my reputation is 29,459, my last answer October 15, 2021, close enough to be considered the same IMO.

For me, the thing that drove me away, is that you'd spend time on writing good answers, to provider lore, etc, but it was increasingly difficult to find "good" questions to answer. There were increasingly many "homework" questions, and far fewer questions where people _had_ put the effort in.

And, an increasing animosity to the people answering the questions.

The community became toxic, and it wasn't the only the gatekeepers, everyone wanted something for free, and StackOverflow suffered from the "open source burn-out" problem.

Certainly I did.

Whenever I took the time to make a long answer on a question, I saw some mod close the question as a duplicate of an existing one - even though the question could have some minute detail that was different, or perhaps there was some recent change in the landscape that meant an answer in 2020 would be different from one in 2013. But no: aggressive marking as duplicate even though clearly as a _community_ it would be much better to have a more recent discussion, perhaps more alternative answers and just allowing the near-duplicates.

This has very interesting implications for everything in the world, btw. Correctness does not track majority consensus.

Stuff like this is basically what drove me away as well. The top-rated answer was frequently not the answer. It was "an" answer to "a" question, but not the "the" answer to "the" question.

That, and I'd ask a variation of a question and it would get flagged as a duplicate of an already-answered question, except the original question and none of the scored answers contained the answer.

I still left eventually and LLM has nothing to do with it. It’s the “closing as duplicate”, it’s the “provide text explanation not just code”, it’s the “deleted as off topic”.

I can’t tell you how many times I was finding useful info in DELETED answers (I see them as moderator). And the existing “vote for undelete/reopen” procedure does not work.

SO did not die because of LLLMs. It died because it was driven to the ground then sold private equity and about to be sold again (mark my words)

On topic: thanks for this great summary of the issue. I too miss the prime stackoverflow days. Although the availability of LLMs has allowed me to learn new things and generate some code again, the passion and love for programming that was on (the early) stackoverflow, I do miss. Also the times of just buying a book to learn about a new tool or language (jQuery in Action teached me a lot) feels different from just asking Claude Code to fix or explain something.

Even so, at some point I made sure to answer a few questions, to get enough reputation to upvote and occasionally comment. I wanted at least to be able to reward / thank those who had posted answers that I found helpful. Even for 5 or 10 year-old answers, if an answer helped me I'd upvote it.

Now we're all "taking" via AI. We don't know who posted a helpful answer 10 years ago that the AI learned from in order to help us. That saddens me.

But in a way that's just the way of the world. We know to credit Pythagorus and Fourier and a few others whose names are attached to their methods. But for the most part we're all constantly standing on the shoulders of anonymous giants whose names are long lost.

IMO the uselessness of the answers just became more and more apparent, so fewer people contributed, so they rotted further. Good riddance, I say - it had a brief couple years where it was excellent because all information was new enough to be useful, and then it started dying like all stagnant things do.

Stack Exchanges that do not have such a bias for time-sensitive information are still going relatively well AFAICT, e.g. math exchange. The model works fine for some things, just not general tech assistance.

The query in this thread further confirms it, by november 2022 (ChatGPT reveal), usage had already dropped to ~109120 questions/month as compared to peak usage in March 2014 at 207169 questions/month.

The trend continued since but their decline is surely not caused by LLM:s.

Had similar thing happen to my colleague too, in both cases they were pretty deep technical questions

Stopped using that ego bubble blower page shortly after

Oh let me remind you, it was the expertsexchange that was actually useful!

Or my favourite entire IP ranges or whole ISPs blocked because eighteen years ago someone used a sock puppet account from their and there is no recourse because the High Doge has decreed it from their throne in Moderator Palazzo - you can appeal but the lord chief's decision is FINAL, and they're all too busy infighting and empire building to care about such trivia as users anyway

Hello, blast from the past! Feels like a lifetime ago.

I'm so grateful for WHATWG listening to the jquery community though, sizzle.js and the CSS query selector API made building dynamic websites so much easier.

Hearing this story is helpful.

There was a lot of blocking and downvoting of questions which was both a good and bad thing.

