Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Let us git rid of it, angry GitHub users say of forced Copilot features

theregister.com

Post reply on HN
Print view View original

221–230 of 315 posts

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#221

Post by ants_everywhere »

globular-toast wrote: post #184
acomjean wrote: post #38

Earlier quoted context omitted.

If I remember initially GitHub (before MS) was free for open source and pay for everyone else. It wasn’t an entirely new idea (source forge?) but it use git which was rising in popularity. I think GitHub added the “pull request” as a really useful add on to git and that really made it take off. Oddly I used selfhosted git at an academic institution. I liked it because it was set up to use “hooks” https://git-scm.com/…

> I think GitHub added the “pull request” as a really useful add on to git and that really made it take off. Negative. The only thing GitHub added to the parlance is "forks" which are essentially like namespaced branches in the same repo.

The use of "fork" predates GitHub as well https://en.wikipedia.org/wiki/Fork_(software_development)

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#222

Post by crazygringo »

dcminter wrote: post #166
crazygringo wrote: post #150

Earlier quoted context omitted.

> I've never seen that with useful tech tools before I've seen it all the time. Version control, code review, unit testing, all of these are top-down. Tech tools like git instead of CVS and Subversion, or Node instead of Java, may be bottom-up. But practices are very much top-down, so I see AI fitting the pattern very well here. It feels very similar to code review in terms of the degree to which it changes developer…

Nope, all of those things were dev driven until they'd diffused out as far as management and only then did they start getting enforced top-down. Often in awful enterprise software ways actually.

But that's what I'm saying.

Obviously developers invented these things and initially diffused the knowledge.

But you're agreeing with me that they then got enforced top-down. Just like AI. AI isn't new or different like this. Developers started using LLM's for coding, it "diffused" so management became aware, and then it becomes enforced.

There's a top-down mandate to use version control or unit testing or code review or LLM's. Despite plenty of developers initially hating unit tests. Initially hating code review. These things are all standard now, but weren't for a long time.

In contrast to things like "use git not Subversion" where management doesn't care, they just want you to use a version control.

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#223

Post by cshores »

tempodox wrote: post #180
goku12 wrote: post #164

Earlier quoted context omitted.

The environmental damage is even worse than the investment. Those who have the money to invest in it usually care only about the returns and not the environment.

Hardly anyone even mentions it in discussions. Even as far as these tools are actually useful, nobody ever asks whether it’s worth the environmental costs.

For Google's Gemini LLM, the energy impact is negligible, with the average prompt consuming the equivalent energy of just three seconds of a microwave's operation.

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#224

Post by grayhatter »

marginalia_nu wrote: post #211
grayhatter wrote: post #198

Earlier quoted context omitted.

because quality has never been the driver of survival. It has always survival. The race is a race of endurance, it's not a competition on merits. Humans would undoubtedly be better without scars, but the same anomaly that gave us scar tissue, gave us faster wound healing. Increasing survival. But scars don't go away, when it'd be better if they did. There's no selection pressure there because it's good enough. IBM st…

This seems like an orthogonal concern. I don't see why quality and correctness would anti-correlate with survivability. I'll ask the same question again, out of all the highly survivable businesses, why are so many seemingly dysfunctional.

That orthogonalality is my exact point. I believe you're correct; quality and correctness aren't negative pressures to survival. If anything, the should support survival, and I'd assume should also have a slight positive pressure on adoption/growth.

But I'd hope you'd admit quality and correctness aren't free attributes? They do have a cost. I can churn out low quality code way faster than I can produce code I'm proud of. If I attach myself to the quality of the code, and get stumped by some bug, become frustrated, and take a break from project_a, to work on something else, and while working on project_b to "clear my mind", I fall in love with project_b, or it gets more popular, or whatever that "pressure" happens to be... project_a has no remaining developers, and it is still dead now. Thus, quality has had a negative impact on it's survival.

Suitability has a tenuous connection and dependence on quality and correctness. (which I believe are synonyms for the same core idea?)

But why are so many businesses (the ones that still survive) so demoralizingly dysfunctional? Because they're run by individuals who don't value quality and correctness above [other attribute]. When given the choice to increase money (which is effectively the exact same thing as market share, and when talking about survival popularity is the same thing as suitability), or increase quality. They will always make the decision that ensures their survival, (by chance, no by intent, that's the orthogonality). Eventually, they'll turn that knob too far, degrade their quality enough and create an ecological niche for someone else to take over. (A competitor that maybe they acquire before it causes a real risk to it's survival/popularity, again choosing to make money/survive, over a decision targeting quality)

Would *you* rather make money, or write something high quality? I use and love marginalia, so I think I can guess the answer. (Thank you so much for building something that actually meaningfully improves the internet btw!) Are there decisions you could make that would trade the quality to become more popular, or make more money? Yes, I'm sure, but you don't seem to be trying to become the next google.

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#225

Post by bink »

nlawalker wrote: post #197
dcminter wrote: post #131

Earlier quoted context omitted.

It's an excellent point - but a lot of the pressure to use AI in orgs is top-down and I've never seen that with useful tech tools before; they always percolated outward from the more adventurous developers. This makes me wary of the AI enthusiasm, even though I acknowledge that there is some genuine value here.

I think it's coming from both places, it's just that the top-down exhortations are so loud and insistent. I wasn't around to experience it but my understanding is that this is what happened in the 90's with object oriented programming - it was a legitimately useful idea that had some real grassroots traction and good uses, but it got sold to non-technical leadership as a silver bullet for productivity through reuse.…

I think that's a good comparison. I was around in the 90s and I do remember OOP being pushed by all sorts of people who weren't coders. It was being pushed as the "proper" way to code regardless of the language, size, platform, or purpose of the program in question.

