Making a (large) epoxy table

While I've been watching Cam's videos for a few years. While I have no particular interest in these tables, I enjoy these videos. So I've watched a bunch of them. AND, this is about making something with your hands.

Timestamps:

00:00 Intro
00:08 Awkward Delivery
01:32 What I'm Building
02:32 Recent Customer Request
03:42 Oversized Machinery
05:16 Epoxy Best Practice
06:38 Don't Be Like Joe Pa
08:28 Emmy Winning Blacktail?
10:51 My One Contribution
12:57 Suspect Measuring
14:55 Workbench Modification
15:31 Epoxy Mix and Pour
18:45 Terrible Idea I Had
20:43 CNC and Potential Problems
22:08 Circle Cutting
25:25 Redwood OF
27:37 How to Clean Epoxy Buckets
29:53 UV Finishing
31:20 Reveal

Time for another ramble: Melancholy, Claude, Bloom, Claude, Ring composition, and Other Stuff

It’s been a while since I’ve done one of these; May 29th was the last one. If you look over there to right at the Blog Archive you’ll see I’ve been on a posting slump, with 3-figure monthly totals from January through June, then a dip to 61 for July, August: 18, September: 15, October: 30, and now 33 for November as I write this, and the month isn’t over. Maybe I’m pulling out of the slump.

Anyhow, I’m feeling a little backed up with things to post about, so it’s time to ramble on and see what’s up.

Melancholy, Mind (Mine), and Growth

That’s the tentative title for my next 3 Quarks Daily article. Starting back in November 2017 I’ve been making occasional posts about my monthly posting habits, which tend to drop during the winter. I’m thinking of using that as the point of departure for my next 3QD piece, which will go up on December 2nd.

During those down times I’m depressed to one degree or another (melancholy). But why? Since those down times have been in the winter, perhaps its seasonal affective disorder (SAD). But that doesn’t square with all of the evidence. There was no down-time in the winder of 2022-2023 and 2023-2024, but there was a slump in the summer of 2023. Something else is going on, and I think it has to do with creativity. To that end I want to discuss the ridiculous blither of tags here, 665 by November or 2023.

Claude

I’ve starting working with Claude, Anthropic’s chatbot. I want to do some posts where I verify some of the work I’ve done with ChatGPT. I’m thinking of posts on stories, ontological structure, abstract definition, and the Girardian analysis of Jaws. I can then gather those into a working paper.

I also want to look at other things. At the moment I’m thinking of seeing how Claude summarizes longish documents. I’m thinking of the Hamlet chapter from Bloom’s Shakespeare book and Heart of Darkness.

Harold Bloom and GOAT literary critics

A year ago I began a series of posts on the theme of the greatest literary critics. I got bogged down in discussing Harold Bloom. It’s time to finish it off.

Bloom may well be as brilliant a literary critic as we’ve had in the last 50 or 60 years. But brilliance is one thing, greatness is another. Brilliance is a function of the individual, while greatness is a function of the relationship between an individual’s work and the arena in which they’re working.

I’m not sure about Bloom’s fit. While he’s got a wide readership, it’s not clear to me that scholars have taken up his work in any significant way. They may cite him – perhaps especially is concept of influence – but that they don’t much use of his ideas in his work. But we’ve also got to consider his work in the larger public arena, where he is hands-down the most prominent literary critic. I’m not sure of how to handle that.

However, if History wants to declare that Harold Bloom is one of the great all-time literary critics, maybe even the GOAT, what do I care? What would really bother me is if future critics should decide to take his work as a model and (attempt to) do more like it. Like most literary critics he’s neglected the study of form and he’s been deaf to the cognitive sciences. There’s little in his work that’s worth amplifying. It’s a dead end.

ChatGPT report

About a year or so ago I started writing a report summarizing my work on ChatGPT. I need to finish that report. I’d estimate the three-fourths or more are done. I’d like to be able to include some work with Claude. I don’t intend a lot on this, just enough to say that I’ve verified some things.

