Sign in Subscribe

Journal

Journal

This journal is a daily account of my creative labors and casual reflections. It’s intended to be frank, not entertaining. It’s also freewritten, quite unlike my other writing. The ideas jotted down here are thoughts going through my head rather than definitive views or even firm opinions. They’re often completely wrong. So don't take them too seriously.

Listen to my most recent music:

September 11, 2026

Fairly smooth day. Did some work on the novel. Reviewed the addendum written yesterday - I am confident this was worthwhile. Also revamped the site homepage and master index. Now much simpler and easier to navigate on mobile (I hope). Letting go of the old homepage painting (my favorite Orientalist painting) was tough because so much passion and work went into that site concept based around my Memoirs, but it is time to move on and close that door. Not in a spectacular mood. Ready to drift into tomorrow…

September 10, 2026

Having realized in recently examining some of my older works that my understanding of the mind-body problem has advanced since my original “Critique” essay, I have appended a new addendum to the end that essay so that readers are not led into confusion by differences between my older and newer views. This addendum essentially recapitulates the points already made in A.13 of Optimalism, but with further remarks added to show how the original “Critique” essay leads into them. You can find it here.

September 9, 2026

Headlines in the world of math and AI today as multiple claimants argue over a breakthrough solution to a Millennium Problem tied to the Navier-Stokes equations. This is extremely advanced math I won’t pretend to be near understanding, but it evinces fresh troubles in our brave new world. An AI company is claiming the solution, and yet it appears very likely their AI stole the essential insights from the chat log of human researchers working on the problem independently. Furthermore, said AI company had its own team of crack human mathematicians working with their AI, and they spent more money just running that AI than it would have cost them to pay a hundred more human researchers to work on important problems full time for years.

In short, it is not at all clear that this major mathematical breakthrough represents a major breakthrough for AI itself to match the headlines so much as a publicity stunt designed to draw in more funding for a cash-burning business that is, financially speaking, on a cliff’s edge - at the expense of humans who may deserve the lion’s share of the credit. It also suggests the problems with theft and attribution I highlighted in AILOM have grown, remarkably, worse over time. Nearly all humans involved in research now use AI as a research assistant to some degree, often just to run more intelligent searches of the literature than were possible in the past. For instance, when putting together the liminal-home meditation, I spent weeks using LLMs to research every obscure visualization tradition I could find, such as 19th century academic architectural visualization and the 16th century ars memoria, locating examples it would have taken me ages to dig up from diverse sections of a large academic library (which I don’t have access to anyway). Yet if AI steals a researcher’s initial insights in the inchoate, pre-publication phase, then the entire incentive structure for doing research falls apart. And because AI training blends the data it ingests in a black box, its true sources can’t be audited directly. So there is always plausible deniability, and this deniability intensifies the problem, because it makes enforcement so difficult.

To be clear, all mathematicians do agree AI is going to have a transformative effect on their field. But so did the calculator. That doesn’t mean every claim of AI intelligence is true. And this one deserves particular skepticism. The whole fiasco seems to me more a disastrous PR blunder rather the triumph announced. The value proposition of the top American AI labs is, supposedly, that they’re more secure than using open-source or Chinese models. But if they’re stealing ideas directly from chat logs as you use them, then nobody doing frontier work in any field will want to use anything but a private LLM based on an open-source model if they have any choice about it. So trumpeting this controversial breakthrough could be putting a nail in hyperscalers’ coffins and ensuring the already likely commoditization of AI models does indeed come to pass.

After musing on these events and their broader relevance, I became newly irritated at the rapaciousness and poor character of the people at the top of our society. I also decided to add credits at the beginning of Optimalism, which I had previously stripped of any authorial references to encourage readers to focus puerly on the logic. Not because people yet believe that AIs can write something with the philosophical novelty and sophistication of this text, but because nobody knows for sure whether they will or won’t someday, and perhaps someday soon; and I want readers to be aware from the start that Optimalism, as well as the illustrations within it, are human creations - and hard-won human creations at that. This is especially worth spelling out because most readers will necessarily have to engage with the more difficult passages via the intermediary of LLM explanation, because they lack the necessary background. They should understand that the LLM is there just an intermediary for a conversation between humans - a translator, that is. I have also restored to the index of my nonfiction my deprecated prior essay A Pragmatical Analysis, so that anyone who wants to can trace the entire development of my work on religion, which dates back to 2020, with the essential kernel of Optimalism already present quite early on.

