CBS, Byron Allen, and the business of bad television

 This is going to take a while.

Picking up on last week's thread, pretty much everything you've read about CBS replacing Colbert with Byron Allen's Comics Unleashed is wrong. Even by the abysmal standards of television journalism, the reporting here has been god-awful. We can say with a high degree of confidence that there is less here than meets the eye, but since none of the people who are paid good money to dig into these stories bothered to do any work, we are left to speculate about most of the finer details.

The meta-story of the demise of CBS late night is just the latest fodder in our long-standing rant about coverage of the television industry, a corner of the news landscape where huge stories go unreported, demonstrably wrong conventional wisdom is allowed to stand for decades, and the most egregiously distorted corporate spin is credulously reported as fact.

In this case, what has been described as a major change to the CBS lineup appears to have been nothing more than a standard and fairly trivial change to the existing syndication agreement between Allen and the network, applying primarily to, as far as I can tell, 15 of the more than 250 CBS stations in the country.

The key distinction at this point is the concept of an owned-and-operated station.

Normally, when we talk about the CBS network, we include all of the CBS affiliate stations regardless of who owns them. The same applies to ABC, NBC, Fox, and The CW. The relationship between these stations and their parent network can get a little complicated, and they do have a certain degree of freedom about what network programming they choose to air, though, to be honest, you'd need to talk to an expert to get the exact details.

As a general rule, however, the affiliate stations run the programming that is fed to them from the networks in the time slots matching the national schedule. I'm not sure what kind of legal requirements there are behind this, but the main reason is that network feed, even from a bottom-tier company like The CW, is still going to bring in better numbers than you could get with syndicated programming. Which brings us to the next key concept.

These stations have a great deal of freedom as to how they fill those hours not provided by the network. They can, of course, produce their own content, such as local news, but it is generally far cheaper and more profitable to go with something syndicated, be it reruns of The Big Bang Theory and NCIS or some daytime talk show or a game show not produced by CBS, NBC, etc.

The cost of these shows varies with their popularity, with the bottom of the barrel being sold on what is called a barter basis. The programming is provided to the stations for free with the understanding that the ad revenue from that show will be split between the station and the syndicator. In some cases, the syndicator will even pay a fee to the stations, making it something of a hybrid between traditional barter and the infomercial model.

Circa 1990, Saturday Night Live's Weekend Update did a bit on the debut of a cable channel with programming that consisted entirely of unscored, unnarrated footage of tropical fish swimming in a large aquarium. The anchor said that the channel carried fish programming 24 hours a day, seven days a week, except for 4:00 a.m. on Sunday morning when it aired the Byron Allen show.

Most barter agreements required stations to air a program a certain number of times during the week but did not specify what time of day they had to run it. As a result, if you see what appears to be a first-run syndication program airing after 2:00 in the morning, chances are that the show's numbers were so bad that the station was simply burning off its obligation in a time slot that virtually no one was watching.

Byron Allen, a truly gifted businessman, has become a billionaire largely because of his understanding of the television industry in general and the barter model in particular. Probably the best example of this is Comics Unleashed, an almost unwatchable show but a brilliant business model. The standard talk-show formula for interviewing stand-up comics has the host make small talk with the guest for a few minutes, then feed them a prearranged prompt so that they can go into their act. Comics Unleashed maintains only the slightest pretense of an actual conversation. Most of the segments consist of Allen turning to the guest and saying something like, "I hear you've been spending lots of time on the road recently," at which point the comic does a three-to-five-minute excerpt of their routine. As Norm MacDonald said about the show, the comics couldn't be more leashed.

The shows are obviously done as quickly and cheaply as possible, with lineups consisting mainly of C-list comics, with a few B-listers and the occasional former star like Jimmy Walker or David Brenner. The real genius, however, lies in how Allen has made the show almost perfectly evergreen. Comics are instructed to avoid politics and anything else remotely topical. The episodes are absolutely interchangeable. Having debuted in 2006, more than 350 are currently in the can, and for those watching at home it is often impossible to tell whether the one they are seeing is 10 weeks old or 10 years old.

In a sense, the frequent lack of recognizable stars might even be considered a feature, not a bug. With a lineup of comics you've never heard of, it is even more difficult to distinguish the old from the new, unless, like the previously mentioned Brenner, one of the comics has passed away. (Comics Unleashed's companion show, Funny You Should Ask, breaks with the pattern by including A-listers like Louie Anderson and Billy Gardell but otherwise follows the same cheap and interchangeable template.)

