staticshock

The main problem here is that real people operate in fuzzy domains. Snapping them into place "with code" won't magically resolve the gray areas inherent to the most valuable real workflows.

Think about the prized "high agency worker." What makes them desirable is the willingness and ability to make well informed, unilateral decisions on matters that are likely not yet organizationally codified, or codified in a way that is "wrong" for the task at hand.

Also, the reason terraform works is because it is _operational_. As in, it's actual code that runs. If it was mere documentation, it would drift like nobody's business. In order to make "organizational code" operational, you would need enforcement (a compliance team?) manually keeping the documentation in sync with reality in all of the meat and thought spaces where real work happens.

The only place where this can plausibly be automated is in digital spaces. In fact, I'm surprised the article doesn't go there: "organizational code" starts feeling way more plausible as definition for AI agents than for real people, specifically because agents are operationalized in digital spaces, where enforcement can be automated.

false; wolfram has been circling the topic of "small yet mighty" rule-based systems for decades, and this is his writing style. if you don't like the topic or the style, you are welcome to move on from it with whatever grace you can muster up.

"wildly misplaced trust" is a subcategory of "trust", which is exactly how i meant it

Instead, the approach that will continue increasing in dominance is hiring referrals and finding jobs through personal networks.

In a world that increasingly resembles The Library of Babel,

- the main way to know what's true is to tune into news sources you trust (monolithic old school media, or personality driven new-school media, social media, etc.),

- the main way to learn what to watch/listen/read is to take recommendations from people you trust, or received through channels you trust,

- the main way to hire or get hired is, increasingly, by exploiting a network of people you trust.

All of this compensates for ambient oversaturation by using the best available (and tunable!) desaturation filter: your trust network.

You're absolutely right!

Oh! I stand corrected.

Yes, the layoffs are a market correction initiated by non-AI factors, such as the end of the ZIRP era.

The world is chock-full of important, society-scale problems that have been out of reach because the economics have made them costly to work on and therefore risky to invest in. Lowering the cost of software development de-risks investment and increases the total pool of profitable (or potentially profitable) projects.

The companies that will work on those new problems are being conceived or born right now, and [collectively] they'll need lots of AI-native software devs.

Efficient markets route around bottlenecks. Technological revolutions accelerate the speed at which that re-routing happens.

In software, we, the developers, have increasingly been a bottleneck. The world needs WAY more software than we can economically provide, and at long last a technology has arrived that will help route around us for the benefit of humanity.

Here's an excellent Casey Handmer quote from a recent Dwarkesh episode:

> One way to think about the industrial revolutions is [...] what you're doing is you're finding some way of bypassing a constraint or bypassing a bottleneck. The bottleneck prior to what we call the Industrial Revolution was metabolism. How much oats can a human or a horse physically digest and then convert into useful mechanical output for their peasant overlord or whatever? Nowadays we would giggle to think that the amount of food we produce is meaningful in the context of the economic power of a particular country. Because 99% of the energy that we consume routes around our guts, through the gas tanks of our cars and through our aircraft and in our grids and stuff like that.

> Right now, the AI revolution is about routing around cognitive constraints, that in some ways writing, the printing press, computers, the Internet have already allowed us to do to some extent. A credit card is a good example of something that routes around a cognitive constraint of building a network of trust. It's a centralized trust.

It's a great episode, I recommend it: https://www.dwarkesh.com/p/casey-handmer

You can just ask ChatGPT what its training cut-off is, and it'll say June 2024.

It is likewise unreasonable to look down on any kind of world model from the past. Remember that you, in 2026, are benefitting from millions of aggregate improvements to a world model that you've absorbed passively through participation in society, and not through original thought. You have a different vantage point on many things as a result of the shoulders of giants you get to stand on.

I don't see a CLI tool in the Typescript repo.

i used to use last.fm with winamp and the like. that needed scrobbling plugins. nowadays, i use it with spotify, and it's pretty simple: (1) make an account on last.fm. (2) go into spotify settings → social → connected apps, and add it in.

Direct quote from RJ Scaringe, founder/CEO of Rivian:

> This is a decision. It's generated, I said there's many millions of decisions, many of them will never get noticed and they're just under the surface. One of those decisions that's been noticed quite a bit is the fact that we've intentionally not included CarPlay in the vehicle. And that's not to say we don't think a close partnership with Apple is important. So we have Apple Music integration, we have a bunch of Apple integrations that are yet to come, we have a great relationship with the team at Apple. But it was more to say, we just felt and continue to feel very strongly about creating a consistent, fully integrated digital experience where you're not jumping between apps, let's say from a CarPlay app back to the vehicle app. And it's quite jarring when you don't have, let's say vehicle level controls when you're in the CarPlay environment. That view we've had since the early, early days. I think that's going to become even more important and more true in a world of integrated AI.

https://cheekypint.transistor.fm/14/transcript

economic value is distinct from social value. the former can be used to describe small groups (e.g. criminals who benefit from a crime), while the latter can be used to describe the negative sum relationship between criminals and their victims.

so, "positive [economic] value [for some]" and "negative [social] value on net" are not mutually exclusive.

Unfortunately we bozos just aren't that imaginative :'(

This seems obviously wrong? Any system whose name includes the word "forecast" was built to predict the future in some domain / over some time horizon / to some level of granularity.

Have you ever tried to submit a PR of your Makefile to the maintainers of, say, an open source npm package? If you have, what compelling story were you able to tell about that Makefile that got the maintainers to merge your PR? And after they successfully merged it, did you continue putting up PR after PR for each successive round of tweaks to that Makefile?

