more conflicting behaviours: the resulting displacement activity is usually unrelated to the competing motivations.

Alas, getting from those characterizations of Facebook’s behavior takes a bit more work than I want to undertake in a blog post. I leave that as an exercise for the reader.

Friday, June 11, 2021

Ross Douthat defends real experience [Andreessen’s been out in the sun too long]

Ross Douthat has an important column, Decadence and Andreessen's Dilemma (Reactions, June 10, 2021). The piece begins:

Early in the pandemic era, the Netscape founder Marc Andreessen published a manifesto that was so on-point for the themes of The Decadent Society that I cited it in the paperback. His argument was that many of the institutional debacles in the West’s response to the coronavirus reflected a longstanding failure to build, a comfort with stagnation and sclerosis visible everywhere from our physical environment to our regulatory apparatus to our transportation infrastructure and education system.

But what kind of building, Douthat wants to know? Building physical things in the real world, but also people and institutions, or cruising the internet to build in virtual reality?

This question divides anti-stagnationists: Figures like my AEI colleague James Pethokoukis or the bullish-on-America Bruno Maçaes tend to regard Silicon Valley and its virtual realities as integral parts of a more dynamic future, while people in the orbit of Peter Thiel are more likely to regard Big Internet in its current incarnation as an obstacle to real-world forms of industry and growth, and in need of some kind of political and social management to become a spur to dynamism instead. Personally I’m mostly on the Thielword side of the debate: Without going all the way to the Butlerian Jihad, I think human civilization needs to exert a kind of mastery over the internet and the virtual, a disciplining and putting-in-its-place, if we want to both preserve certain basic human goods and make non-virtual leaps forward more likely — because left to its own devices the internal techno-logic of the virtual realm will pull us deeper into decadence.

As important as the internet is to me – I suspect that it’s more important to me than it is to Douthat – I’m with him on this. He goes on to quote a recent interview in which Andreessen defends the coming primacy of virtual world:

A small percent of people live in a real-world environment that is rich, even overflowing, with glorious substance, beautiful settings, plentiful stimulation, and many fascinating people to talk to, and to work with, and to date. These are also *all* of the people who get to ask probing questions like yours. Everyone else, the vast majority of humanity, lacks Reality Privilege – their online world is, or will be, immeasurably richer and more fulfilling than most of the physical and social environment around them in the quote-unquote real world.

After giving Andreessen’s point due consideration, acknowledging for example that “I can find people who share my obscure interests in an instant” – Douthat argues:

But there’s a problem: There are a lot of experiences — call them “core experiences” — that people still prefer to have in the world of flesh and blood, ranging from the banal (real drinks >> Zoom drinks) to the transcendent (visiting Chartres Cathedral >> doing a virtual tour of Chartres Cathedral), from the excitement of travel or the thrill of concert-going to the physical-cum-emotional intensity of falling in love and having sex (not necessarily in that order). And there are core experiences that just don’t translate into virtuality at all: You can some kind of sexual intercourse on the internet, depending on how elastically defined, but you can’t bear and raise a child.

Now say what you will about 20th century America, but it did a pretty good job of democratizing these kinds of core experiences. Whereas while the age of the internet has clearly expanded the range of available experiences, for core experiences democratization sometimes seems like it’s in danger of going into reverse, with experiences in reality, from the museum to the ballpark to the classroom, becoming the high-cost “premium” experience, while a mediocre virtual equivalent is offered the masses and the proles.

Yes.

Why the internet is important to me

While I have a variety of interests and talents – I’m a good musician, and a good photographer, and have spent countless hours on both – I am primarily an intellectual, a thinker, explorer, and a writer. I got my Ph.D. in 1978 for a dissertation entitled, “Cognitive Science and Literary Theory,” years before anyone else in literary studies way paying attention to the cognitive sciences. I got a faculty position and The Rensselaer Polytechnic Institute in Troy, NY, and was turned down for tenure in, I believe, 1985. I spent two or three years trying to get another academic post and failed. Since then I’ve continued doing my intellectual work while doing a bit of this and that to make a living.