I worry for future innovators and researchers who come of age entirely within the AI era. Our society has no plan to deal with attribution issues in the future, and AI looting more generally is an injustice our rulers have decided not to punish. That may facilitate progress in the short-term (not because it actually speeds anything up, but because a politically impotent class has to take the loss). Yet if we don’t arrive at true ASI quickly (a scenario I consider possible but unlikely), the resulting destruction of incentives will be not just an injustice, but a net negative for progress as well.

September 8, 2026

Had a good writing day today but don’t want to say much about it yet. Will write more here soon.

September 5, 2026

One of the major improvements I alluded to yesterday is that Optimalism now has “tooltip” style previews for all the references, thanks to some vibecoding magic. This sounds like it would be a merely cosmetic effect, but it actually makes it much easier to read, at least if you’re on a PC. I don’t think they show up on a phone, unfortunately. I’ve also discovered that LLMs have now absorbed the text to some extent, or at least that they can retrieve it in full from the web. Unfortunately they are very lazy about this and hallucinate with much more frequency than if you paste the text in directly - and, they seem to rely on a cached version that’s not completely up to date. Whether these issues will improve in the future or not, I don’t know. But for now, if you do want to try discussing Optimalism with an LLM based on its web apprehension, you need to remind it to check the full text in every prompt; and for the time being I can’t recommend this approach: it’s still best to paste passages in. But, once you do paste passages in, the web access makes it easier for LLMs to look up references in the full text that aren’t in the prompt itself, and this usually improves the accuracy of their reading. I reckon LLMs will be able to give a very accurate explanation of Optimalism in a few years when their context windows expand. Or, if it becomes discussed enough that they consider it important to process it thoroughly in their training, which I naturally don’t expect to happen anytime soon. Normally, of course, I wouldn’t be keen to have LLMs process a damned thing I’ve written because I’m strongly against information looting—which is, at heart, what they do. But in this particular case, LLMs are the only interpreters that can give a reasonable number of people a chance of understanding the work. So, let it rip, I guess.

September 4, 2026

Have still felt very creatively uninspired of late, so I took a few days to touch up earlier works. This turned out to be less of a waste of time than I feared, and indeed I was able to make some genuine and unexpected improvements.

I gave some thought to the question of when a work is “done.” And, I came up with a clear answer. A work is done when the best possible change cannot improve it. That is a real definition of done. This point can be attained for simpler works. For long and complex works where the interactions between parts increase exponentially, it is effectively an impossibility; so in practice one has to make a judgment about diminishing returns. And yet because the value of mechanically reproducible works follows a power law, most people overestimate how quickly returns really diminish, and fail to seek an appropriate level of perfection, leading to great and avoidable waste. I find Ravel’s views to be most correct here, and keep a page of his remarks with me as a reminder.

Be that as it may, I am going to forge ahead again tomorrow and hope for better success.

August 31, 2026

Have not been in a great mood lately but also don’t want this journal to be all venting. So, sparse entries for the moment.

On a positive note, today I had an interesting realization. I have thought rather deeply about residential architecture for a number of years. Today I finally came to accept something I’ve resisted. The monoslope is the strictly dominant solution for house architecture. The only reason no one has noticed or accepted this is tradition. I could write an article explaining the details of why this is the case. However, I have sworn off any more nonfiction. So, it will have to be my private realization. There is one exception, of course: the riad. But to my mind the riad is just a monoslope eating its tail.

Added D.28 to Optimalism today. On the surface this is a new addition but really it’s just bringing forth a concept that was latent in the existing propositions and needed to be foregrounded with a term. However, I may expand on it with a brief addendum or proposition added to Chapter VI. We shall see.

To be clear, Optimalism is a consequence of working out the logic to an extent never previously attempted. It is not a translation of my accrued wisdom. I don’t consider myself very wise. I am not even sure wisdom is much of a thing. And I am still in the process of learning what the logic means. I think… my philosophical insight has leapt ahead of my ability to actually process it in life.

August 27, 2026

In the process of doing an extended interview that also includes another author. Pre-written responses rather than live tape, and rather time-consuming. It took most of the day to compose my first response. I assume it will be published in a few weeks.

August 26, 2026

I expanded VI.6N of Optimalism with a few brief but consequential sentences. Made minor edits to two of the addenda. Also improved the visuals with a super-cool leaf effect. Still not in a mood to say much here. Better luck tomorrow…

August 21, 2026

Need a little time off from the journal but will be back next week.

August 15, 2026

A few weeks back I mentioned how I’d researched Vajrayana visualization techniques and found some similarities to the meditation I’d put together for Optimalism, but ended up not using anything, or hardly anything, from that research because I didn’t see added value. This morning I started to wonder whether I hadn’t followed through enough. So, I went deeper into researching those techniques from another angle. This time I found, on a more abstract level than visualization itself, some transferrable elements in the approach. Consequently, I added a few new points to the advanced techniques section of the meditation, which I think are helpful.