What an extraordinary amount of unnecessary effort that would be. My workflow does not belong to the repo, it belongs to me. The only thing that belongs in the repo is all the shared workflows (or the elements they comprise.)

nope, this is Mira Murati's thinking machines. she used to be the CTO of OpenAI, but started her own thing once she [I think] realized that she was missing out on the gold rush by staying there.

I do this and like it: https://github.com/staticshock/dotfiles/blob/main/Makefile

I also tend to put a Makefile into the root of any repo I work in (ignored via .git/info/exclude), so that shell commands relevant to my workflow can accumulate under that Makefile. I've been doing this for at least 10 years and love it. Most repos have some sort of a cli interface, but that doesn't mean that they're any good. Make is nice way to wrap all of them, and to standardize on at least the build/test/run commands across many disparate repos.

Here's an example of one of those from a long-abandoned project: https://gist.github.com/staticshock/0e88a3232038d14a2817e105...

Arc's fluid transition between tabs, bookmarks, and favorites was excellent UX. It finally helped me tame 100s of open tabs by helping me treat "apps" as distinct from "pages".

Every new browser should be doing something to recognize the fact that browsers are an application platform, and they should be absorbing more desktop OS features in the direction of native-feeling applications.

One of the things I want from a browser is to fade to the background and to expose those applications as natively as possible at the OS level (e.g. I want to be able to put browser-based applications into my dock without having to download special electron wrappers for them.)

Some churn is fads, but some is legitimate (e.g. "we know how to do this better now".) Every living system is bound to churn, and that's a good thing, because it means we're learning how to do things better. I'm happy to have rust and typescript, for instance, even though they represent some amount of churn for c and javascript.

The spirit of this post is great. There are real lessons here that the industry will struggle to absorb until we reach the next stage of the AI hype cycle (the trough of disillusionment.)

However, I could not help but get caught up on this totally bonkers statement, which detracted from the point of the article:

> Also, innovation and problem solving? Basically the same thing. If you get good at problem solving, propagating learning, and integrating that learning into the collective knowledge of the group, then the infamous Innovator’s Dilemma disappears.

This is a fundamental misunderstanding of what the innovator's dilemma is about. It's not about the ability to be creative and solve problems, it is about organizational incentives. Over time, an incumbent player can become increasingly disincentivized from undercutting mature revenue streams. They struggle to diversify away from large, established, possibly dying markets in favor of smaller, unproven ones. This happens due to a defensive posture.

To quote Upton Sinclair, "it is difficult to get a man to understand something when his salary depends upon his not understanding it." There are lots of examples of this in the wild. One famous one that comes to mind is AT&T Bell Labs' invention of magnetic recording & answering machines that AT&T shelved for decades because they worried that if people had answering machines, they wouldn't need to call each other quite so often. That is, they successfully invented lots of things, but the parent organization sat on those inventions as long as humanly possible.

i tend to draft everything lowercase, and then go back and uppercase things depending on how much of a formal vibe i'm going for. capitalization rarely helps me formulate an idea, and so my writing often splits into phases: (1) formulation, (2) polish.

also, it's worth noting that proper capitalization does not automatically yield text worth reading. from that perspective, i like lower case text as a form of rebellion against the artifice of rules; any rebellion against particular aesthetics is fair game in my book. more generally, i'm skeptical of process advocacy in cases where the process seems to be done for its own sake.

on the flip side, good grammar helps me parse sentences, so i do sympathize with arguments in its favor.

I'd love a version of this where I can paste my full list of extensions, instead of a box where I can only paste one. The latter is tedious, so I'm not that likely to do it.

I think this is best seen as an art project, not as a business idea. I'm also assuming that the cost to run such an app are near zero, so the break-even cost would probably be, like, one person a month generating one new thing.

Confusingly, the about page suggests that this is an AI-driven storytelling app: https://www.thiings.co/about. Maybe that's a vestige of prior iteration of the app, though.

The concept minds me of the million dollar homepage, where you could pay a dollar per pixel of advertisement. Here, you're paying a dollar (that's a guess; i didn't check) to get a new object listed in the potentially infinite grid, which gives you a unique URL for that object, an emoji-like reference image, and a one-paragraph description for aliens, were they to land on earth and ask what that object was for.

Basically, looks like an art project that monetizes participation.

In addition to having the money, Google also needs the incentive to spend that money on such projects. If the perceived return on capital is low (or negative!), the incentive is simply not there.

Because it got voted up by people who found it interesting

As a rule of thumb, when the pause between two adjacent sentences implied by the period seems a bit too halting, try a semicolon.

I find it easier to think about the utility of semicolons from the perspective of individual sentences to be combined. And therefore, if a properly used semicolon were to be replaced with a period, the two resulting sentences should still make sense in isolation, they'd just feel ever so slightly too far apart.

"Semicolons bring the drama. That's why I love them." Does it work? Yes. Is there a clear through-line from one sentence to the next? Yes. Does the pause between the sentences feel a bit too abrupt, though? Maybe!

("Establish clear through-line" can be another nice reason to combine sentences with a semicolon.)

Ignoring for the moment the fact that this particular article is a book summary, a task LLMs excel at, it's interesting that the (warranted) comparison to LLM outputs casts doubt on the credibility of the writing, and in some ways maybe even caps its implied utility.

Just like how human beings choose different ways of presenting themselves to the world (e.g. masculine/feminine, gay/straight, goth/punk/preppy) as a form of social signaling, today's LLMs emit a certain "I'm AI" signal that humans pick up on, and human writers will likely have to continue evolving the counter-position(s) to that signal.

If the results of relatively simple, unsophisticated prompts get better at passing for human-written articles/blog posts/forum comments/etc, that'll increase the fraction of human writing that falls into this uncanny valley, and exacerbate the need for stronger counter-positioning over time.