Commentators focused on that 85% drop-off from Colbert's finale are missing the bigger and bleaker picture. That debut number was based on a tremendous amount of publicity for a show that, despite a 20-year run, most people have never seen. More importantly, it appears that, due to that publicity and the suddenness of the Comics Unleashed announcement, virtually all CBS affiliates chose to carry the program in the 11:35 slot. While Byron Allen may have some sort of condition in his contract with CBS requiring those 15 owned-and-operated stations to leave him in Colbert's spot for a while, this almost certainly does not apply to the over 200 other CBS affiliates, which will very probably find something more profitable to air in that hour and will send Comics Unleashed back to its natural habitat.

Comics Unleashed has always done terrible numbers and it will return to terrible numbers in the future, but that's what it was designed for. Allen has designed a show that can turn a healthy profit with audiences in the tens of thousands. Both he and the show will be fine.

The local news shows of CBS affiliates (the source of about half their stations' ad revenue) will be screwed, but that's a topic for another post.   

 

I voted this weekend because...

1. I consider it my civic duty;

2. I was concerned about some of the races;

 

3. Voting here is secure, quick, and convenient;

 

4. I wanted the state of California to stop sending me those annoying texts:

Mark, if you live at ________________, records indicate you may not have returned your ballot for the 6/2 election yet. Your closest drop-off location is ____________, Burbank, CA _____. Return your ballot ASAP if you haven't: sos.ca.gov/elections/where-and-how
-OurBallot
Stop to end 

Exponential growth in demand is one of the sustaining assumptions of the AI bubble. This looks like something else

Following up on our tokenmaxing post.

Companies are starting to question whether soaring AI spending is delivering meaningful returns. An AI consultant tells us a client recently spent half a billion dollars in a month after failing to put usage limits on Claude licenses for employees.

[image or embed]

— Axios (@axios.com) May 28, 2026 at 10:15 AM

 


Aditi Bharadewriting for Business Insider: 

In a Rapid Response interview released on Saturday, Uber's operations chief, Andrew Macdonald, said it was becoming harder to justify AI costs within the company.

He said that Uber CTO Praveen Neppalli Naga went viral after telling The Information in an April interview that Uber had already blown through its Claude Code budget for 2026.

The comment led to what he described as a "head-exploding moment," sparking discussions about AI token consumption within the company and the trade-offs it creates, such as on head count.

He said that, based on talks with Uber's senior engineering leaders, he realized higher token usage did not translate into a proportional increase in useful consumer features.

...

Macdonald added that AI can seem free if you're "just a user sitting there coming up with interesting use cases" without paying for it. But ultimately, the company foots the bill.

...

Duolingo, for instance, walked back its decision to include AI usage in performance reviews after employees asked whether they had to use AI for the sake of using it.

"It felt like, rather than being held accountable for the actual outcome, we were trying to just push something that in some cases did not fit," Duolingo CEO Luis von Ahn said in a podcast interview in April.

Coding veteran Carl Brown has long been one of the sanest voices on the subject, so it's not surprising he has one of the best takes.

Apparently insider trading is now illegal

 

from ABC News: 

A Google employee fraudulently made more than $1 million by using inside information to place Polymarket bets on what users were searching for on Google, according to a federal criminal complaint unsealed Wednesday in New York. 

...

Spagnuolo, 36, is charged with commodities fraud, wire fraud and money laundering.

"Unlike the counterparties to his trades, Spagnuolo knew the outcome of these wagers before the trading public did because he had accessed Google's confidential, commercially valuable internal data," the complaint said.  

He correctly bet -- using an account under the name AlphaRaccoon -- that Google's most-searched person in 2025 would be the singer known as D4vd, according to the complaint. At the time he placed that bet, the prediction market Polymarket "assigned a near-zero probability to d4vd being 'the #1 searched person on Google this year,'" the complaint said.

 

A few things surprise me about this story.

Two or three months ago, I was at a neighborhood get-together, and I found myself in a conversation with someone who works for TMZ. The discussion turned to d4vd's case. The tabloid news show had been running with it since early September, when the badly decomposed body of a 14-year-old girl was found in the trunk of his Tesla. This brings up my first question: how was this story assigned a near-zero probability on Polymarket?

Putting aside the insider-trading aspect, which we will get back to, this is a remarkable failure of prediction. Even a fairly cursory bit of research into the big stories of 2025 would have revealed the completely unsurprising fact that lots of people were interested in this incredibly lurid story involving a highly popular recording artist. This is a case of a market failing to price in publicly available, easily accessed data.

While Yogi Berra (or perhaps the even more widely quoted apocryphal Yogi Berra) was largely correct when he said, "It's tough to make predictions, especially about the future," some predictions are tougher than others. This is one of the persistent problems in discussing and evaluating prognostications, be they model-based or otherwise. Sometimes predictions are so easy as to be effectively worthless, and it's important to give credit accordingly. The person who predicts the sun will rise in the east may have 100% accuracy, while the one who predicts it will rise in the west may only be right one in 100 times, but it's the second guy we should be impressed with.

