Don't play to the sci-fi "this thing's trying to outsmart me" tropes.

But in practice I never saw anyone crack the embedding-generation-and-comparison problems well enough to actually get better results than grep for things like "find similar code and see what it does."

(You also don't need that advanced a model to use "grep over a pile of files", but the models today can run MUCH faster than GPT 3.5/4 were running over the APIs back then, making "summarize all five hundred of these matches from those files" much more usable.)

So many claims about how it would lead to far better lives for everyone, but the working conditions and general affordability have basically gone down for 40 years. Imagine bringing back the white collar work in the 80s, with a private office with a door, and people whose jobs were to help coordinate and schedule things even if you weren't an exec, instead of you just having a phone to answer all hours of the day.

One of the greatest tricks of the modern era in the US has been to convince everyone that making the slice of pie bigger for the richest people is necessary to grow the economy.

Are you saying you can't think of examples from the past of the introduction of technology changing labor dynamics or organization? Say, mechanical agriculture?

The changes are hardly going to always be good - there's no determinism of "new technology means society will get 'better'". But they've often been periods of change, and such periods are when it's easier to influence the direction if you happen to care about the direction of the change.

How so? Without technology we wouldn't have a "work week" in the first place and work would be much more directly tied to survival of the community and generally less negotiable in the first place. The "flipping out" came about precisely because technology changed what work was and what the conditions around it were while people noticed that those new expectations and conditions didn't actually seem necessary for their survival (or even much to their personal direct benefit vs the benefit of business owners).

Any technology that lets more be done with less time is an opportunity for a population to make an attempt to claim some of those gains for themselves.

"Salaried" vs hourly is increasingly a scam anyway, but all that benefits stuff is something that would have to evolve. And it could, if people find the political will.

This ignores a lot of historical fighting (sometimes literally!) to get it down to 5 in the first place.

If everyone sufficiently "flips out" about it being 5 then the problem of "reduce the context switching problem" is something the owner can try to figure out.

Cause otherwise, you could find a perfect solution to that problem, and still not have leverage to make ownership actually change anything vs just raise expectations that much higher.

(Meanwhile, some companies are trying to import 996 and push it past 5 for white-collar work anyway, so any sort of non-political, non-disruptive action seems doomed to fail since the status quo is moving the wrong direction.)

Running software in the cloud gives you certain reliability and scaling advantages that would be very hard to replicate locally. Running some code agents in the cloud vs local hardware, if the local hardware gets "good enough," breaks the other way - offline usage, alone, would be hugely valuable to many people and companies.

It'd be very interesting to see where various players would decide to make a call "local is good enough" though. Buying the hardware isn't a small bet, if it's not something that ends up as part of your standard computer.

Using Cursor to hop between models, I've found Opus to be generally better at really tricky debugging than GPT 5.5 or earlier models, but not reliably better at execution because of these things. I'm not sure Composer 2.5 is quite there yet for the execution side, but it's getting pretty close to those other ones, such that I'm definitely still in a "debug and plan with slow, execute with faster ones" operating model for working on hard shit.

4-7x isn't a tiny markup, but how does that compare to high-margin internet businesses like AdSense? Meta and Google do hundreds of billions in ad revenue a year, and after taking out the publisher's portion (60-80% per some searching), I wonder what the ratio of the remaining tens-of-billions is against the compute cost and headcount required to run it.

And how much room for maintaining or improving that margin do they have if the cheap competitors also continue getting better? Is there a "good enough" point where the easier inference tasks are all moving to vendors massively undercutting them, and then they don't have the volume necessary to justify spending on further cutting-edge development?

1) submitted changes that don't need any more revision than their previous human-written ones when it comes to code review?

2) no increase in bug incidents

3) no slow-down in peer work or future work caused by humans-or-agents having to fight increasingly overly-complicated, poorly-factored, copypasta-style code or god methods? (this might not be evident yet)

(Another question is how well is this person doing their job as a reviewer, making sure to keep the product quality bar high, without looking at code?)

Anyone in an org with coworkers no longer writing code needs to be making sure their managers have a pulse on the long-term health of the product to see who's doing it well (lots of test coverage, shipping only super-high-quality, refined-from-multiple angles stuff) or just being lazy (shipping first drafts that continually add debt to various files and methods).

Rust had a "vibey" community long before vibecoding. In particular, it's long been fairly non-serious about yolo importing a bunch of crates to solve things (since the standard lib is small) which is kinda the same problem as having all those things just vibecoded. Either way, most projects weren't reading all of that other code!

However, I don't think the rest of the post supports the Pratchett-as-expert-on-memories bit of that little aside, so it could be critiqued for not being followed up on or further supported. It's not a "make an argument for Pratchett as an expert on memory" post, it's a "tell a story of his writing connecting with me" post, in which case, meh.

But apparently the post was edited again, and "The Author, refusing to let the Narrator off the hook" is gone (along with some other things), which is a bummer as it seems like one of the most Pratchett-like of the attempts. In context I don't think it works, but generally it fits well with Pratchett's writing style and his relationship with his protagonists.

I haven't seen "just absorb a giant ball of context and do the right thing the first time" be cracked yet, even for Opus 4.7.

At the end of the day, code is code, and we have decades of lessons about how to make code more reliable and maintainable. Composable small modules, not god methods, are still the way to go, and they reward devs who use them to get focused context for agents with faster - and often better - results.

BUT.

"Ignore anything but actual problems" is a terrible stance to take generally for software and dependency selection. Incidents are fairly sparse, process is much easier to observe. So if you can find connections between process and incident possibility, that's a very reasonable heuristic. And it's easy to find examples of overaggressive LLM usage introducing problems into software.

Doing this particular one is definitely expecting the market squeeze to continue. "Worst case" is back to more "normal" depreciation. Where I'd expect to only be able to recoup more like 18k. But... if you look at GPU prices the last 3 years... it's not a crazy assumption that it won't drop that fast.

iPhone example since those are easiest to find in quantity: new iPhone 16 Pro Max for $1200, Gazelle would want $866 for "execllent" condition. Lost ~28% for one-model-back. iPhone 15 Pro Max, though: excellent priced at $667 here, only down another 23%, and gives you basically half-priced-upgrade if you can sell it for that and roll into the newest.

So to have never-more-than-one-model-old rough estimate at today's value-holding you'd be out $3600 for three new phones, with getting 1732 of that back, or 1868 for it (with a $334-per-year incremental cost of upgrade).

For never-more-than-two-models-back you'd be out $2400, getting back $866, for net $1534 spend, with a $167 incremental per-year upgrade cost once you buy the first one. Pretty good if you keep the phone in excellent condition and are happy to budget a bit over $10/month to be on a every-two-year upgrade train.

Well, you'd also eat the tax...

https://buy.gazelle.com/products/iphone-16-pro-max-256gb-unl...

https://buy.gazelle.com/products/iphone-15-pro-max-256gb-unl...