@skinnylatte Oh god does it make more sense if you realize everyone in charge is either someone who grew up on 1970s scifi, or is sucking up to someone who did?

@simon_brooke tbh what lots of environments allow is menial. But if people use coding tools they’re going to need to fix it for the tools at least.

@IngaLovinde @futurebird @maco @EricLawton @david_chisnall Yup. Not mostly for coding, where the demand is pretty direct, but ChatGPT is wildly popular itself _and_ wildly integrated even in the most ridiculous places. So even in the places it's hated, someone's inducing demand.

(Certainly not all programmers, but people proud of the craft and of doing a good job.)

@EricLawton @unlambda @maco @futurebird @david_chisnall i disagree with some strong asterisks. But I do think that making something you can really be proud of in detail is a bit uphill.

But also this enables a lot of time to spend on the planning and documenting side of things, which are absolutely things that programmers have wanted more time to do well, and processes that support. This might be providing them.

@EricLawton @stilescrisis This. I see it undermining knowledge in a _myriad_ of ways very quickly.

But also affixing us to 2025 technology because it will be good at known quantities more than novel things.

@unlambda @EricLawton @maco @futurebird @david_chisnall All of this. I remain with a position of "this is a net negative for society”. However the utility to a programmer is pretty hard to deny.

@futurebird @maco @EricLawton @david_chisnall Oof. A sales pitch embedded in it sounds miiiiiserable.

As far as pricing ... man it's hard to tell. The training of models is very expensive, and energy-consuming. That has to be amortized somehow. But the actual running takes only a little more than 'home computer' level. (and cruddier models do run on home computer scale things)

@maco @futurebird @EricLawton @david_chisnall That's complex and ever-changing, as business tuning is wont to do.

You usually pay for input and output tokens both, and thinking is part of that. But most people are using plans that give them some sort of semi-metered time-based access — five hours of time with the model and a token limit within that. It's a strange system.

Tokens are roughly in the lex/yacc sense, but they're a new thing, for LLM models. They're not precise parser tokens with parts of speech, but they are roughly "words”. Not exactly, since language is morphologically complex, and programming languages carry semantics in other granules, but the idea that they're words is not wrongheaded.

Others are going flat-fee (F/e, something like z.ai hosted GLM-4.7 is a flat fee per month, and quite low.)

(Also that one is interesting because cost to operate it figures are quite public. The model is public, the hardware requirements are about $15000, so you can do the math on it pretty easily to see what capital costs would be. Also environmental! Like that's 3 high end server GPUs, so a lot of heat, but also to humanize it, "less than heating a house" amounts of energy by far.)

@futurebird @EricLawton @david_chisnall @maco Yeah. And it really just is more and more precise force of the same sort. It does however end up at a qualitatively different place, with different impacts to the system of programming work itself because of it.

@futurebird @EricLawton @david_chisnall @maco Are they though? The only sensible way to evaluate it is as a system — nobody uses the raw LLM, it's always through layers of API, tokenization, and now models or at least separate "trains of thought" leveraged against each other to refine the output. Using the tooling to conform output is a good hack to keep the systems able to deal with new things by using new tools instead of needing new training.

And it's not exactly an extra check — it's embedded in a feedback loop.