The internet afforded me the opportunity to build an intellectual life that would have been impossible without it. Not only has it given me access to books and articles, but it has given me access to people and avenues for publication that don’t require me to jump through the hoops of academic convention, though I do publish in the formal academic literature every now and then.

Back in 2006 I started blogging at an academic group-blog called The Valve. I posted articles and commented on articles by others. The Valve went bust in 2012 or so, by which time I’d created New Savanna. But I remain in touch with some of my colleagues from those days even though we’ve never met in person. I’ve set up pages at three sites for the distribution of academic papers, Academia.edu, SSRN, and ResearchGate. And then there’s Twitter. And more, much more.

But I’ve made my point. Without the internet I would be an isolated intellectual. The internet has given me the means of actively functioning in an international intellectual community.

But none of these experiences compare with a core experience I had in graduate school. While I was enrolled in the English Department at SUNY Buffalo, I got my deepest education in the Linguistics Department under the tutelage of David Hays. I was a member of his research group for four years.

We met weekly at his house, Twin Willows, on the shore of Lake Erie. At the beginning of the meeting each participant, student, faculty or, for that matter, visitor, was allowed to put one item on the discussion agenda. We then moved through the agenda in order around the table. If we didn’t finish the agenda, the remain items went to the head of the agenda for the next meeting.

Depending on when we held the meeting in the morning or afternoon, we’d prepare and eat a meal together. Hays would search the kitchen for comestibles, point out what was available, and we’d set to work fixing a meal. When it was over, we’d wash the dishes. The shared meal was essential to the group’s functioning.

We accomplished real intellectual work in those meetings, unlike what happened in the graduate seminars I took, which, despite the small class ‘discussion’ format, were primarily opportunities for faculty to expatiate on whatever. I can’t imagine that happening in a Zoom meeting, or some similar online format.

And I certainly cannot imagine the kind of discussions Hays and I subsequently had, after I’d graduated, taking place online. We continued to work together. He’d left Buffalo to live in New York City, so it was easy to travel there from Troy, New York (where I remained after I’d failed to get tenure). We’d talk and talk, often in his apartment. But we’d also go to Fort Tryon Park nearby and chat while walking the park. As you know, peripatetic discussion is an ancient and venerable intellectual tradition. It works.

Our most profound interaction, however, took place when we were all talked out but had not been able to resolve matters. Hays would take a seat in one place, me in another, and we’d each lean back and think. And think. One of us would hazard a remark or a question, we’d have a little discussion, and that it was back to our reveries. When things went well, which happened often enough, but not always, one of us would have an idea that the other could take up. We’d toss the ball back and forth, reach some conclusions, and Hays would sit down at the computer and write up some notes. Then we’d break out the scotch – four fingers worth in a tumbler – and drink while preparing the evening’s meal.

I can’t imagine doing such a thing in a virtual meeting, I don’t care how high the visual resolution, how good the sound, or how fast the internet connection, you just can’t get the highly personal and idiosyncratic interaction that is necessary.

And then there’s music

It seems that I have been a musician all my life. Many of my deepest and most profound experiences, both joyful and, to be honest, a bit terrifying, have happened either while making music or listening to live performances. I will only mention two occasions and then let my band mate Rick Rourke speak.

In the early 1970s I played in a rock bank called St. Matthew Passion. It was out last gig. During the introduction to “She’s Not There” I went blank. The music got more and more intense until Wham! I felt myself dissolve into white light and pure music. It felt good. I thought about that experience for days, weeks, months, and years afterward. I might even say that I’ve never fully ‘recovered’ from that experience. What could ‘recovery’ possibly mean? It changed me.

In 1987 David Hays and I went to hear Miles Davis perform at Avery Fisher Hall in New York’s Lincoln Center. The band was hot, the performance was stunning, and Miles and his band left the audience in grateful shock. “Time After Time” was the strongest song in a set of strong performances. The tempo was slow, very slow, and the dynamics were low, mostly, for there were times when the sound swelled to fill the hall. But the music was so powerful it filled the hall no matter how loud or soft the sound. Even at its softest, which was very soft, Miles’ sound was so intense that you’d think it could suck sound right out of the air. When Miles and the band had finished performing “Time After Time” the audience was completely and actively silent. It took us a few moments to return sufficiently to ourselves so that we could offer up the customary applause.