I’d like to finish this by the end of this year.

Why’s ring-form composition important?

That’s tricky. It has to do with the fact that literary works are extended in time, unlike the visual arts, which are static in time but extended in space. You can’t take the whole thing in at a glance like you can a painting.

There are constraints on how a literary work can unfold in time. There is a sense in which (the nature of) the end is inherent in the beginning. Ring-compositions are even more tightly constrained. Dylan Thomas consciously and deliberately plotted the ring-composition of the rhyme scheme in his “Author’s Prologue.” Rhyme is not about meaning; its patterns are arbitrary with respect to meaning. But Coleridge did not consciously work out the ring-compositions in “Kubla Khan,” not Conrad in Heart of Darkness. These patterns ARE NOT arbitrary with respect to meaning. On the contrary, they are central to how meaning is constituted.

Other Stuff

More Cobra Kai: Follow-up on my post where I explore the Freudian angle, saying a bit more about Girard, and extending that into history.

* *

More on meaning in LLMs: I’ve suggested that Ilya Sutskyver conflates mistakenly (and unknowingly?) conflates cognitive and semantic structure with the structure of the world and so suggests that robust next-token prediction requires knowledge of the world. In this post I argue that making that distinction is, in fact, difficult, and involves what I’ve been calling the word illusion. I first confronted the problem as an undergraduate when I was trying to understanding the difference between the signified, and mental structure, and the reference, something in the world, of a sign. I may not have gotten deep intuitions about that until I began studying cognitive networks in graduate school.

* *

Gila-monster venom and computational irreducibility: The idea is to start with a NYTimes article on drug discovery that starts with Gila-monster venom and ends up with Wolfram’s concept of computational irreducibility. This is about search, computation, and the complex and irregular structure of the (natural world).

* *

My work with Ramesh: Notes on conceptual ontology and hypergraphs in conceptual space.

* *

LLMs and literary study: Can we use LLMs to analyze the thematic structure of literary texts?

In particular, can we use them to examine Bloom’s these about Shakespeare as “inventing” the human. Are there themes that appear first in Shakespeare? We need more than Bloom’s vigorous assertion on this. We need to examine the thematic structure of prior texts and of Shakespeare’s texts and show that there are things new in Shakespeare. Do those new things then continue if texts after Shakespeare? To do this properly we need to examine a lot of texts.

I’d also like to know if LLMs could be used to find ring-composition in literary texts. It is by no means obvious to me that they can.

Look at me!

It will be interesting to see how AI affects medical practice

Not too many years ago Geoffrey Hinton confidently predicted that radiologists would soon be replaced by AI. That didn't happen. But now...

The New York Times a small study (50 doctors, a mix of residents and attendings) in which ChatGPT-4 outperformed physicians in diagnosis based on a case report:

...doctors who were given ChatGPT-4 along with conventional resources did only slightly better than doctors who did not have access to the bot. And, to the researchers’ surprise, ChatGPT alone outperformed the doctors.

“I was shocked,” Dr. Rodman said.

The chatbot, from the company OpenAI, scored an average of 90 percent when diagnosing a medical condition from a case report and explaining its reasoning. Doctors randomly assigned to use the chatbot got an average score of 76 percent. Those randomly assigned not to use it had an average score of 74 percent.

The study showed more than just the chatbot’s superior performance.

It unveiled doctors’ sometimes unwavering belief in a diagnosis they made, even when a chatbot potentially suggests a better one.

Of course, reading an x-ray and analyzing a case report are very different activities. Still...

There's more at the link, including a brief look at INTERNIST-1, an old-school AI system developed in the 1970s for diagnosis. It's clear to me that AI his here to stay, in general, and certainly in medicine. What's not at all clear is just how it's going to be used. Obviously, that will change over time as AI capabilities develop. While thinking about that you might look at Hollis Robbins' post, AI and The Last Mile:

While we worry about AI replacing human judgment, the real story may be how AI is creating a market for that judgment as a luxury good, available only to those who can pay for the “last mile” of human insight. What do I mean by this?