The fact that, if not for the guy with inside information, Polymarket would have apparently gotten an easy prediction this wrong raises questions in the "dog that didn't bark" sense.

The other thing that surprises me is that this particular case of insider trading was illegal. To be clear, I'm not saying that it wasn't wrong—it certainly was—or that the FBI should not have gone after this guy—they certainly should have—but given how incredibly forgiving courts and regulators have become toward insider trading and other financial crimes, I honestly don't see this rising to that level. 

 From Wired:

 According to the complaint, Spagnuolo placed trades on Polymarket from around October 2025 to December 2025 using internal Google data. In one instance, he netted $1.2 million trading on who Google’s most-searched person of the year would be in 2025, correctly predicting that the winner would be D4vd, a once obscure singer who became the subject of intense public scrutiny after he was suspected of murder. (D4vd was ultimately charged in the case in April.) ["Once obscure" is an odd framing for a then rapidly rising young performer, but it is technically true in the vast majority of cases. -- MP]

“Unlike the counterparties to his trades, Spagnuolo knew the outcome of these wagers before the trading public did because he had accessed Google’s confidential, commercially valuable internal data,” FBI agent Brandon Racz wrote in the complaint.

 Given that the courts currently hold that “insider trading is not about fairness, it’s about theft,” why is the FBI even bringing up the counterparties to the trades?

Hello Goodbye

I have some thoughts on the finale of The Late Show with Stephen Colbert. Actually, I have lots of thoughts, but I'm going to save the discussion of what this means for the future of late night and of CBS/Paramount (particularly Paramount+) for a later post—I might even do one specifically on the extraordinary career of Byron Allen and how he fits into all of this—and play TV critic for now.

If you haven't watched the final episode of The Late Show with Stephen Colbert, you should, particularly the last 10 minutes. It was extraordinary television. Even more so if you caught it over the air, because the medium of broadcast television is at least in part the message here.

Perhaps the biggest difference between broadcast and streaming media comes down to the distinction between personal versus shared experience. Watching something at the same time as millions of other people is very much a new idea historically speaking.

Our modern conception of time was largely shaped by two technologies: the railroad and mass communication. Up until less than 200 years ago, time was set and experienced locally. Noon occurred whenever the sun was directly overhead, and the fact that 12:00 in Memphis might be 11:30 in Oklahoma City was a matter of no great concern.

Trains led to time zones; then radios and televisions led to synchronization. The telegraph started the process, but it was the media that streamed directly into people's homes that had the big cultural impact.

That sense of experiencing the same news and cultural events at the same time peaked in the '60s. Soon after that came second-run syndication, home video, satellite, and finally streaming, all greatly diluting that feeling of synchronization. But even now, broadcast television can convey a feeling of sharing a moment, particularly moments of historic importance.

It is, of course, true that these newer technologies, plus things like gaming and social media, have eroded broadcast television's numbers, though it is important to note that the actual viewership estimates are far more complicated than most commentators admit or even realize (for instance, shows that first appear on broadcast television, like High Potential, often score streaming numbers that crush most of the streaming originals), and that this erosion has been far slower than most industry watchers predicted. In the late seventies and early '80s, it wasn't unusual to see people predicting that at least one of the big three networks wouldn't make it to the 1990s. Instead, the next two decades saw that number go from three to five.

It is safe to say that, back in the early '80s, few pundits would have guessed that two of the most consequential stories of American politics and media would involve not just broadcast television but the dinosaur format of the late-night talk show.

The Kimmel suspension was arguably the first time during Trump 2.0 when a major corporation, one that had readily complied, reversed its position and stood up to the administration. Disney realized it had more to fear from its own customers than it did from Brendan Carr.

Kimmel was a humiliating defeat for the administration; Colbert was a Pyrrhic victory. Forcing him off the air greatly amplified his message and burnished his reputation while inflicting considerable costs on the Ellisons and CBS/Paramount. Combine this with the self-inflicted damage that Bari Weiss has done to the once-respected and profitable news division, and it's safe to say that the company is worth considerably less than it was when David Ellison bought it with his father's money. If things continue going the way they have been, Byron Allen might end up owning CBS after all.

In terms of the legacy of both the show and the host, the cancellation was the best possible outcome.

For serious history buffs, the obvious analogy here is that The Late Show with Stephen Colbert is to the anti-Trump resistance as The Smothers Brothers Comedy Hour was to the anti-war movement.

Somewhat unfairly, we remember that latter show almost entirely because of the circumstances of its demise. As long as historians study the anti-war movement and the counterculture, the Smothers Brothers will merit a section in the history books.

I think it was with COVID that Stephen Colbert truly stepped into the "America's dad" role. Every night we could have a Zoom call with a friendly middle-aged family man who openly shared our frustrations and our fears.