During the mid-1980s I was a member of The Out of Control Rhythm and Blues Band. We gigged in upstate New York, mostly in Troy, Albany, Schenectady, and Saratoga Springs. We played small clubs and not so small clubs, private parties, weddings, and major concerts (e.g. opening for B.B. King). Here’s a video that Steve Rosenbaum, then of Our Town Television, now Managing Director of NYC Media Lab.

That’s me on trumpet (c. 0:29). We’re playing at The Metro, in Saratoga Springs, one of our regular gigs. A bit later Rick Rourke, who’d been on the road a number of years as a full-time musician, sums it up (c3:19): “It’s the closest you can get to really being feeling totally happy with yourself.”

The closest you can get to really being feeling totally happy with yourself. That’s core experience.

If Andreessen doesn’t understand this, then he has no business talking and writing about the future of internet technology. He may be brilliant, he may be rich, but I fear that he is not wise. He doesn’t understand the limitations of his experience.

I hear he’s a tall man. Perhaps his brain has spent too much time too close to the sun and, like Icarus, he’s falling. Or perhaps he has spent too much time surrounded by people who defer to him because of his wealth and power. Whatever the problem, we cannot allow ourselves follow him and his fellow digerati down that rabbit hole to perdition.

Sunday, April 18, 2021

Who's connected to the internet, world-wide? [from 2014]

Thursday, October 15, 2020

Facebook or Freedom, Part 6: Ideas on the table [so far]

This is mostly a book-keeping post to list the various ideas I’ve gathered on the problem of social media, with Facebook as my prime example.

What I’m seeing at the institutional level is: 1) break them up, 2) third party oversight of some kind, not government, but not directly tied to the for-profit platforms either, and 3) independent not-for-profit digital “public parks.” What if we link 2 and 3 somehow? Or are they too different for such a linkage to make much sense? In any event, they might be the beginnings of a new institutional sphere.

Note: This post does not, however, collect all my Facebook or social media posts.

Institutions

Digital public park

We need to create the digital equivalent of public parks. These would be free to all and perhaps funded through philanthropy (think of Andrew Carnegie). They would require careful tending.

What we need online is the electronic equivalent of a public park [digital public infrastructure]

Break them up (monopoly)

Asking/forcing the major platforms to police themselves will ultimately prove unworkable. It is more properly a function for the state; and it would also lock-in large size and create a barrier to entry as only large platforms can come up with the overhead revenue to do this.

Facebook or freedom, Part 4: Notes on Doctorow, How to Destroy Surveillance Capitalism

House recommends anti-trust action against tech giants

Oversight Board

We now have two oversight boards, one set up by Facebook, and the other set up independently of Facebook. Could these develop into a new kind of institutional actor, perhaps in concert with one or more digital public parks?

There is a third party board created to critique Facebook. What authority does it have, would it have, over FB? None that I can see.

Facebook critics form their own oversight board [Facebook objects]

Facebook has also established an oversight board, and set it up as a legally separate entity with a multi-million dollar endowment. What real authority does they have, would it have, over FB? None that I can see. Just moral suasion.

Facebook Oversight Board

Facebook announces new 'values'

Phenomenology of interface use

This is based on my own experience and notes how I’ve reacted differently to the forced change in the Facebook interface and the forced change in the Blogger interface. Two factors: 1) I use them in somewhat different ways; this is about technique. 2) I interact with people more of Facebook than on Blogger.

Facebook or freedom, Part 5: A different platform changes its interface, Blogger

Facebook or freedom, Part 3: The game goes on [Media Notes, special edition #1]

Facebook or freedom, Part 1: Who gave you permission to mess with my mind?

Users need to take more responsibility for the interface, and they need tools to do so.

We need decentralized social networks and AI tools to give users more control over them

Other things

There are places in the world where Facebook is essential internet technology.

Facebook or freedom, Part 2: Interlude, If I were Facebook, I’d be thinking…you will comply

There are other uses to which Facebook’s data can be put.