The challenge of mail delivery from the post office to each home or from a communication hub to each individual end user is known as a “last mile” problem. In the paper newspaper era, the paper boy was the solution to the last mile problem, hawking papers on street corners or delivering papers house by house in the early morning before school. The postal carrier is a solution to the last mile problem. DoorDash is a solution to the last mile problem in the food business. [...]

What I’m calling “the last mile” here is the last 5-15% of exactitude or certainty in making a choice from data, for thinking beyond what an algorithm or quantifiable data set indicates, when you need something extra to assurance yourself you are making the right choice.

Why the music industry sucks [Mary Spender]

Well, Hi there!

3QD Catch-Up: Affective Technology, Georgia on My Mind

Whenever I publish a new article at 3 Quarks Daily I also publish a somewhat shorter post here at New Savanna. That post links to the 3QD post and comments on it, or perhaps adds some further thinking. But that doesn’t always happen. I’ve missed giving notice for my 3QD articles in July, August, and September. 

The articles for July and August continue the series on Affective Technology that I began in June. That first article was about Poems and Stories. In July I published Emotion Recollected in Tranquility. The phrase is from Wordsworth: “I have said that poetry is the spontaneous overflow of powerful feelings: it takes its origin from emotion recollected in tranquility.” That is, there is an emotion that we experience at some moment in time. Then, at a later time, when we’re doing something else, or perhaps nothing beyond lazing about as we let our mind wander as it will, that emotion calls to us. We hear it, take hold of it in words, and bring it back, creating a poem in the process. But before that I talk about state-dependent memory, the idea that our ability to remember something depends, in part, on our mental state at the time of remembering. Remembrance works best is there our current mental state resonates with our mental state during the original experience; this resonance is (hypothesized to be) a function of neurochemistry. But what if you can’t establish that resonance? Does that mean that your memory of that experience is gone, at least until you can establish resonance? I conclude by discussing Shakespeare’s Sonnet 129, “The Expense of Spirit.” 

That sets us up for the third article in the series, Coherence in the Self, which I published in August. If our memories are neurochemically sensitive, that implies that our ability to remember the events of our lives depends, in part, on establishing neurochemical resonance with past states of mind and the events that engendered. If we can’t establish that resonance, then those events are lost. Our autobiographical self is shattered, a topic I take up with a discussion of dissociated identity disorder (DID), aka “split personality.” I then go on to argue that, by presenting us with a wide range of emotions, literary texts (and, by implication, other expressive culture) help us to construct neurochemical “scaffolding” that helps us keep in touch with our past. This strikes me as more fundamental than the (moral) improvement folks are forever attributing to (good) literature, though there may be a bit of that as well. Who knows? After discussing play I conclude considering Kenneth Burke on how we use literary texts to make sense of our lives. 

September’s post is a bit different. I discuss five recordings of “Georgia on My Mind”: the original by Hoagy Carmichael, Ray Charles’s 1960 version, a harmonically-dense version by Jacob Collier, a beautiful jazz rendition by Dexter Gordon (tenor sax), and I conclude by another jazz version, this one by trumpeter Lew Soloff. That video is why I wrote this post. Notice that he begins with an a cappella trumpet line he took from the intro to the Ray Charles version. 

 

Soloff’s style is interesting. It’s not swing, bebop, hard bop or cool, but it’s not modal or free either. In a way, it’s a kind of 1960s mainstream, but extended and opened-up (such as Soloff’s passage with only his mouthpiece or his use of the plunger mute) with a freedom that hardly existed before free jazz. That’s what makes it so interesting, that and Soloff’s soulfulness.

On the ethnographic study of expressive culture: If a Martian anthropologist were to study American culture, would she assign special privilege to the canonical texts of the Western traditions?

