Live data from Hacker News

What AI did to stackoverflow in a graph

data.stackexchange.com

341–350 of 615 posts

Re: What AI did to stackoverflow in a graph

#341
post #11

Seing a bell curve and singling out a factor that appears only for the 15% of the total time demonstrates some pretty extreme tunnel vision Edit: https://postimg.cc/n9nZGLmb

Yes.

At the same time, this is graph is something that really should not look anything like a bell curve. So the format is probably just a coincidence.

Except if the "all the questions have been asked" hypothesis is correct. What I really doubt.

Re: What AI did to stackoverflow in a graph

#342
AI didn't kill SO, SO was already declining due to toxic users and lack of organization. The toxic users were dunning-krugerian anal-retentive mini-Napoleons who delighted in locking anything slightly off-topic and merging slightly-related questions, and the lack of organization was a tendency for people to give incorrect answers but still get upvoted (due to Google search surfacing the page in the wrong context).

Re: What AI did to stackoverflow in a graph

#343
post #217

Earlier quoted context omitted.

How good are the questions?

This is why people use AI so heavily. Because no matter how bad or ill formed a question is, it wo answer it the most "caring" way possible, and never berate you for it being a stupid question, common sense, etc (as long as it is about some forbidden topic of course)

On top of that, the AI will take a plain no-context error message and give it a college try to figure it out. A lot of times it will be right.

On SO that experience is going to be “we closed this because you didn’t form a good question.”

And of course, that’s true, but it demonstrates the wide gulf in user experience between the two platforms.

Re: What AI did to stackoverflow in a graph

#345
post #314

Earlier quoted context omitted.

How else do you help people learn if not by giving feedback? As an avid helper in a programming related discord channel I need you to understand that a lot of the people who come to these places for help are incredibly stupid. It's not uncommon to have to converse with a person for 15+ minutes just to get them to give you enough information to understand what their problem is. Like they will just paste an error messa…

You just described the complete opposite of the StackOverflow experience.

I disagree.

Here’s an example: my account on StackOveflow has enough reputation to answer questions on SO, but on other StackExchange sites that are very related I can’t do that just because I spent more time on StackOverflow.

The whole setup is basically repelling you from engaging by design. The site should already know from my SO reputation that I’m trustworthy enough to answer stuff on the other similar tech related stackexchange sites.

It was built for a time when you actually needed to filter out low quality questions and answers, but now that the users have abandoned the ecosystem the bouncer at the door makes a whole lot less sense.

Re: What AI did to stackoverflow in a graph

#346
post #334
post #313

Earlier quoted context omitted.

I remember having to fight a mod for him to restore a reply penned by Mike Pall, to a LuaJIT question. Mike Pall is the author of LuaJIT. The reply had been either deleted or edited to the point of being wrong (memory is foggy), because Mike Pall wasn't an expert at SO, and had somehow not used the site exactly as intended. The mod was very dismissive and patronizing.

I was going through some answers on a stackoverflow thread, and noticed that every single one had been edited by the same guy, just.. adding his own personal opinions and 'corrections' to them, and in the process making them universally worse and less correct The idea that answers should be editable, and the gamification of stackoverflow, was an absolutely terrible combination

Editing used to be fine, back then like 15-16y ago. Personally I posted mostly everything as community wiki (no reputation gain) as I got the mod-alike reputation anyways.

At some point the (say 2013-2014 or so) the site deteriorated quite massively, though - as folks considered stackoverflow CV worthy material...

Re: What AI did to stackoverflow in a graph

#347

Earlier quoted context omitted.

In the age of gratuitously encouraging LLMs and constant slop I sometimes miss the way in which stack exchange punished people for putting in low effort. I recently asked a LLM a question simply out of newfound habit. I realized while reading the line that this command seemed very familiar. I had bookmarked the exact same command in stack exchange many years ago. Of course the fact that stack exchange traffic has dra…

> Taking personal offense to someone insulting your question is something you should learn to get over. I did get over it when I stopped bothering with SO. Sorry, but a community whose ethos is "you need a thick skin to belong" is not one most people even want to belong to.

Yes. And this is exactly my point. The path of least resistance is often preferred. Maybe in most cases this makes sense. But it can also lead to one wading in slop.

Just beware an operating principle based on what "most people" want. Nearly all human societies that I am aware of bake in a caution to this impulse into their grounding philosophies. Because humans have always ultimately recognized that while seductive at first, taking guidance merely from "what most people want" is a path to decay.

Garbage in, garbage out. The inalienable right vs the mob. Three men make a tiger (三人成虎). Does the market always know best? Low quality, low effort questions are ultimately destructive to the ends of something like stackexchange. No need to apologize, they evidently didn't want you there either. Again, it's not personal. They may be gone tomorrow. Zoom out just a little bit farther and you will, in all likelihood, not be far behind.

Re: What AI did to stackoverflow in a graph

#348

I don't have the time to, but I'm surprised there aren't a lot of comments on the decline before chatGPT was released, but after SO was sold to Prosus [1][2] Even the curious growth spike in activity happened just before the acquisition. I wish I had time to do this analysis a bit deeper, but you can look for SO activity up until when chatGPT was released, it is really noticeable. --- [1] Stack Overflow acquired by P…

I don’t think we need to blame that sale. The same steady decline had been going since 2016. Itms clearly there in the graph, but everyone trumpeting the ChatGPT angle conveniently ignores the preceding 5+ years of continual decline.

Everyone trumpeting the "bad SO moderation" angle is delusional. Stack Overflow was far better than conventional forums. Without LLMs, SO would have stayed popular, there is no doubt about it. The massive decline was clearly and obviously caused mainly by LLMs.

Re: What AI did to stackoverflow in a graph

#349

Any social organization needs to carefully consider their inclusion-exclusion curve with intentionality. I think a lot of people might balk at the word "inclusivity" today, but StackExchange had ridiculously high barriers to participation, making it inclusive to the long-time users on the site, but exclusive to the newbie participants who found themselves blocked for asking questions. They slowly killed the site in t…

If it wasn't so controlled it'd might as well have just been reddit. I agree there's a balance, and maybe they edged over the line, but I was consistently happy to have the following be the outcomes 1. Answers were reasonably close to correct, usable, informative (teaching) 2. Your site score came to mean something -- I once had a hiring CTO say "Oh you have some popular answers on the techs we use" 3. Progressive un…

> If it wasn't so controlled it'd might as well have just been reddit.

And why do you think most people (and LLMs) just Google " reddit" ??

Re: What AI did to stackoverflow in a graph

#350

The graph actually peaked in 2014. That's a decade before AI became a thing while the Software engineering workforce grew a lot since then. I think Stackoverflow was the last incarnation of romanticized old-school Q&A forums where you first had to earn your badges before being treated with respect. Luckily today's new projects have much better documentation, issue tracker's etc. And apparently AI is able to work with…

> The graph actually peaked in 2014. That's a decade before AI became a thing while the Software engineering workforce grew a lot since then.

That's obviously because initially there were a lot of basic newbie questions to be asked, and those continued to be relevant in search results, so the amount of new questions went down over time, as there were fewer relevant unanswered questions left.

Post reply on HN