Measuring culture worldwide with a sample of two billion humans using data from Facebook

Saturday, September 19, 2020

Facebook or freedom, Part 3: The game goes on [Media Notes, special edition #1]

It’s been 12 days since my last Facebook post, on September 7, and I’ve had at least four 48 hour reprieves since then, maybe 5 – I’ve not been keeping a strict count. What do I mean by a reprieve? I switch to a new page on FB and Wham! they hit me with the new interface without warning. But so far they’ve also given me the option of switching back, which I’ve done. When I do so, however, they always as why I’m doing it, is it because you don’t like something or want something else? No, I reply, it’s because I don’t like you arbitrarily messing with my mind.

I assume that one of these days I’m not going get that option. When, though, and why do they keep giving my 48 hours and then not following through? Do they want to keep on collecting my comment – I’ve only got one – on why I’m doing back? There’s got to be some logic here.

I’ve been watching this new docudrama on Netflix, The Social Dilemma, which I don’t much like (more on that later), which is about social media, including Facebook of course, and how the tech wizards behind the screen do everything to control us, including running little experiments where they make some change in what we see and note how we respond. That makes sense. From which it follows that my behavior is being closely monitored, not necessarily by a person – I can’t imagine that I’m important enough – but more likely by some program module.

Is that module primed for each individual user? That is, FB has been tracking our behavior on FB and across the web, right? So it’s built up some statistical profile of each person’s activity. I’d assume it’s making guesses about how we’re going to behave in various circumstances. We’ll either confirm the guess, thus confirming its (Bayesian) priors, or not, in which case it updates those priors. Either way the algorithm ‘learns’ something.

So, what kinds of guesses is FB making about my behavior with respect to this interface change? When people make a comment about why they want to keep the old interface, how do they deal with those comments? Are they looking for something in particular? I’d assume that very few – none? – get read by a human. What’s their AI-module checking for?

What’s their plan for me on the next time they switch me over? Are the going to give me the option of switching back? If so, what’s their prediction about whether or not I’ll accept the offer? If they predict that, yes, I’ll switch back, do they already know how many more times they’ll give me to stay with the old interface?

I don’t know. Just how sophisticated is their manipulation software? I don’t know that either. In a sense, neither do they.

* * * * *

And that brings me to The Social Dilemma. I’m about two-thirds of the way through, and it took me two, maybe three, sittings to get that far. Will I finish it? I don’t know, but maybe Facebook, Google, and of course Netflix know, or at least have their guesses. Maybe they’re even betting on my behavior. Now there’s a concept, have the different social networks bet on user behavior? How will the bets be paid off? Shares of stock? User information?

The Social Dilemma switches back an forth between scenes where various experts from industry and academia tell us about how social media – Facebook, Twitter, Google, Instagram, etc. – tracks and manipulates our behavior and dramatized scenes where we see people respond to the manipulations of a nefarious AI, personified by Vincent Kartheiser. I don’t feel that I’m learning anything new. Yes, yes, they’re doing all this stuff and yes, yes, they’re really good at it. But there’s no real detail, no substantial argumentation, just all this assertion sandwiched in between this cheesy dramatization, which is useless.

Coming into this program I KNEW there was a problem, tracking, manipulation, fragmentation and polarization in the civic sphere, and so forth. I was looking for something to give form and substance to all these things I already know in a loose and baggy was. The Social Dilemma isn’t doing that.

Three useful reviews:
The Oremus piece reminded me of Marcuse's notion of repressive desublimation, where an oppressive regime allows the expression of some dissent (desublimation) but not so much as to threaten in regime in any material way (repressive). It's a kind of tax the oppressors levy on the oppressed for the privilege of speaking truth to power, but nothing more than speech.