This was originally posted to The Valve on 7 Feb 2007 under the title: Discipline: The Aesthetic and the Ethnographic. I’ve edited it slightly so as to insert the proverbial Martian anthropologist into the mix. The distinction I’m making here, between the aesthetic and the ethnographic approach to culture, should be compared with a contrast I've made between ethical and naturalist criticism.

Scott Eric Kaufman’s recent post about the graphic novels issue of MFS Modern Fiction Studies got me to thinking about disciplinary matters. On the face of it, as Scott noticed, the special issue seems concerned about canonical legitimacy, hence the prominence given to a Pulitzer-winning text, Art Spiegleman’s Maus. My initial impression – after reading Scott’s post, the first and last articles in the issue and abstracts for the rest – was that there was nothing about manga in the issue. Since, as far as I know, manga constitutes the single largest body of graphic narrative in the world, that absence seemed rather glaring.

But, I suspect that my interest in manga (and anime) is rather different from the interest the MFS authors have in graphic narrative. I’m not terribly interested in claiming lasting canonical legitimacy for this or that manga. It’s clear that some of these materials seem intended for pre-teens, others for young teens, older teens, adults, males and females, and so forth. And some of them may well deserve a place in some high-art canon. But at the moment I’m not interested in arguing those distinctions. I treat manga as examples of human culture, and that’s why I’m interested in them. Through studying them we can learn something about the mind and about culture.

It’s a matter of how one frame’s one’s inquiry, of how one justifies the range of texts open for consideration. For the purposes of discussion, I’ll use the terms “aesthetic” and “ethnographic” for these two stances. Note that this distinction is not necessarily about the methods one uses in studying a body of work.

The Aesthetic Critic

The aesthetic critic is concerned to study those texts that are the best a given culture – or all humankind – has produced. It seems to me that the academic study of literature has traditionally chosen its objects of study from the aesthetic point of view. It is the study of a so-called canon of texts that have passed the test of aesthetic excellence.

The nature of that test is not very clear. As I said in the comments to Scott’s post:

I think that judgments about good and bad literature are arrived at on an intuitive basis and communally negotiated in terms of richness, complexity, reflexivity, and so forth, while quoting appropriate passages here and there to illustrate what’s being talked about. But, such examination and discussion are applied ONLY to those texts judged to be good. It is simply assumed that the bad ones do not measure up on those rather vague standards.

One of the standard clichés is that good texts can support many readings; that’s a measure of their goodness. And, what do you know, we have many readings for the good texts. That’s because the community that believes this doesn’t bother to provide readings for bad texts.

As for the value of these readings, my guess is that it has more to do with the critic than with the text or texts being criticized.

What we do not have anywhere, as far as I know, are explicit criteria applied to a wide variety of texts with the evaluations being done in some way that permits objective comparison between valuations.

The work thus done, the preservation and promulgation of a body of cultural material, of texts, seems to me necessary and valuable one. But one might have questions about institutionalization about, for example, how the primary winnowing of modern fiction into a canon has deserted the journalistic sphere in favor of the academy. But, for the moment, I want to treat that as a secondary issue

 

Given this, how has this discipline proceeded? On the face of it, broaching such a topic in the space of a brief blog post is absurd, with the absurdity multiplied by the fact that I’ve read little of extant literature on the professionalization of literary studies. Nonetheless, I want to float some informal observations.

In the first place, explicit aesthetic evaluation is not and has not been a major part of the discipline. We do not spend a lot of time arguing for the excellence of William Shakespeare or Jane Austen. We simply assume it.

In the second place, and more importantly, we have done and continue to do various kinds of work around and about the canonical texts. There is literary biography, various kinds of literary history, source and influence study, study of tropes and imagery, and so forth. However this is framed, there’s a great deal of accumulated knowledge of how these canonical texts relate to one another, to external events, to the human mind, and, for that matter, to non-canonical texts. Thus, while the selection of a universe of discourse may be steeped in mystery, the profession has quite a bit to say about that universe that is not so mysterious.