If anything, Colbert steered the show into this role of a traditional, reassuring, old-fashioned late-night talk show, sticking with the familiar format and reliably throwing in some dad jokes as the years progressed. This was perhaps best demonstrated by introducing the recurring segment with his wife, Evie. It was the sort of thing that could easily come off as cloying or fake, and I'm sure people who hated Colbert would have had that reaction had they watched the show. For fans, however, the exchanges felt spontaneous and charming.

Viewers and TV critics will inevitably go back and forth over the way the show wrapped up its run, particularly the last two episodes. In many ways, the penultimate episode would have been the more conventional send-off, with a star-studded lineup of guests honoring the departing host. By comparison, most of the actual finale was comparatively goofy and low-key.

The closing musical numbers were indicative of the two approaches. We could argue about where Bruce Springsteen and Paul McCartney rank in the pantheon, but obviously both are very close to the top. But while the caliber of the performers was comparable, the tone could not have been more different. "The Streets of Philadelphia" was and is a gut punch, at least for now the definitive protest song of the second Trump administration. The closing numbers of the finale were something quite different.

While the finale was not free of politics, the focus was on history, nostalgia, and an appreciation of what the show had been and what the theater that housed it meant.

I don't want to get overly nostalgic for the period, but the ability of broadcast television to create shared moments for a nation was probably exemplified by The Ed Sullivan Show more than by any other program. Ed Sullivan strove to make his show a cross-section of the best and biggest in popular culture. From plate spinners to avant-garde rockers, from vaudeville comics to George Carlin, from big Broadway dance numbers to Diana Ross singing "Love Child."

All of this came together perfectly in the last 10 minutes of the show, from its suggestion of the afterlife as a bunch of friends sitting around in the dark singing old songs (complete with an uncredited Jon Batiste and Elvis Costello) to the absolutely perfect number with Paul McCartney returning to the site of his American debut over 60 years ago, performing the apt "Hello, Goodbye," and closing with the image of the Ed Sullivan Theater disappearing and leaving a miniature replica inside a snow globe.

Remember that company that everyone with a NASDAQ index fund is about to be forced to buy?

The initial numbers are out and they aren't pretty.  

Joe Wilkins writing for Futurism:

Elon Musk has just pulled back the curtain on the biggest public stock offering in history, and the numbers are ghastly.

SpaceX, which is expected to go public on Nasdaq in June, just released the first round of financial summaries all companies are required to share when they’re about to sell stock to the public for the first time. The documents reveal Musk is targeting a raise of at least $80 billion — for a proposed valuation of $1.75 trillion — which would immediately make the rocket company one of the top 10 most valuable conglomerates in the US, Axios calculated.

With that kind of valuation in mind, one might expect SpaceX to be massively profitable going into its debut — but that’d be dead wrong.

According to the financial statement, the company lost $4.9 billion in 2025, even though it brought in around $18.7 billion in revenue. It’s not like that situation is about to turn around in time for the IPO, either: over the first three months of 2026, SpaceX posted further net losses of $4.3 billion.

 

 Allison Morrow writing for CNN (from her newsletter): 

Musk is practically unfireable
Musk will be SpaceX’s chief executive, chief technology officer, and chairman of the board. And he seems to have gone out of his way to ensure he won’t have to deal with any of the shareholder brush-backs he’s faced at Tesla. The filing shows Musk holds the majority of super-voting shares known as Class B stock.
 
Musk will be able to “elect, remove or fill any vacancy” among the top shareholders on the board and “have the power to control the outcome of matters requiring shareholder approval, including election of all our directors, and to control our business and affairs.”
 
Musk controls 85% of the shareholder vote, according to the filing, which means he’d have to vote to fire himself. 

This is a good time to remind those who haven't been following closely that Musk has a long history of using companies he controls to boost the stock of his other companies or to buy out his failed efforts. If this goes through, don't be surprised if SpaceX suddenly feels the need to order a million or so Optimus robots.

 

 

Matt Levine writing for Bloomberg (from his newsletter);

“We believe we have identified the largest actionable total addressable market (‘TAM’) in human history,” says the prospectus for SpaceX’s initial public offering, and, duh, of course. Every other company in human history has had essentially the same business model, which is “we will sell goods or services to people on Earth for money.” The upper bound on every company’s TAM has been something like the number of people on Earth times the amount of money they can spend, the gross domestic product of the Earth.