For example, and from Oremus' article:
But if The Social Dilemma largely succeeds in answering its opening question (“What’s the problem?”), there’s a second, crucial, stage-setting scene that the film seems to forget about as it goes on. It’s when the film’s central real-life figure, the “humane tech” advocate Tristan Harris, recounts how he grew disillusioned with his work as a young designer at Google, and eventually wrote an explosive internal presentation that rocked the company to its core — or so it seemed. The presentation argued that Google’s products had grown addictive and harmful, and that the company had a moral responsibility to address that. It was passed around frantically within the company’s ranks, quickly reached the CEO’s desk, and Harris’ colleagues inundated him with enthusiastic agreement. “And then… nothing,” Harris recalls. Having acknowledged ruefully that their work was hurting people, and promising to do better, Googlers simply went back to their work as cogs in the profit machine. What Harris thought had served as a “call to arms” was really just a call to likes and faves — workplace slacktivism.
But also, as an example of the film's narrow tech-bro-centric POV:
As the activist Evan Greer of Fight for the Future points out, the film almost entirely ignores social media’s power to connect marginalized young people, or to build social movements such as Black Lives Matter that challenge the status quo. This omission is important, not because it leaves the documentary “one-sided,” as some have complained, but because understanding social media’s upsides is critical to the project of addressing its failures. (I wrote in an earlier newsletter that the BLM protests should remind us why social media is worth fixing.)
* * * * *

Meanwhile, why not let people have the interface they want? In fact, why not offer several interfaces? After all, Facebook is a huge company; they could afford to do this, no?

All Facebook needs is to be able to pass ads to users, makes suggestions, and track usage. Does everyone have to have the same interface to make this happen?

But then there’s that general principle: We’ve got you under our thumb. That makes sense. I don’t like it at all – which is why I’m putting up (ultimately futile) resistance – but I understand it.

Somewhere out on the web there’s got to be images of Zuckerberg tricked out in Borg cyber gear.

Monday, September 7, 2020

Facebook or freedom, Part 2: Interlude, If I were Facebook, I’d be thinking…you will comply

Siva Vaidhyanathan, Don’t Delete Facebook. Do Something About It. NYtimes, March 24, 2018.

But even if tens of thousands of Americans quit Facebook tomorrow, the company would barely feel it. Facebook has more than 2.1 billion users worldwide. Its growth has plateaued in the United States, but the service is gaining millions of new users outside North America every week. Like most global companies, Facebook focuses its attention on markets like India, Egypt, Indonesia, the Philippines, Brazil and Mexico. At current rates of growth, it could reach three billion users by 2020.

People in those countries are getting value out of Facebook; in some places, it’s one of the few reliable ways to keep in touch. In much of the developing world, Facebook is also the only news source that matters. This should horrify us. But it’s not a problem that will be solved by indignant Americans leaving the service. [...]

Hope lies, instead, with our power as citizens. We must demand that legislators and regulators get tougher. They should go after Facebook on antitrust grounds. Facebook is by far the dominant social platform in the United States, with 68 percent of American adults using it, according to the Pew Research Center. That means Facebook can gobble up potential competitors, as it already has with Instagram, and crowd out upstarts in fields such as artificial intelligence and virtual reality.





I certainly don’t know how Facebook thinks about these things – I’m thinking particularly about the current face-lift – but surely they must put their business front and center. And their business is delivering eyeballs and tracking data to advertising. That means they’ve got to keep as many users on the platform as possible.

Thus, to the extent that hate speech, bat-shit crazy conspiracy theories, political scamming, and so forth, drive up audience numbers without, at the same time, driving (too many) others away, that’s all good for business. Some employees and executives may know that this kind of stuff is not healthy for society, but, hey! business is business, amirite?

In THAT context, what about unilaterally changing the user interface? It’s tricky. The number of people who are really angry about it, as I am, is probably rather small. But they’ve been through this before and they know that some people won’t like it. So they give you the option of temporarily going back to the old interface, just to show you that they too feel your pain. And of course it is easy for them to frame the change as something they are doing for you, for us, the users. They’re making Facebook a better place.

The happier we are, the more eyeballs and tracking data they can deliver to advertisers.

The idea is to demonstrate that they own and can control this space, this commons, and that you will comply because, what choice do you have?

* * * * *

See  Facebook or freedom, Part 1: Who gave you permission to mess with my mind?

Sunday, September 6, 2020

Rambling into Fall: Facebook & the interface, brain-to-brain communication, AI and rank 5 culture

A lot’s happened since my last ramble (August 22). I’ve changed my attitude toward AI; Elon Music and Christof Koch have triggered one of my hobby horses, direct brain-to-brain thought transfer; and Facebook has decided to change its user interface. It’s start with the last and work through them in reverse order.

Facebook! Don’t mess with my mind!