The Ethnographic Critic

The ethnographic critic, whom I’ve modeled on the proverbial Martian anthropologist, is not focused specifically on the sublimely beautiful. Being from Mars, this anthropologist has no personal stake in any particular human community. She does not live here as a citizen of Sierra Leone, Iran, Sri Lanka, Peru, Norway, or for that matter, of the United States of America, and so she is not trying to advance the values of her home state. Rather, she is studying earthlings. Period.

If the text is a human one, then it is within the scope of her study. Thus the range of texts within the purview of the Martian anthropologist, of the ethnographic critic, is much larger than that within the purview of the aesthetic critic, who serves a particular cultural tradition. The canonical texts of specific cultural traditions are no more than a proper subset of the ethnographic critic’s range. In fact, that range is so large that it may be crippling, an issue which Franco Moretti faced in Graphs, Maps, Trees, in which he espoused “distant reading” in which individual texts figured only as data points in a statistical distribution. I’ll return to this issue shortly.

As you might have suspected, I think of myself as an ethnographic critic. I am interested in the human mind, the brain, and culture; and how they interrelate. In particular, I have spent a great deal of time with the cognitive and neurosciences and so find even very simply texts – such as those created for children – to be absorbingly complex. When you look at culture in cognitive and neural terms, phenomena that are beneath the purview of the aesthetic critic all of a sudden seems almost hopelessly complex. That is to say, I don’t need the richness of Coleridge or Dickens to hold my interest. Conversely, I haven’t the foggiest idea of what it would mean to account for a Shakespeare play in cognitive terms, nor does this cause me any anxiety.

But that’s incidental. I don’t think one needs to be interested in the cognitive sciences to work from an ethnographic point of view. One simply needs to be interested in culture wherever and however it is. I would think that students of popular culture and of cultural studies have adopted the ethnographic point of view – though there may be some aesthetic anxiety that is assuaged by reading popular forms as existing in opposition to (evil) high cultural institutions.

It seems to me that one the most troublesome issues that arises within the ethnographic stance is how to choose just what to study, for it would be impossible for study everything within the very generous scope of the ethnographic perspective. The anthropologist doing fieldwork in a small-scale society faces a daunting task; but one might imagine that a team of anthropologists might “cover it all” over several decades. But the huge multiplicity of texts generated by complex modern cultures, how can one possibly study everything? And if one doesn’t study everything, how do you pick just which texts to study? Obviously, one has to sample the space – as students of popular culture have been doing.

Coca-Cola and AI: It's the Real Thing! [Still] [Media Notes 144]

Alex Vadukul, Coca-Cola’s Holiday Ads Trade the ‘Real Thing’ for Generative A.I., NYTimes, Nov. 20, 2024. The opening two paragraphs:

With temperatures dropping, nights growing longer and decorations starting to appear in store windows, the holidays are on their way. One of the season’s stalwarts, however, is feeling a little less cozy for some people: Coca-Cola, known for its nostalgia-filled holiday commercials, is facing backlash for creating this year’s ads with generative artificial intelligence.

The three commercials, which pay tribute to the company’s beloved “Holidays Are Coming” campaign from 1995, feature cherry-red Coca-Cola trucks driving through sleepy towns on snowy roads at night. The ads depict squirrels and rabbits peeking out to watch the passing caravans and a man being handed an ice-cold bottle of cola by Santa Claus.

And then:

The internet pile-on arrived swiftly as consumers decried the commercials as uncanny valleyesque perversions of the company’s classic ads.

“Coca-Cola just put out an ad and ruined Christmas,” Dylan Pearce, one of the campaign’s many critics, said on TikTok, adding, “To put out slop like this just ruins the Christmas spirit.”

“This is legit heartbreaking,” another user, De’Vion Hinton, posted on X. “Coca Cola has been the gold standard in branding and advertising for decades.”