I have been having some kind of blood-sugar issue lately trying to easy my diet with more honey and less meat. I don’t think that’s working very well; my mood has been off, and particularly as the day wears on. But - I still can’t manage white rice. My plan is to try rice once a fortnight. Hopefully at some point I will see improvement in my tolerance.

I had another look over the script that I posted yesterday. I’m pretty happy with it. Also made some minor visual tweaks to Optimalism and the website in general. I have lately been reflecting on large fonts. My feeling is that large fonts are a good idea because they discourage craning. And, they are prettier as well. I’m aware that my website font reads much larger than is currently customary. But the more I thought about that, the more I decided to lean into it, especially with the new font design and layout. I suspect many prominent sites keep the small craney fonts just to improve the visibility of ad sidebars, and everyone else copies them. Well, I am not showing ads. And I want to read things at arms’ length.

Amusingly, someone tried to post my GIMBY article on Reddit. And - it was censored by mods: deleted. I strongly discourage anyone intelligent from writing anything original for the public. It is not worth your sweat. There is no return, no effect. They are so retarded that they can’t even tolerate… promotion of low-density residential zoning! The only sensible use for intelligence seems to be siphoning money away from dumber people via distraction apps and financial engineering.

August 14, 2026

Feeling fuzzy-headed at the time I write this entry so I’m not sure how coherent it will be, and will keep it short. Today I rewrote the new addition I mentioned in the last entry because, as usual, I wasn’t happy with how I’d written it the first time. Also took some time to think about architecture, and that maybe I will casually publish some of my reflections and designs at some point - I had originally gathered material for a full article, and it seems a waste to not use it in any form, but whatever form it does take will be casual and easy and perhaps in a journal format rather than finely wrought like my articles.

More importantly, I think I’m about done with the script for the current piece. Going to print it below. I think… it’s very good. Though likely there will be further revisions and additions as the music comes along - as anyone who reads this journal has noticed, I never know when something is done. And of course, in reality this is meant to be presented with music, not read silently. I would reserve it to be read with the music rather than sharing it now, but I don’t expect many people to listen to this piece of instrumental chamber music anyway, so I guess it doesn’t really matter…

Missed Rendezvous, Eiffel Tower

I.

When I lay my ear
Against that dusty white tile
I could hear them thrum

To summon me below:
Muffled groans from the metro.

You were - where were you?
Lost in your endless spring:
In bed flirting with smoke rings,

Less careless than caring less
And thinking nonsensical things.

That's how you found me
Slumped over a piano:
Lazily lucky.

And some whim eked casual grins
From the same halves of our mouths.

We’d agreed to meet
By a turnstile underground
Amid the torrid crowd:

You arrived too late somehow,
Overslept I suppose, and now . . .

Here I am, riding alone.
Where are you, baby - alone too?
Tower glimmering above:

A Christmas tree, a missed rendezvous,
Fulgurant and frivolous like you.

Wheels thrumming, doors humming,
Packed to standing with more coming,
And tiny snowflakes

Blown from a black sky far
Down the stairs and into the cars.

❧

II.

One sunless room,
And a neighbor's woman wont
To bang my thin walls.

I folded your scrawl again.
Five rings . . . a coquette’s drawl.

Crosstown traffic to
A luxurious apartment
In a louche arrondissement,

Done up with unhemmed fabrics like
Some oriental restaurant.

Your fine tower view
Got no use that afternoon.
Your ceiling, later.

The bistro below lit it yellow
And squawked into night as you slept.

I stole out early.
Dawn made empty streets’ angles
Look bent out of true.

I walked almost halfway home.
Then, lost an autumn with you.

We can only feel
That shimmering-sky feeling
A few times before

It leaves forevermore,
And our sun is spent.

Yet time doesn’t stop
The moment it stops meaning
Anything at all -

It’s a song that drones on too long
After the singer is gone.

Heaven knows which love is real,
But I know which melody
You sing for me.

And while I have a chance to feel,
I want to hear it deafeningly.

Once illusion comes
To repossess her own
What remains to ring

Is only a thin tin echo.
So sing to me true or false - but sing.

We'd agreed to meet
At midnight underground
Amid this torrid crowd,

But you arrived too late somehow,
Overslept I suppose, and now . . .

Here I am, riding alone.
Where are you, baby - alone too?
See it glimmer above,

Like a signal tower for our lost love,
To send the words I couldn't say to you.