Facebook has announced that it’s changing its interface. It’s done that to some people already. Others, like me, have been warned. But our respite is only temporary.

I’ve objected to this on two grounds: 1) it’s inconvenient, and 2) general principle: it’s my mind. The first is real but not, at this point, all that consequential. I neither like nor dislike the current interface; it is simply what I’m used to. Changing it will force me to think a bit where now I can act intuitively and automatically. While I don’t know how bad that will be – I do quite a bit on Facebook – I don’t anticipate major problems. Though, who knows, maybe they’ll cripple a feature I like. Twitter did that with their last interface change. I like the Moments feature, where I could group tweets under a single topic. Contrary to what it says in the help notes, it is no longer possible to add new tweets to existing moments. Has Facebook done something similar? I don’t know.

However, I think the matter of principle is deeper and thus more important. But I’m not sure how to argue it. What I’ve said is Facebook is messing with my mind, and I’ve justified that by reference to the idea that all sorts of media, books, sound recordings, movies, etc., are extensions of our minds. I’m wondering if that’s quite right?

What I’m thinking at the moment is that we think of minds as essentially private. You can’t read my mind and I can’t read yours. And, yes, that is true. But aren’t minds fundamentally social as well?

Wittgenstein famously argued that there is no such thing as a private language. Hence language is inherently social. Inner speech is a form of language and hence is, in some way, social and not merely private. That needs clarification and development, but I’m thinking that maybe that’s the way to go.

Social media are also private media, in a sense. I know Facebook has a bunch of settings you can use to control who sees what. I’ve looked at them once or twice in very specific situations, but for the most part I’ve not paid any attention to them because I don’t use Facebook for more or less private purposes. I’m a thinker and a writer and I’m happy to do both in public. That is, it’s my mind, but I’m using it in public and in a public way.

This line of thought goes in two directions: 1) the institutional and civic nature of social media, and 2) the nature of the technology. On the first, it’s not at all clear to me that our current institutional structure is adequate for using and governing this technology. On the second, I think the user needs to be in control of the interface (beyond the blither of settings FB makes available) in a way that requires perhaps a substantial rethinking of the nature of the technology. I suspect that AI will play a role in the future realignment.

Brain-to-brain thought transfer

The idea’s been around for awhile, along with the idea of direct brain-to-machine linkage. I don’t know when it first showed up in science fiction, or how wide spread it is, but I first became aware of it as a scientific and technical proposal in the work of Rodolfo Llinás in the mid-2000s. More recently Christof Koch and, most visibly, Elon Musk have joined the parade.

I don’t get it. There is an obvious technical objection which, as far as I can tell, none of these thinkers have addressed: how does a neuron tell whether an impulse is coming from some other neuron in the same brain or some other brain? If neurons can’t tell the difference, how can there possibly be thought transfer? I’ve offered a more elaborate and explicit argument, but that should be enough. Yet it hasn’t even been considered?

I don't know. I note that no one is suggesting it as something that will be done in the near-term or even mid-term future – though come to think of it, yikes!, Musk told Joe Rogan five to ten years – rather, it’s kinda’ far in the future where who knows what will happen. That is, it’s in a realm where we can’t really think about things rigorously. So, yeah! let’s throw caution to the winds.

Anyhow, it’s right up there with other fantasies of tech-bro religion, the super-intelligent computer (malevolent or benevolent) and uploading the contents of one’s brain to the cloud and thus becoming immortal.

AI and the future of technology

Finally, just the other day I started talking about the evolution of rank 5 culture and linking that, however loosely, with AI. Rather, it’s something a bit beyond AI. For AI is just a bunch of techniques. I’m looking for some one thing, comparable to the role of writing in the emergence of rank 2 or Arabic notation in the emergence of rank 3, etc. Whatever this new thing is, it is as likely to emerge out AI as anywhere else.

To get there AI has to break free of its chess-centric ways and develop a bit of curiosity about the basis of current statistical success, which I’ve been pointing toward in my recent thinking about GPT-3: GPT-3: Waterloo or Rubicon? Here be Dragons, and What economic growth and statistical semantics tell us about the structure of the world. As I said before, don’t believe the hype, but there’s a breakthrough there. We just have to find it.

More later.