SpaceX has much higher ambitions. “Our mission,” it says, “is to build the systems and technologies necessary to make life multiplanetary, to understand the true nature of the universe, and to extend the light of consciousness to the stars.” The upper bound on SpaceX’s TAM is the gross domestic product of the universe. Elon Musk is going to make humanity a multiplanetary species, and then we’re really cooking. Why sell goods and services to a few billion people crammed onto Earth, when you can sell goods and services to a few trillion people spread among the stars? “We believe that our current space efforts will catalyze transformative breakthroughs that could reshape terrestrial industries and lead to the emergence of new trillion-dollar markets on the Moon, Mars, and beyond,” says the prospectus. Every other company is stuck in our dreary modern reality, but SpaceX operates in a Star Trek universe with obviously larger economic potential. [1]

I’m sort of kidding. Here is SpaceX’s actual description of its total addressable market [2] :

We estimate that our quantifiable TAM is $28.5 trillion, consisting of $370 billion in Space from space-enabled solutions; $1.6 trillion in Connectivity across $870 billion in Starlink Broadband and $740 billion in Starlink Mobile as well as additional opportunities in enterprise and government; $26.5 trillion in AI across $2.4 trillion in AI infrastructure, $760 billion in consumer subscriptions, $600 billion in digital advertising, and $22.7 trillion in enterprise applications. For illustrative purposes of sizing our addressable market opportunity, we exclude China and Russia from our global estimates.

The entirety of SpaceX’s estimated market for “space-enabled solutions” is about half of that for “consumer subscriptions” to AI chatbots. Substantially all — 79.6% — of SpaceX’s total addressable market is for enterprise AI applications. SpaceX looks around at the economic possibilities of extending the light of consciousness to the stars and thinks “we’re gonna build tools to automate spreadsheets and sell them to investment banks.”

When you get to the slides at the end, keep in mind that in order for SpaceX to live up to its projections, not only will generative AI have to go from losing money to being an enormous cash cow, but xAI will have to dominate the market. Unless you're scoring on non-consensual sexualized images generated, it is no off to a good start. 

Patrick Boyle (who weighed in previously here) revisited the IPO last week and was rather... critical.

 

 

Longtime Musk critic Thunderf00t had the clever idea of bringing Musk's own chatbot Grok into the discussion.

Remember what we said about xAI needing to dominate the competition? 

   

Memorial Day Repost

 

A good day for a recommendation

There is, of course, no such thing as the military perspective -- no single person can speak for all the men and women who have served in the military -- but if you are looking for a military perspective, my first choice would be Lt. Col. Robert Bateman who writes eloquently and intelligently on the subject for Esquire. Here are Bateman's recent thoughts on Memorial Day.

When the guns fell silent in the Spring of 1865, they all went home. They scattered across the country, back across the devastated south and the invigorated north. Then they made love to their wives, played with their children, found new jobs or stepped back into their old ones, and in general they tried to get on with their lives. These men were no longer soldiers; they were now veterans of the Civil War, never to wear the uniform again. But before long they started noticing that things were not as they had been before.

Now, they had memories of things that they could not erase. There were the friends who were no longer there, or who were hobbling through town on one or two pegs, or who had a sleeve pinned up on their chest. There were the nights that they could not shake the feeling that something really bad was about to happen. And, aside from those who had seen what they had seen and lived that life, they came to realize that they did not have a lot of people to talk to about these things. Those who had been at home, men and women, just did not "get it." A basic tale about life in camp would need a lot of explanation, so it was frustrating even to talk. Terminology like "what is a picket line" and "what do you mean oblique order?" and a million other elements, got in the way. These were the details of a life they had lived for years but which was now suddenly so complex that they never could get the story across to those who had not been there. Many felt they just could not explain about what had happened, to them, to their friends, to the nation.

So they started to congregate. First in little groups, then in statewide assemblies, and finally in national organizations that themselves took on a life of their own.

The Mid-1860s are a key period in American history not just because of the War of Rebellion, but also because this period saw the rise of "social organizations." Fraternities, for example, exploded in the post-war period. My own, Pi Kappa Alpha, was formed partially by veterans of the Confederacy, Lee's men (yes, I know, irony alert). Many other non-academic "fraternal" organizations got their start around the same time. By the late 1860s in the north and south there was a desire to commemorate. Not to celebrate, gloat or pine, but to remember.

Individually, at different times and in different ways, these nascent veterans groups started to create days to stop and reflect. These days were not set aside to mull on a cause -- though that did happen -- but their primary purpose was to think on the sacrifices and remember those lost. Over time, as different states incorporated these ideas into statewide holidays, a sort of critical legislative mass was achieved. "Decoration Day" was born, and for a long time that was enough. The date selected was, quite deliberately, a day upon which absolutely nothing of major significance had occurred during the entire war. Nobody in the north or south could try to change it to make it a victory day. It was a day for remembering the dead through decorating their graves, and the memorials started sprouting up in every small town in the nation. You still see them today, north and south, in small towns and villages like my own home of Chagrin Falls -- granite placed there so that the nation, and their homes, should not forget the sacrifices of the men who went away on behalf of the country and never came back.

Since ed reform mania died down, we haven't had much call to mention Goodhart's Law

 From XKCD

 

 

Nancarrow and McLaren

Tuesday's post about a late sixties technology that attempted to build a home theater revolution on something similar to super-8 film got me thinking about how clever 20th Century artists played with the media.