❧

Missed Rendezvous,
Eiffel Tower

J. Sanilac

August 11, 2026

Haven’t been in the mood to write much here the last few days. Better luck tomorrow.

Yesterday I added a few new paragraphs to The Illusion of Dominance between the addendum and appendix.

August 7, 2026

Below, a promotional thread I think is worth preserving because I frame an issue that’s been bothering me for a long time in an especially clear way.

Foundational Stupidity

The single most important fact about society is the stupidity of the average person. Everything flows from that. Yes, really - and no: you can’t just ignore it. Vital projects require collective action. So to get them done, you have to recruit the dumb.

Now... there are three ways to recruit the dumb.

First, the influencer way: just be dumb yourself. That way you’re speaking the same language. Ok, good for you, but the dumb leading the dumb is... dumb. As far as solving any real problems goes, this is a non-starter.

Then there’s the smart, honest, and well-intentioned way. You patiently explain true ideas in ways the average dummy can understand.

Unfortunately, all evidence leads me to believe this fails hard - not least because of the third way.

That third way is to be cunning, manipulative, and evil. You take advantage of the stupidity of the average person by tricking them for your own personal gain. This third way is the tactic that dominates. So, its masters wind up running the show.

But J, you say: isn’t there a fourth way? Couldn’t you have good intentions and be cunning and manipulative? Well, no. Not really. People who are honest are usually compulsively so. The class of people who are cunningly dishonest but have good intentions is trivially small.

And we also have to understand why evil people control public opinion. For one, they’re playing with fewer rules. That already puts them ahead. Better still, they don’t care what the truth is. So they work within the topography of stupidity. They never have to push anyone uphill.

Meanwhile the honest and well-intentioned are like Sisyphus on a steep slope. Very often the truth is uphill. And that would be hard enough already. Thing is, the evil are pushing the exact same boulder roughly down. This is a battle the good are not going to win.

There’s one oversimplification in the above: low intelligence is not the only issue. Lack of what you might call “veridical virtues” is a big contributing factor.

These virtues give you the wherewithal to see and choose the true and good even if it’s cognitively uncomfortable. They’re somewhat trainable. But because the evil are usually in charge… they don’t get trained. Especially today. Surprise, surprise. Most people don’t have them.

In other words, I’m not just talking about intellectual stupidity. The average person today is dumb in a much more profound way.

Of course, Hollywood is always telling the opposite story: good defeats evil, the true argument defeats the “strong” argument. This endlessly repeated narrative is why people who should know better overlook the consequences of stupidity for so long.

(This, and the misguided inclination to assume the best about other people.)

Let’s sum up. Society is the story of stupidity and its consequences. Keep that foundational model in mind from the start, and everything you see will make sense. Refuse to keep it front and center because it’s a “mean-spirited” attitude… and life will remain a mystery.

Yes, I learned this the hard way. But fortunately, now you don’t have to.

August 3, 2026

Did some minor touchup on the website design. Also improved the appearance of Optimalism in particular and more significantly, with some very appealing details. Looking quite pretty now - and pretty much exactly how I want it to.

There has been more news regarding AI advances in math along the lines I discussed not long ago. I consequently wrote a reflective thread for social media as I thought it was a topic about which I could say something of broad interest. I am going to reprint it below.

In the next few journal entries I am going to address some of these topics in more detail rather than just give a daily account of what I’m up to, because I think they are important issues that may soon be quite connected to what everyone is up to.

The Fall of Mathematics and the Human Desire for Meaningful Work

AI now appears to be conquering higher math. Why are mathematicians having a meltdown? Is it just about their paychecks? In this thread I’ll explain.

First of all, I’m not a mathematician but a starving artist surprised to see mathematicians join me in the brilliant-but-still-mostly-useless biotrash category. Surprised to see that our motivations were much closer than I had known.

AI is going to put many people of all different sorts out of work. From one perspective this is progress. From another side, it entails serious financial losses for certain groups. People who invested heavily in skill development will find that investment coming up zero.

And yet what’s going on here is different. It’s not mainly about lost investments. Mathematics and the arts are not lucrative fields. They’re poorly paid fields for highly talented and motivated people, whose main concern was to find work that is primarily meaningful rather than primarily lucrative.

Certainly some of us could have contributed more to the economy by doing better-renumerated jobs. But others among us give society a great deal for little return. Or, for a modest return, to the extent that meaningful work is part of the “pay package.”

Meaningful work is near the top of the pyramid of human desires. Not all feel it equally, and not all have the chance to make it a reality. But setting aside pure hedonists, it’s something most of us would prefer to have.