Alex Hirsch, an animator and the creator of the Disney series “Gravity Falls,” expressed a sentiment that other creative professionals have shared online, noting that the brand’s signature red represented the “blood of out-of-work artists.”

The company, however, is sticking by the ads.

“The Coca-Cola Company has celebrated a long history of capturing the magic of the holidays in content, film, events and retail activations for decades around the globe,” a spokesman for the company said in a statement provided to The New York Times. “This year, we crafted films through a collaboration of human storytellers and the power of generative A.I.”

Ah, remember back in the previous century when Coke decided to revamp its formula and produce New Coke? Ah, those were the good old days. But the best days are yet to come!

There's more at the link. Me? I think I'm going to have a good old Diet Coke.

Behold!

Why is Cobra Kai better than one might have expected? [Media Notes 143a]

When Cobra Kai first showed up on Netflix back in 2020 I noticed it – Oh, the karate kid continues, that’s nice – but had no interest in watching it. But then I read something, perhaps this piece in The New York Times, indicating that it was actually rather good. So I decided to give it a try.

I liked it, I actually liked it. It’s a martial arts movie, and I’ve got a minor interest in such movies. But it’s got more going for it.

You may remember that back in 1984 there was a film called The Karate Kid, which was about a teenaged boy who is mentored by a middle-aged Japanese handyman and karate sensei. The film did fairly well and was followed by a couple of sequels. I have a vague recollection of having watched, I don’t know, that last half of the original film on TV at some time. But that was the extent of my involvement.

Cobra Kai is a sequel to those films, but with a twist. Rather than pick up where the original series left off, it starts when the teenaged protagonist of the original film have reached early middle age. Here’s how the Wikipedia entry explains the premise:

Thirty-four years after being defeated by Daniel LaRusso in the 1984 All-Valley Karate Tournament at the end of The Karate Kid (1984), Johnny Lawrence suffers from alcoholism and depression. He works as a part-time handyman and lives in an apartment in Reseda, Los Angeles, having fallen far from his wealthy lifestyle in Encino. He has an estranged son named Robby, from a previous relationship, whom he has abandoned. In contrast, Daniel is now the owner of a successful car dealership and is married to co-owner Amanda with whom he has two children: Sam and Anthony. However, Daniel often struggles to meaningfully connect with his children especially after his friend and mentor Mr. Miyagi passed away prior to the series’ beginning.

After using karate to defend his teenage neighbor Miguel Diaz from a group of bullies, Johnny agrees to teach Miguel the way of the fist and re-opens Cobra Kai. The revived dojo attracts a group of bullied social outcasts who find camaraderie and self-confidence under Johnny’s tutelage. The reopening of Cobra Kai reignites Johnny’s rivalry with Daniel, who responds by opening the Miyagi-do dojo, whose students include Sam and Robby, leading to a rivalry between the respective dojos.

Whereas The Karate Kid had Johnny Lawrence as the bad guy (“villain” doesn’t seem quite the word) and Daniel LaRusso as the good guy, Cobra Kai is not so polarized. Both men get to be jerks and both get to be decent and sympathetic. But they maintain their antagonism and work it out through their respective dojos, Cobra Kai and Miyagi-do.

So we’ve got action on two strata, among the adults and among the teenagers. The teens are motived by their own romances and beefs and join one dojo or another in the process of working things through. But Johnny and Daniel pick up the conflict from their youth and use their dojos as a vehicle for working it out. That conflict is cloaked in different martial arts styles. Cobra Kai blunt, aggressive, and brutal while Miyago-do strives for ‘balance’ and is more ‘spiritual.’ The stylistic difference is obvious in the fights ¬– of which there are many – and is the occasion for much conversation throughout the series. There’s even an attempt to blend the two styles in opposition to yet a third dojo which shows up in the middle of the series.