Wednesday, February 11, 2015

Analog recording without analogs  

[slightly edited repost]

And now for something completely different... 

 Two examples of artists who manually created their works on media normally used for analog recording.


Conlon Nancarrow

Nevertheless, it was in Mexico that Nancarrow did the work he is best known for today. He had already written some music in the United States, but the extreme technical demands his compositions required meant that satisfactory performances were very rare. That situation did not improve in Mexico's musical environment, also with few musicians available who could perform his works, so the need to find an alternative way of having his pieces performed became even more pressing. Taking a suggestion from Henry Cowell's book New Musical Resources, which he bought in New York in 1939, Nancarrow found the answer in the player piano, with its ability to produce extremely complex rhythmic patterns at a speed far beyond the abilities of humans.

Cowell had suggested that just as there is a scale of pitch frequencies, there might also be a scale of tempi. Nancarrow undertook to create music which would superimpose tempi in cogent pieces and, by his twenty-first composition for player piano, had begun "sliding" (increasing and decreasing) tempi within strata. (See William Duckworth, Talking Music.) Nancarrow later said he had been interested in exploring electronic resources but that the piano rolls ultimately gave him more temporal control over his music.[6]

Temporarily buoyed by an inheritance, Nancarrow traveled to New York City in 1947 and bought a custom-built manual punching machine to enable him to punch the piano rolls. The machine was an adaptation of one used in the commercial production of rolls, and using it was very hard work and very slow. He also adapted the player pianos, increasing their dynamic range by tinkering with their mechanism and covering the hammers with leather (in one player piano) and metal (in the other) so as to produce a more percussive sound. On this trip to New York, he met Cowell and heard a performance of John Cage's Sonatas and Interludes for prepared piano (also influenced by Cowell's aesthetics), which would later lead to Nancarrow modestly experimenting with prepared piano in his Study No. 30.

Nancarrow's first pieces combined the harmonic language and melodic motifs of early jazz pianists like Art Tatum with extraordinarily complicated metrical schemes. The first five rolls he made are called the Boogie-Woogie Suite (later assigned the name Study No. 3 a-e). His later works were abstract, with no obvious references to any music apart from his own.

Many of these later pieces (which he generally called studies) are canons in augmentation or diminution (i.e. prolation canons). While most canons using this device, such as those by Johann Sebastian Bach, have the tempos of the various parts in quite simple ratios, such as 2:1, Nancarrow's canons are in far more complicated ratios. The Study No. 40, for example, has its parts in the ratio e:pi, while the Study No. 37 has twelve individual melodic lines, each one moving at a different tempo.

Norman McLaren

McLaren was born in Stirling, Scotland and studied set design at the Glasgow School of Art. His early experiments with film and animation included actually scratching and painting the film stock itself, as he did not have ready access to a camera. His earliest extant film, Seven Till Five (1933), a "day in the life of an art school" was influenced by Eisenstein and displays a strongly formalist attitude.

That included painting on the optical sound track.

In the 1950s, National Film Board of Canada animators Norman McLaren and Evelyn Lambart, and film composer Maurice Blackburn, began their own experiments with graphical sound, adapting the techniques of Pfenninger and Russian artist Nikolai Voinov.[2] McLaren created a short 1951 film Pen Point Percussion, demonstrating his work.[3] The next year, McLaren completed his most acclaimed work, his Academy Award-winning anti-war film Neighbours, which combined stop-motion pixilation with a graphical soundtrack. Blinkity Blank is a 1955 animated short film by Norman McLaren, engraved directly onto black film leader, combining improvisational jazz along with graphical sounds. In 1971, McLaren created his final graphical sound film Synchromy.[4]

 

More irrational than the shoe company example but admittedly not as embarrassing

 

pretty depressing that there's basically nothing a company like Ford can do on the car business side of things to get a 25% stock bump, but say you're putting CATL cells in a box for data centers and boom

[image or embed]

— e.w. niedermeyer (@niedermeyer.online) May 14, 2026 at 11:19 AM

 

This is another one of those cases (and they have gotten awfully common over the past year or so) where simply running the numbers shows there is simply no way that a certain market move can be considered rational.

For a company like Ford, which, to put it in technical terms, is really big, it is extremely difficult for a new product line to substantially increase the reasonable valuation of the company.

With smaller companies, it is at least possible to justify huge relative increases in valuation. If a third-tier construction company announced it was going into the data center business, there are at least plausible scenarios where you would see a manyfold relative increase in revenue. It's the classic big-denominator-versus-small-denominator question.