The tragedy of AI is not (mainly) that it’s sending skill investments to zero and putting mathematicians and soon many others out of business. It’s that it’s eliminating meaningful work - rather than eliminating meaningless work. If material wealth were held constant, this would be a straight loss, and a far cry from the reduction in drudgery provided by previous waves of automation. Of course, material wealth is not being held constant, and that matters a great deal - but it’s a different topic.

In the future we may need to manufacture meaningful work out of the vast meaningless wealth that AI will (hopefully) create. Otherwise an important human want will go unfulfilled. Yet sadly, there is no reason to assume this is possible. If machines can do everything better, then meaningful work in the strict sense of work can only be a fraud. Captain Cook in a satellite-mapped swimming pool with artificial waves is no longer Captain Cook. From this perspective we are witnessing humanity suffer a permanent loss.

Now, I can think of ways to cope with this in the arts. Artists won’t stop arting, but they will art in a different way. Vast and laborious projects will be out. Instead we’ll pursue projects where the process itself is rewarding (note that - in reality although not in the public imagination - this includes only a small fraction of artistic projects normally pursued in the past). Perhaps there are similar meditative and internally rewarding projects in mathematics, though I’m the wrong person to ask about that.

Artists and mathematicians are groups the public takes a dim view of. This is largely due to misconceptions, but it’s the case nonetheless. So you, Joe Public, may not think this thread is relevant to you. And for now, maybe it isn’t. Maybe “meaningful work” is a concept that doesn’t register with you anyway. But one day - perhaps not soon, but someday in our machine-run future - we’ll all have to find new ways to fulfill the upper reaches of our hierarchy of human needs and desires. Some will make due with endless reveling, but many others will want stronger stuff in their lives too. Artists and mathematicians are in this sense just the advance guard of society to come.

August 2, 2026

Tiring day. I finished the new font - I had left out the ligatures in yesterday’s version and there was also a little touchup in order. Its final name is Fairefeel. But after some hours I failed to build in the inky thickening I wanted in a way that didn’t cause a crash of some sort or another. Instead I resorted to using browser CSS. So, it does look right in the browser. And, it looks right in certain other programs, especially when the font is enlarged. But I couldn’t justify spending another half a day messing with it just to add one pixel on the side of each letter in a version nobody but me will see. Maybe some other time. It’s pretty good anyway, and I prefer drafting with a very large font anyway, so for the purposes that matter - drafting and web presentation - it is fine.

After this, I propagated the changes I’d tested here on the journal to the rest of the site. I think it looks quite nice. On the other hand, I learned more about the limitations of digital type by trying to work past them. Our tech overlords are not making a sincere effort to get the beauty of the printed page onto the screen - and right now it’s just not possible. But that is a story I don’t think I will write out in detail here…

Also went for my usual weekend hike today. Unfortunately I wasn’t feeling well enough to run. Disappointing because it would have been a perfect day for it. Oh well.

Hopefully done with my typesetting days and back to the usual projects tomorrow.

August 1, 2026

Spent the morning redoing the journal’s formatting, yet again. However, I made actual progress this time in several regards. First, I figured out how to get it to justify the body text only when viewed on larger screens. More importantly, I finally fixed the problem with this font that’s been bothering me for years. Example below so you can see what it is. The corrected version is at the top.

It’s based on a particular midcentury typeface that feels like an art-deco update to a classic. However, the digitization that I’ve been using since the website began shortened the lowercase ascenders. In the original they were taller than the capitals. This change created a “square” look. Well, I finally gave up on finding another font and decided to edit it myself and raise the ascenders back to where they were originally supposed to be. Now it looks elegant instead of square. The difference is especially noticeable in sentences that start with “the” (like this one). I’ve renamed it Fairfeel.

I am going to give myself some time to get used to this before I propagate it through the rest of the website. I also need to see if this lighter weight looks right on the other pages.

Not done for the day yet—still time to do some writing. Also, feeling somewhat healthier today.

July 31, 2026

Today I worked a bit on “Missed Rendezvous.” While I can’t claim I was highly productive, the work I did was good. A small excerpt below. Not the most important page, but might amuse readers of my orientalist writings. In the evening I felt rather unwell despite having attempted to take strict care of my health this week. However, I am not going to worry about that. Instead I am going to assume it’s just a small hiccup.

I wanted to add one thought to my remarks on the war from a few days ago. Many people are citing a 300 or else 250 operational minimum to the SPR. After looking into this in more detail, I think the minimum might be very much lower and perhaps closer to half those numbers; and this would allow the reckoning to be delayed by several months past the October date I had previously envisioned. If so, we will hear expert analysts panic too early, again.