Which is to say, things get complicated. More actors, more complex interactions, kids moving from one dojo to another as their relationships between one another and with their elders shift. By the second or third season the two dojos are fighting of karate dominance in San Fernando Valley. By the sixth season – in process, 10 episodes are online, the last five will come next year – we’re talking about dominance of world-wide (teen) karate. Somewhere in this process we see genuine evil, with people getting seriously injured and even killed.

So, what do we have? A conflict between two teenage boys gets reignited in their early middle age where it becomes amplified into a war for dominance of the karate world. On the one hand, it’s rather ridiculous, adults using their dojos as vehicles for working out their conflicts. How Freudian! [Indeed. Think about that for a minute, think about it very seriously.] The series is aware of this and indicates that awareness in various ways. 

It’s brutal, but also lighthearted and funny. And there’s plenty of fighting, well-choreographed, and well filmed. It’ll be interesting to see how things end.

* * * * *

And exercise for the reader: Like so many things these days, this series just begs for a Girardian interpretation. On the one hand we’ve got mimetic desire operating on individual relationships. But we’ve also got larger scale sacrificial rhythms revealing themselves in the various gang battles and organized tournaments. Who are the scapegoats?

Scientific American throws shade on Elon

Well, not so much the magazine as their senior opinion editor, Daniel Vergano, who wrote an article entitled, Elon Musk Owes His Success to Coming in Second (and Government Handouts) (Sept. 13, 2023).

The first three paragraphs:

The superstar success of Elon Musk, the world’s richest man, with a fortune now estimated at $246 billion, poses quite a puzzle. How can someone whose wealth and fame rest on innovative technologies—electric cars and rocket launches—be such a doofus? Never mind his cursing at advertisers, getting high on a podcast, fostering white supremacists or challenging another peculiar rich guy to fistfights, the globe’s most famous technology mogul has spent much of the past two years blowing $44 billion on Twitter (now X) only to drive it into the ground.

And yet this same character remains critical to Ukraine’s defense against Russia, U.S. national security space launches and the American automotive industry’s hopes of withstanding a flood of cheap Chinese electric vehicles. His Dragon capsule will retrieve astronauts stranded on the International Space Station by his doddering space competitor, Boeing, NASA announced in August. This same genius crudely proposed impregnating the pop star Taylor Swift, in front of millions of people, following the second presidential debate in September. Again, how can someone be so essential, so rich and, at the same time, such a dingbat?

A look at history and economics suggests that Musk, now age 53, won his fortune from a “second-mover” advantage, well known in innovation scholarship. With regulation, marketing and basic principles established, a savvy second comer—think Facebook trampling MySpace or Google crushing Yahoo—can take over a burgeoning industry with a smart innovation pitched at a ripe moment. Musk also enjoyed a great deal of good fortune, otherwise known as handouts from Uncle Sam and less astute competitors.

The rest of the article fleshes out that last paragraph.

I wouldn’t use the words “doofus” or “dingbat” but, yes, he’s a bit strange. So?

As for the rest, OK. Am I an Elon-fanboy? No. That’s not how I am.

Does he mess up? Yes, what he's done with Twitter is not a success story. And no matter how much he blathers about free speech, free speech is not his game at all. He's egocentric. I think it’s incredible. Still, look at what he’s done.

Vergano’s critique is weak tea.

POW! Tomato

How far can next-token prediction take us? Sutskever vs. Claude

One of my main complaints about the current regime in machine learning is that researchers don’t seem to have given much thought to the nature of language and cognition independent from the more or less immediate requirements of crafting their models. There is a large, rich, and diverse literature on language, semantics, and cognition going back over a half century. It’s often conflicting and thus far from consensus, but it’s not empty. The ML research community seems uninterested in it. I’ve likened this to a whaling voyage captained by a man who knows all about ships and little about whales.