You're probably saying at this point, "But the amount of money being spent on data centers is so unprecedented that it could move the needle even for the biggest of companies." That's true, but there are a couple of things we need to keep in mind. First, we need to distinguish between the amount of money currently being spent, the amount of money being invested but not yet spent, and the amount of money companies are proposing to spend. All of these numbers are huge, but they grow progressively more huge as you go down the list. Add to that the pervasive circular financing and creative accounting that defines the industry. (Things have gotten so bad that the CFO of OpenAI no longer reports directly to the CEO. That can't be good.) The upper bound of the numbers you've been hearing about depends on these companies raising hundreds upon hundreds of billions of dollars in the next three years. If that does not happen, the potential market is much smaller.

Even more important, Ford is talking about becoming yet another player in a highly competitive, relatively small sliver of the data center business. Most of the cost of these centers goes to GPUs, with power coming in second or third. Furthermore, most of the money spent on power goes to generation, not storage.

Don't get me wrong, in absolute terms we are still talking about a lot of money, but in relative terms, for a company with revenue of close to $200 billion a year, there is simply no way to justify the bump we saw in the stock.

This is the culmination of a decades-long trend of increasing market irrationality, originally driven by bubbles, deregulation, and demographics, becoming truly dangerous during the post-financial-crisis asset bubble until it reached its current febrile state, where the dumbest of dumb-money retail investors determine the direction of the markets.

This will end badly, and the longer we have to wait for the crash, the worse it's going to be. 

"Tomorrow's just a future yesterday." TV edition

I love this sort of stuff.  

A few points:

1. I was a huge fan of James Burke's Connections back in the day and his framing of technological progress is still an influence. 

2. Burke was correct about the impact of satellites on television, but he was somewhat optimistic about the adoption curve. It would take more like 15 years for all of those channels to start overwhelming viewers.

3. I'm a bit surprised that developers were still working on this kind of reverse kinescope in the late 60s, years after videotape had been adopted by broadcasters.  

4. Super-8 came out a few years before this and the Fisher-Price Movie Viewer came out five years later so the idea of a film-based home/classroom video system probably didn't seem too impractical. 

5. I'm sure that the picture quality was better with this system but one of the lessons of the history of home video  is that consumers were surprisingly indifferent to how good the images looked. 

6. The Prisoner had just wrapped. Maybe that's where they got the chair. 

1968: Is this the FUTURE of TELEVISION? | Tomorrow's World | Retro Tech | BBC Archive 

AI growth and tokenmaxxing

It is difficult to convey the abundance of warning signs suggesting that the AI boom of the mid-twenties is in fact an AI bubble. Note, I'm not saying that AI won't eventually reshape our world (I believe it will). I'm not even saying that large language models won't someday live up to their promise (though I am far more skeptical on that point). What I'm saying is that there is increasingly little chance that the industry will go from a money sink to a money gusher in the next three or four years, which means that the chances of our avoiding a larger and more painful version of the dot-com bubble are growing vanishingly slim.

Pretty much every part of the booster narrative looks, at best, sketchy under close scrutiny. Take perhaps the main pillar of the bull case: token and revenue growth. The following, the beginning of a massive recent tweet from a prominent investor, lays out the standard view.

 

(I love how he tries to frame his argument as a conservative case by suggesting a slight tapering off of his impossible-to-sustain growth curve.)

The idea that you can sustain this kind of exponential upward path is silly, but the problems with his numbers actually start before he begins to extrapolate. Where exactly has this growth been coming from?

Given the unprecedented circularity of current AI financing and business deals, it is always wise to consider the possibility that the numbers you're seeing are the result of Peggy Bundy accounting, but let's drill down further and see how these tokens are actually being burned through.

 

 Joe Wilkins writing for Futurism:

Many employers, in turn, have begun mandating AI use on the job, some even going so far as to fire those who don’t hop on board in order to justify their big-time spending on tech industry contracts. While the forced adoption of AI has major implications for the financial viability of the tech overall, it’s also giving office workers a perverse incentive to increase their AI use for non-productive tasks.

Case in point, the Financial Times reports, Amazon’s office staffers are increasingly using the company’s in-house AI agent MeshClaw to run personal tasks in a bid to get their quotas up.

In an attempt to get more than 80 percent of its developers to use AI every week, Amazon has introduced employee-specific AI usage targets in addition to a broader “token consumption” leaderboard that tracks how much each employee uses AI (in machine learning parlance, tokens refer to basic units of data used by AI models to understand text.)

But according to staffers interviewed by the FT, employees are gaming the system by increasingly using the mandated AI systems to automate personal tasks, a tactic known as “tokenmaxxing.”

...

For its part, Amazon told the FT that “thousands of Amazonians to automate repetitive tasks each day,” adding that the retailer is “committed to the safe, secure and responsible development and deployment of generative AI for our customers.”

A quick jaunt through Team Blind, a message board for verified employees of companies like Google and Apple, shows that the practice is widespread — or at least, widely acknowledged.