I am going to revert this formatting closer to the original journal formatting, which I suppose I’d gotten used to. But it may swap unpredictably over the next few days while I make up my mind. The experimental new one one has/had certain merits.

Beyond that I have taken time to update myself on AI. This is important, so I’m writing it up extensively below.

For most of this year I was too occupied with Optimalism to closely follow the progress of AI. I have now taken time to update, and am somewhat surprised; or at least surprised enough to modify my predictions. I’m going to share my thoughts below.

My public remarks in the past have been quite bullish, to the point that they seemed ridiculous to many. For instance, in March 2023 I wrote that AI girlfriends would become a thing as soon as the late 2020s, with the likelihood increasing each year thereafter. So I can hardly be categorized as an AI bear. Privately though, I doubted whether AI’s impact would really extend very far beyond the social chaos itself, which seemed a near certainty. Ever since GPT5 promised much and delivered little, LLMs seemed to be rolling the top of the S-curve. Recent events have caused me to revise that view.

The main limit to AI’s advance has been the training data. For a game like go, AI could learn superhuman skill by playing an incredible number of games against itself. That is, it could generate infinite data synthetically. For the real-world tasks that have the most economic value, this is much harder. Whence the stall.

In the field of mathematics, AI engineers now seem to have successfully conquered this problem with “neurosymbolic” supplementation, which both extends LLMs’ capabilities and allows the creation of synthetic data. Don’t ask me to explain the details, but that’s the gist of it. As a result, the likelihood that AI will surpass peak human performance in at least significant subsets of mathematics within a few years is very high.

For most of us success in abstract mathematics is not of particular interest. But what’s downstream of that success is universally relevant. The mathematical abilities thus gained will allow AI to improve itself and the hardware that runs it. This is known as recursive self-improvement. It doesn’t mean immediate broad superintelligence, nor a straight shot to automating the entire economy, and its effect might be limited in both scope and degree; but it does enable acceleration in numerous fields of research and technology. It implies we are entering an unprecedented new period, where particular technologies may advance rather suddenly; and what we come out with on the other end is difficult to predict on the whole—though as for my longstanding predictions about AI girlfriends, they now seem like a sensible central case.

Despite the above, AI valuations have looked bubbly to me for some time, and still do. Not because the technology doesn’t do anything. Because commoditization from open models threatens hyperscalers’ ability to generate a financial return; and because their numbers don’t evince a likely near-term profit in the first place, while their gamble on AI becomes more and more expensive as each month of capex rolls on. Yet the potential of these recent advances all but ensures the US government will keep them running whether they crash or not (and whether investors are made whole or not). They are going to squeeze this lemon till the juice runs down their legs, as the saying says. And so, wherever we are going, we are definitely going there.

I don’t know that destination any better than anyone else, or what bumps will occur on the way, or whether another AI winter is just around the corner, after a significant fraction of higher math has been conquered by machines. I certainly don’t put any stock in Elon’s claims that by 2035 nobody will have to work. And while it is much more certain that by 2035 some of us will be out of work, I also wouldn’t want to predict who.

Amidst all this change and uncertainty, I simply enjoin everyone once again to think about what really matters. We are not digital, but social and biological. And our destiny is uniquely tied to advances in society (ha) and biology. Now more than ever, it is important that we direct advancement and material gains toward these domains. To cure disease, to cure aging, to cure ugliness, and to arrange society so each of us can find love and community. I’m not going to tar all tech bros with the same brush, but some of them are so in love with the machine, or the money, that they miss the forest for the mechs. We need to take rejuvenation, beautification, and yes, romance as seriously as they deserve and as soon as they deserve, and not forget what matters most in the mad rush of digital advances.

July 29, 2026

Having just archived the past for symbolic reasons as well as practical ones, I’m going to write a longer entry to start this new journal series, which will also sum up some of my recent concerns.

I’ve made very unimpressive progress in the last few weeks, leading me to question whether the evinced lack of inspiration and motivation is simply a personal failure, or a sign that I’m pursuing the wrong direction. With this question in mind I’ve written a series of notes to myself that I have not shared here. These notes mainly served to reveal the muddiness in my thinking, because after writing them and revising them to the point that they were clearly stated, I decided they were wrong. This is often the way I use notes; and if someone were to page through my notebooks later to discern my viewpoint, they would come away with a very misleading and even backwards impression, because they so often represent discarded viewpoints rather than adopted ones—and the eventually adopted views, I typically do not write down at all. Moreover, deciding on a course of action in life is a complex optimization problem. One cannot solve it by writing down rules and weights. But attempting to do so can reveal gaps in the consciously acknowledged rules and weights, and these hidden variables can then be better brought to the surface, improving one’s clarity of purpose. In that sense the notetaking was successful despite the notes themselves being wrong.

