This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
Trillions of dollars are being spent on building datacenters for inference. Amazon software engineers are inventing bullshit work for AI to inflate their internal usage scores.
I’m no expert, but isn’t there a fatal flaw here? Most of the work LLM inference is used for is essentially busywork that wouldn’t exist in an automated economy. It’s writing emails, it’s code reviews, it’s asking dumb questions, it’s transcribing or summarizing research or zoom meetings. Even in software engineering, a lot of LLM tokens are used in the kind of inference that a hypercompetent solo-coding model with limited or no human oversight just wouldn’t need.
Think of an office with 10 human employees working in, say, payroll, constantly sending each other emails, messages, having meetings, calling and speaking to each other and other people, summarizing documents, liaising with other departments, asking AI question about how to use various accounting tools, or about the company’s employee benefits package. Now say this department is automated. An AI model acts as an agent to use an already-existing software package to do all the payroll work. No emails, calls or meetings - or at least far fewer. The total inference work required goes down. And the existing software package doesn’t use AI (even if it may have been coded with it), because you don’t need AI to compute payroll data once you have sufficiently complex and customized software for your business.
In the same way, if we imagine our automated future, super high intensity / high token usage inference is actually not really universally required in a lot of occupations. It will be for some multimodal work (plumbing, surgery, domestic cleaning in complex physical environments), but for many tasks, one-and-done software coded either by AI or that already exists can just be deployed at low intensity by an agent. The AI that replaces your job might at first do a lot of coding, but as time goes on, the amount of novel inference required will diminish. Eventually, software coded in a one-and-done way by the AI may actually handle almost all the workload, and token usage for generation may be very limited to just some high level agent occasionally relaying instructions or performing oversight.
In this scenario, why would we expect inference workloads to shoot up so dramatically? Much enterprise AI usage is currently “fake” in the sense that it would not be performed in a fully automated environment. It’s a between-times thing.
It's the same at the company I work for. The board member in charge has introduced AI usage KPIs, and now everyone is using LLMs for random shit. The new KPI is that 5% of all new code must be written by an LLM. Which is achieved by running a post-commit hook that is actually quite clever. There are lots of tools out there that are used to detect LLM-produced writing or code. Well, the same tools can be used to reject code that is too human if you flip the final check!
An acquaintance of mine got slapped with the same thing recently. Management has since walked it back because it caused an avalanche of technical debt, but at no point did they ever explain why that kpi was instituted in the first place. Did you get any kind of explanation of the goal they're trying to hit?
Broadly I imagine it was:
I have some sympathy for this perspective, having seen two very skilled devs just become fundamentally obsolete and impossible to work with because they refused to give up using tools from twenty/thirty years ago.
That isn't true at all in my experience. But it is true that more senior devs are less impressed by "new and shiny", instead being very critical about "what problems does this solve better than my current tools do".
One of the things that annoys me about mandating LLMs is that, generally speaking, you have to hold tech guys back from adopting new stuff. They are notorious for going all in on things which have issues for the company (security and compliance flaws, etc) and have to get walked back. They will even set up shadow IT departments just to get stuff done better. If LLMs are truly as useful as the hype says, there's no need to mandate using them. The people for whom they solve problems will trip over themselves to try to use them.
It really, really depends, especially since I’m talking senior in age terms ie 50+. But we eventually had to fire a 3d artist for refusing point blank to learn the industry-standard tool that the rest of the team was using, for example. He was perfectly fast on what he had but it didn’t scale and he couldn’t work with the rest of the team.
That aside, a lot of the programmers I worked with considered themselves gurus, and were very invested in the practices that they had been taught at their expensive computer science degree. They were legitimately good at what they did but they clearly considered LLMs an inferior replacement for their skills. The kind of people who insist on VIM over an IDE and will argue for days about whether Python private functions should be prefixed with underscore.
Tl;dr: a lot of programmers are genuinely in a rut, and a lot of others are more interested in writing beautiful code than solving problems.
My people! If the world loses them, it will be poorer for it.
I am not a professional developer. But in my IT experience, it’s helpful to have a mix of wary skeptics and early adopters for many kinds of technology. Strongarming your skeptics before you have to is a mistake. And while I believe much that I’ve heard about the benefits of coding agents, nobody knows what the final picture will look like. The grumpy old fogeys aren’t prophets, and I am not saying that they are, but they come by their battle scars honestly.
Granted! And I'm been the wary skeptic on a lot of things. In this case, given the unique and potentially transformative nature of the technology, I have a certain amount of sympathy for the managers who decided that the greybeards needed some experience with AI products so that they can judge from a position of knowledge not prejudice. Ideally that should employ the carrot rather than the pointy stick, but occasionally you still need the pointy stick.
I also coerce my interns into using it (and pay for the subscriptions myself). In that case it's more for knowledge lookup more than code, because in my opinion getting used to having a personal tutor permanently on call is the best gift I can give them.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link