“I burn tokens to s**t my [project manager],” one Amazon employee wrote in a post from May 8. “Whenever my PM says stupid s**t, I launch 10 sub agents to s**t him. Great use of GPUs.”

Asked by a Microsoft worker what that means, the Amazon staffer replied “just paste the Slack conversation history and tell the agent to analyze the guy using 10 sub agents.”

Though less malicious, there are plenty of other threads from workers across the tech industry asking how others are maximizing token usage — receiving lots of free advice in the replies.

 This isn't just Amazon.

 Deirdre Bosa and Jasmine Wu writing for CNBC.

Meta and Shopify say they have created internal leaderboards that track how many tokens employees use. Nvidia CEO Jensen Huang has said he’d be “deeply alarmed” if an engineer earning $500,000 a year wasn’t using at least $250,000 worth of compute — measuring what an engineer spends on AI instead of what they produce with it.

Once companies start measuring AI adoption by volume, employees optimize for the metric instead of the outcome.

“If your goal is to just burn a lot of money, there are easy ways to do that,” said Ali Ghodsi, CEO of Databricks, which processes AI workloads for thousands of enterprises. “Resubmit the query to ten places. Put up a loop that just does it again and again. It’s going to cost a lot of money and not lead to anything.”

 

Only slightly more absurd than some real-life cases

A classic from XKCD

A great Charles Bronson story.

This has nothing to do with any of the topics we focus on here at West Coast Stat Views. I have no good excuse for posting it other than, as previously mentioned, it's a great Charles Bronson story, and I've always been a big fan of Bronson, particularly before the Cannon Films years.

As long as he was actually given something to do, pretty much all of his film and television work before the 70s was wonderful, wry, economical, and with real sensitivity. His performances on the Twilight Zone and Have Gun Will Travel were series highlights. He was often the best thing in those big, manly ensemble pieces, the Magnificent Seven, the Great Escape, and the Dirty Dozen (his closing comment was perhaps the best line in the last one). He more than held his own with Henry Fonda and Jason Robards in Once Upon a Time in the West.

He also turned in first-rate performances in lesser but still worth-a-look movies like Master of the World, the Mechanic (complete with 70s nihilism and a surprisingly obvious homoerotic subtext), and Red Sun (forget the critics, if the idea of a buddy action comedy with Bronson and Toshirô Mifune sounds like fun, you'll enjoy this movie).

Among others who shared my high opinion of the actor. Ingmar Bergman was on record as admiring his work and calling him "scandalously underestimated."

Charles Bronson's childhood was something out of Dickens which explains this wonderful story from Kurt Russell.

 

A tale told in tweets (Make sure to use the Muppet announcer voice when you read the word "space")

 

Are data centers in space a dream or AI’s next big thing? SpaceX, Blue Origin and tech firms see orbital server farms addressing Earth’s power and land issues but costs could be out of this world.

[image or embed]

— The Wall Street Journal (@wsj.com) May 12, 2026 at 9:17 AM

PICHAI: “.. There’s no doubt to me that a decade or so away, we’ll be viewing it as a more normal way to build data centers.” @wsj.com $GOOGL www.wsj.com/tech/spacex-...

[image or embed]

— Carl Quintanilla (@carlquintanilla.bsky.social) May 12, 2026 at 9:59 AM

I wonder if there's some reason why Google would want to encourage all of this silly data-centers-in-space talk...

[image or embed]

— markpalko.bsky.social (@markpalko.bsky.social) May 13, 2026 at 12:47 AM

finance.yahoo.com/news/why-spa...

[image or embed]

— markpalko.bsky.social (@markpalko.bsky.social) May 13, 2026 at 12:51 AM

Truly incredible stuff here. These people are brain dead www.wsj.com/tech/ai/data...

[image or embed]

— Rude Law Dog (@esghound.com) May 12, 2026 at 9:27 AM

BAGNAROK is Ed Niedermeyer's term for Musk's scheme to use the SpaceX IPO to create a massive short squeeze and force everyone with a NASDAQ index fund to buy his insanely inflated stock.

Keeping the data-centers-in-space hype following may be a necessary condition for BAGNAROK (tm) just as credulous reporting on humanoids is a necessary condition for keeping Tesla valuation from imploding. (And Google makes a lot of money if this goes through.)

[image or embed]

— markpalko.bsky.social (@markpalko.bsky.social) May 13, 2026 at 1:13 AM

Google/Alphabet has even more money money on the line when it comes to keeping the AI bubble going and this definitely helps keep that river of hype flowing.

— markpalko.bsky.social (@markpalko.bsky.social) May 13, 2026 at 2:11 AM

Data centers in SPAAAAAAAACE! youtu.be/EmI77ZBeJrQ?...

[image or embed]

— markpalko.bsky.social (@markpalko.bsky.social) May 13, 2026 at 2:11 AM