This week I was reminded that my student loans, incurred many years ago now, are still outstanding after all those years despite being only moderate in size; and that had I chosen some easier career path, in tech or finance, I could have paid them off within a single year more than a decade ago. As my window overlooks the local drug dealer and a wide variety of frank welfare abusers who pass their time patronizing the former and raising children at others’ expense instead of working, it is natural to reflect on the wisdom of my course, especially to the extent that all of the above are visibly wealthier than I despite laboring away at least sixty fewer hours per week. My health has been in a poor state for some time, and slow mending has limited what I can do, while also imposing unpredictable mood swings downstream of however my body happens to be malfunctioning at any given moment. Poor health is a somewhat separate issue; but it sharpens my feeling of much work with little reward, and shrinking opportunities.

The correct response to these misgivings is not as obvious as it appears. My numerous creations would have been impossible if I had taken a more prudent path earlier. Their high points do seem to me to have exceptional value. While their social influence is of course thus far negligible, one cannot exclude the possibility that some among my few readers will one day find themselves in a position to effect real change. And, I still have the opportunity to create yet more works of value. So the final outcome of all this is impossible to know at present.

Because I had the rare ability and opportunity to create what I have, the choice to neglect these and pursue personal gain instead could easily have led to regrets of another sort. And with ability and opportunity there comes, I feel, some higher obligation. I recently read an interesting comment from Michael Pettis, who wrote that of his time successfully trading on Wall Street, the only genuine value he contributed to the American economy was to use the proceeds to fund a nightclub. While we obviously need a financial sector, today much of it is just skimming money off the top of the economy, such that the net contribution is negative. In other words, it is parasitical. I mention this only to illustrate that the best use of one’s skills is not an irrelevant question, and the Randians who reduce it to dollars accrued are mistaken: much value can be represented financially, but some cannot, and some financial “value” is illusory and negative.

It seems to me, then, that I would need to live my life not twice, but perhaps a dozen times to be certain about the right course of action given my particular cluster of strengths and weaknesses and the goods and evils of the various choices I could have made; and the first time is not even halfway done.

Be that as it may, we only live in one direction. So my bitterness here has more to do with my current outlook than with any chance of making better (or worse) decisions a second time around, and compels me to wonder the following. Given daily evidence that the worst people are rewarded while the best are punished, how does one continue trying to do good? And harder still, given our society’s continual slide to greater depths of stupidity and the general decline of virtue, how can one justify producing anything for such a world rather than skimming as much money off the top as possible, as do so many other capable people? These questions may not have definitive answers, but one is forced to come up with an answer, or fall into paralysis. And that answer has to be chosen under great and unavoidable uncertainty about the outcome; because as I discussed a few weeks ago, the reception of artistic and intellectual works is inherently unpredictable, and feedback on the way to success is noisy and often opposite to the truth.

The answer I have come up with for myself is twofold. The first part is that a number of my original ideas have genuine utility for society, and still deserve a wider hearing. I ceased my promotional efforts after the public made its undying rejection of Dispelling Beauty Lies indisputably clear, but I have other and more important things to offer as well. Foremost of these is my insistence that we give better attention to the new opportunities arising in medicine, and rejuvenation in particular. Our society is leaning into information technology while leaving on the table important opportunities to improve and extend our health, which is easily the most important and valuable thing we have. This is something I intend to address in the upcoming novel. People should see utopian visions of what a world with better medicine might look like, but writers are almost solely giving them digital dystopias—and that is a gap I actually have the power to fill, and which might be beneficial at the very least to encourage readers dedicated to these fields.

The second part relates to the aesthetic realm. We are living in a world where entertainment is cheap in both senses—which reinforce each other. It is cheap because the cost is spread over a huge population, but for the same reason it is emotionally and aesthetically cheap. It is full of lies, and in many cases moral messaging that directly harms the character of naive viewers. AI slop is now flooding every channel with junk. Perhaps worst of all is that the predominant entertainment is optimized for addiction and sucks everyone toward the lowest common denominator. I would estimate that a majority of “independent creators” today are producing pornography in one form or another; but most influencers somehow manage to be even worse. So-called high art and academically-funded art are as useless as they’ve ever been. The result is an aesthetic food-desert with infinite free Cheetos with a side of kale. One can object that it has been this way for a hundred years, but erosion has made it gradually worse over time. This is especially tragic for the right half of the bell curve—people with intelligence and aesthetic sensitivity who, in a different society, would have filled their leisure hours with much richer and more rewarding stuff.