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#226

Post by ants_everywhere »

stavros wrote: post #98
layer8 wrote: post #25

Earlier quoted context omitted.

While Microsoft is certainly to blame, GP is also right that the problem wouldn’t exist if people hadn’t continued en masse to have their code hosted on a centralized proprietary and (since 2012) VC-funded platform in the first place. As an aside, I don’t really see GitHub as a whole as a community. It’s a go-to place with network effects, but network effects doesn’t by itself imply “community”.

People will respond to incentives. They had an incentive to host their code in a place that easily let them do things that were extremely high friction before. People aren't morally reprehensible because they prefer convenience over hardship. People like using easy things, and they like making money. This means that people will make easy things so other people will give them money. If you don't like it, make easy thi…

> People will respond to incentives....People aren't morally reprehensible because they prefer convenience over hardship

To clarify my point isn't that anyone is morally reprehensible. My point is that using a free VC-backed service is like selling an implied option. You don't know when they're going to invoke the option, but eventually they will. And often it will be when you've gotten used to the income from selling the option.

It's not a question of morality or judgment, it's just meant to be a description of what the game we're playing is.

> If you don't like it, make easy things that work the way you like them, run them ethically, and don't sell them to anyone.

I'm trying to

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#227

Post by bink »

karlshea wrote: post #145
Rover222 wrote: post #97

Earlier quoted context omitted.

Not really your main point, but Tesla self driving is quite incredible, despite what internet clickbait says. They have a clear path to full autonomy with vision-only systems. But yeah, certainly 5-7 years behind the initial schedule. Which I guess was more of your point.

You’re still falling for it. They have a clear path to vision-only autonomy IN THE BAY AREA. Let’s see it work in Minnesota in the winter where you can’t see lane markings, everything is white, and the camera lenses immediately get covered with road salt spray.

Heck, I'm concerned how they're going to work in the Bay Area after an earthquake or cell network outage.

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#228

Post by stavros »

ants_everywhere wrote: post #226
stavros wrote: post #98

Earlier quoted context omitted.

People will respond to incentives. They had an incentive to host their code in a place that easily let them do things that were extremely high friction before. People aren't morally reprehensible because they prefer convenience over hardship. People like using easy things, and they like making money. This means that people will make easy things so other people will give them money. If you don't like it, make easy thi…

> People will respond to incentives....People aren't morally reprehensible because they prefer convenience over hardship To clarify my point isn't that anyone is morally reprehensible. My point is that using a free VC-backed service is like selling an implied option. You don't know when they're going to invoke the option, but eventually they will. And often it will be when you've gotten used to the income from sellin…

Sure, but nobody knows this. Everyone just thinks that things will be as they are now. I don't think the average developer knows what enshittification is, but Doctorow really nailed that one.

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#229

Post by cshores »

j45 wrote: post #158
AlexandrB wrote: post #44

I find it weird how companies talk out both sides of their mouth on AI. On the one hand it's this magical tool that will make you 10x more efficient at your job, and on the other it's something they have to market heavily and shove in your face at every turn - sometimes outright forcing you to engage with it. These two things don't seem compatible - if the tool was that good people would be beating down their doors t…

Just like someone who can figure out how to write code to solve a problem can do it while other programmers say it's not possible and just do it some other way, the same is true of AI. One of the major issues I'm seeing is how much technical people haven't been involved in the application of AI, which leaves non-technical people to pontificate and try. With any new tech, after the hype is gone, what remains, is adopt…

I generally agree, but I think there is a real disconnect. Middle and upper management often do not understand how developers and engineers are actually supposed to use these tools.

For example, I work in operations, so most of what I touch is bash, Ansible, Terraform, GitHub workflows and actions, and some Python. Recently, our development team demonstrated a proposed strategy to use GitHub Copilot: assign it a JIRA ticket, let it generate code within our repos, and then have it automatically come back with a pull request.

That approach makes sense if you are building web or client-side applications. My team, however, focuses on infrastructure configuration code. It is software in the sense that we are managing discrete components that interact, but not in a way where you can simply hand off a task, run tests, and expect a PR to appear.

Large language models are more like asking a genie. Even if you give perfectly clear instructions, the result is often not exactly what you wanted. That is why I use Copilot and Gemini Code Assist in VS Code as assistive tools. I guide them step by step, and when they go off track, I can nudge them back in the right direction.

To me, that highlights the gap between management’s expectations and the reality of how these tools actually work in practice.

Re: Let us git rid of it, angry GitHub users say of forced Copilot features

#230

Post by watwut »

crazygringo wrote: post #222
dcminter wrote: post #166

Earlier quoted context omitted.

Nope, all of those things were dev driven until they'd diffused out as far as management and only then did they start getting enforced top-down. Often in awful enterprise software ways actually.

But that's what I'm saying. Obviously developers invented these things and initially diffused the knowledge. But you're agreeing with me that they then got enforced top-down. Just like AI. AI isn't new or different like this. Developers started using LLM's for coding, it "diffused" so management became aware, and then it becomes enforced. There's a top-down mandate to use version control or unit testing or code revie…

No, not just like AI. The difference is that these things were pushed by people on the bottom for years and run successfully before management top caught up. Like, years and years.

AI does not have such curve. It is top down, from the start.

Post reply on HN