As a symptom of this, I offer this video clip from a 2023 conversation between Ilya Sutskever and Dwarkesh Patel in which next-token prediction will be able to surpass human performance:

Here's a transcription:

I challenge the claim that next-token prediction cannot surpass human performance. On the surface, it looks like it cannot. It looks like if you just learn to imitate, to predict what people do, it means that you can only copy people. But here is a counter argument for why it might not be quite so. If your base neural net is smart enough, you just ask it — What would a person with great insight, wisdom, and capability do? Maybe such a person doesn't exist, but there's a pretty good chance that the neural net will be able to extrapolate how such a person would behave. Do you see what I mean?

Dwarkesh Patel

Yes, although where would it get that sort of insight about what that person would do? If not from…

Ilya Sutskever

From the data of regular people. Because if you think about it, what does it mean to predict the next token well enough? It's actually a much deeper question than it seems. Predicting the next token well means that you understand the underlying reality that led to the creation of that token. It's not statistics. Like it is statistics but what is statistics? In order to understand those statistics to compress them, you need to understand what is it about the world that creates this set of statistics? And so then you say — Well, I have all those people. What is it about people that creates their behaviors? Well they have thoughts and their feelings, and they have ideas, and they do things in certain ways. All of those could be deduced from next-token prediction. And I'd argue that this should make it possible, not indefinitely but to a pretty decent degree to say — Well, can you guess what you'd do if you took a person with this characteristic and that characteristic? Like such a person doesn't exist but because you're so good at predicting the next token, you should still be able to guess what that person who would do. This hypothetical, imaginary person with far greater mental ability than the rest of us.

The argument is not clear. One thing Sutskever seems to be doing is aggregating the texts of ordinary people into the text of an imaginary “super” person that is the sum and synthesis what all those ordinary people have said. But all those ordinary individuals do not necessarily speak from the same point-of-view. There will be tensions and contradictions between them. The views of flat-earthers cannot be reconciled with those of standard astronomy. But this is not my main object. We can set it aside.

My problem comes with the Sutskever’s second paragraph, where he says, “Predicting the next token well means that you understand the underlying reality that led to the creation of that token.” From there he works his way through statistics to the thoughts and feelings of people producing the tokens. But Sutskever doesn’t distinguish between those thoughts and feelings and the world toward which those thoughts and feelings are directed. Those people are aware of the world, of the “underlying reality,” but that reality is not itself directly present in the language tokens they use to express their thoughts and feelings. The token string is the product of the interaction between language and cognition, on the one hand, and the world, on the other:

Sutskever seems to be conflating the cognitive and semantic structures inhering in the minds of the people who produce texts with the structure of the world itself. They are not at all the same thing. A statistical model produced through next-token prediction may well approximate the cognitive and semantic models of humans, but that’s all it can do. It has no access to the world in the way that the humans do. That underlying reality is not available to them.

What does Claude have to say about this?

I gave Claude 3.5 Sonnet Sutskever’s second paragraph and had a conversation about it. I wanted it to see if it could spot the problem. Claude saw various problems, but couldn’t quite find its way to what I regard as the crucial point. In the end I had to tell it that Sutskever failed to distinguish between the structure of the world and the structure of the semantic and cognitive structure expressed by the text.

My text is set in bold Courier while Claude's is plain Courier.

Neil deGrasse Tyson talks about his life, the cosmos, and stuff

From the YouTube page:

For a lot of people black holes and string theory were topics that were filed in the mental box labelled ‘things I will never be able to get my head around”. However, all changed when Neil deGrasse Tyson began appearing on TV screens.

0:00 Intro
02:02 Early context
05:47 Your parents direct influence
12:39 Your father being racially abused
23:36 How to decide what I want to do with my life
26:52 What are you concerned about with the human race
30:05 Social media polarisation
42:40 Do we matter
47:48 Where does happiness and meaning come from?
54:46 Whats required for a happy life for you?
01:00:17 The perfect way to tell stories
01:13:39 What do you struggle with
01:17:32 Mental health
01:30:04 The last guest’s question

Sailing on the Hudson