I don’t have the power to change this larger ecosystem. If people really want to be retarded (and they do), they are going to be retarded. And even with the best efforts, I don’t see the financing there to rebuild a healthier ecosystem. The elites who in past times would have protected and even funded the arts are today proud philistines, who in some cases actively arrange to damage them, perhaps due to some lingering teenage jealousy. Nevertheless, I should like to provide a small crack into an alternative and give those who have the patience and sensitivity—a few still exist, and will surely grow tired of eating Cheetos eventually—some poetry suitable for our time. Easier said than done when you are sick in the ghetto with shrinking opportunities; but I will have to arrange for this musical poetry with miraculous feats of imagination.

Thus, my concluding thought is to move forward with these two worthy goals in mind, and rest on the idea that they retain their value regardless of the various other difficulties and discouragements I happen to be facing. As for whether advances in AI will make it all useless, that same question is presently forcing itself on a number of fields beyond the creative, and the answer is that nobody knows for sure.

With these thoughts set down, I’m going to shift topics altogether and record some remarks on coming challenges of another sort.

When the war with Iran began my opposition was immediate, and recorded in earlier entries. At the time I estimated that a small nation in the Middle East had used a political-control network established over decades - really going back to WWII, and spanning blackmail, media influence, direct political financing, and calls on divided loyalty - to push the US into extremely foolish, and I will even say evil, acts of aggression. Again as noted in past entries, I have since given these views a larger context without abandoning them. The rulers of the American empire have multiple goals not limited to Middle-Eastern geopolitics. And the easiest way one can go wrong in sociopolitical analysis is to explain events with single causes. This is how conspiracy theories lead to erroneous views even when they contain (and often they do) elements of truth. Sociopolitical events always have multiple causes, because there are always multiple forces at work. Here the second relevant cause, and perhaps the most relevant, is our rulers’ desire to defeat China and ensure the American empire remains preeminent into the future. I have come to see their strategy as sophisticated and diabolical, not mere stupidity on the part of the executive as we are expected to believe.

The Chinese economy is dependent on mercantilist overproduction and exportation of the resulting demand deficit. The US strategy, as I see it, is to destroy global demand by dramatically reducing the oil supply. After this happens China can no longer export its demand deficit, because the rest of the world will stop buying, and China will fall into a prolonged crisis. Surplus countries will then begin a massive program of dirigiste capex to stimulate and rebuild their local economies. The US is in an especially good position to do this, and the coming crisis will justify any amount of stimulus desired. Whereas China, having already done too much, cannot run that playbook without digging themselves an even bigger hole. Their excess of manufacturing and infrastructure then becomes a drag rather than a boon. When all is said and done, the US will use this as an opportunity to reindustrialize, while China will suffer two or three lost decades, by which time demographic decline will bite hard and lock it into a downward spiral for a long time to come.

Such is my theory. Like most theories, it is probably wrong. However, it neatly explains why we have had months of kayfabe negotiations that seem unlikely to have ever been sincere: for the US, the main point is closing the strait, and winning or resolving the war is a secondary matter. Closing it for how long is hard to say, but long enough to create serious pain, and perhaps indefinitely.

Regardless of whether this explanation is exact, we are increasingly staring into the maw of a global depression, entailing food insecurity for the developing world. Yes, some people have said this for months now and nothing has happened; I myself thought that harm would be evident more quickly. Oil reserves and financial manipulations have postponed but not cancelled the pain; the government shorting oil futures has presumably ensured it will be worse in the end, because they have forestalled adaptation. But the window wherein they can continue to delay the pain is closing, and it is difficult to see how it can be extended into next year. A global depression in 2027 seems likely already, and more likely with each passing day, though there are a few months left before one can say it is entirely certain. Presumably this will be an inflationary depression; though if I ran the government and wanted to squeeze maximum value out of it, I might scare people of the reverse long enough to draw some advantage from selling cheap debt.

I am not sure quite how to end this entry. My circumstances are not good. Worse circumstances are coming. Everything I do may turn out to be useless because LLMs flood the zone, and is already fairly useless because today’s public has no interest in the truth about… anything, really. Somehow in that environment I am going to attempt to create a little light. I am hoping my poor recent progress was from muddled thinking as much as anything else, and writing this all out will set me straight. On the other hand, when I read back it does seem rather dire.

Well, that’s what I have. We’re obliged to strive onward and upward, whether for this life or the next.

July 28, 2026

I’ve finally archived the past year of entries. Still accessible by the link at the bottom of the page. Testing new formatting, but it will take me a few weeks to decide if I want to keep it or revert to the usual.

Read older journal entries here.