If you had an expertise on a certain topic, the questions were often very simple and repetitious (e.g. I knew a decent amount at relational databases), but also you'd get people asking for whole system designs which got a bit ridiculous.

It wasn't great for asking about little known products or libraries etc and I often found no one responded to questions on certain js libraries or BI tools. Which was quite annoying, maybe because it appealed more to certain types of developers.

I guess you served your time and most people don't want to be answering questions every day for ever! There was a guy called Tom Kyte from Oracle who ran a question site for over 20 years - it must have taken a lot of his time, but if you're out there Tom it was very much appreciated!

Culture or no culture LLMs killed SO for good, there's simply no need for it anymore

That happened often enough that I just gave up. I rarely felt the need to ask questions cause I was pretty good at what I did so I quit altogether. For fun, I signed in and noticed about a 2000 uptick in my standing and three "best answer" awards.

For me, I think I left for a few reasons. All of them are some flavor of "the design is user-hostile".

The first reason is the "traps" they put in the moderation review queue process. These always felt like shit, even if you were right. And for some reason they frequently picked questions where you could reasonably pick multiple reasons to close a question... and they'd yell at you for getting it wrong! Really, this needlessly punitive experience is the heart of why SO was so awful, because everything is like that across the whole site. Some of the network sites aren't like that, but it's rare. And there were just so many Meta questions where the upvoted response was "this would be hard to do right, so instead we're going to do it the way we all know is wrong but it's easy."

The second reason is just how awful the onboarding is. The site constantly had problems with people either not posting code or posting screenshots of code. I don't know why they didn't make a beginner form, similar to what you often see on GitHub issue trackers for large projects (i.e., always include this, always include this, include this if it's relevant, etc. and it's in the textarea box when you start a question). Instead they assume that you know how to ask a question, give you a completely blank form, and then aggressively policy anything where you didn't do it right. I think the one thing to come out of that site is that people do not inherently know how to ask questions. [Which, if anybody has been responsible for responding to IT tickets, oh, boy, do they not know how to ask questions or report problems. "I'm getting errors running my report in the Finance application. I need this report soon! Help!" No mention of which report. No mention of what the report parameters were. No mention of which error message. They don't know what the error message means so, obviously, it must not be useful to anybody.]

The third reason is the site finger-pointing. "This doesn't belong here so I'm locking and closing your question," is possibly the most ridiculous thing ever. Why doesn't it move the question to the other site and then make a stub link? More awful, user-hostile experience.

But, I think the final nail in the coffin is just how aggressively awful rpg.stackexchange.com is, which seeks objective, hyperliteral answers to game questions that are, by design, subjective and intentionally meant to be played with interpretation and nuance. The result is a handful of highly active users will nit pick any answer or question they don't like. Really, I think it's the GameFAQs Forum clique problem, to a significant extent. Some people really have nothing better to do except quash the utility of a site they have a small amount of control over, and it largely ruins the whole experience for the majority of other people. And, unlike a subreddit, you can't just go make your own version on the same topic. It doesn't work like that.

Rather than culture, for me it was the moderation body.

There were a bunch of people that were policing the dumbest minutie, and being assholish about that.

I occasionally would get my answers rewritten to have a completely different tone. My tone was not polemical or controversial in any way. It was just my tone. It was me. And somebody was not okay with that. They would edit MY TEXT and potentially put words in my mouth.

That's absolutely unacceptable.

Ten years on and i'm still salty about that, and i cheerish stackoverflow's death every time it gets brought up.

StackOverflow: Rest in piss, you won't be missed.

They exploit the basic human needs and build whole exploiting platforms on it and then kick you out from the nest like a cuckoo does with other bird chicks. The abnormal modding and other shenanigans also killed my interest quite quickly, but in hindsight I have to say, good riddance. People are naive and gullible, and give out decades of experience on a whim, because they think that everyone is as good and benevolent as them. While in reality there are a lot of truly evil and selfish people out there and at the end of the day using their knowledge against the good ones.

And now here we are in the AI madness, where a similar exploitation is happening just at a much bigger scale, using the benevolent and naive ones information against them.

“I noticed $email looked like $('#email')” isn’t exactly an exciting bit of history, even if you spend multiple paragraphs trying to pump yourself up.