SO was a governance disaster. It was the infighting it was the strike, it was bad auto moderation tools. And the Stack Exchange strike wasn't really about AI. AI was just the final trigger after years of management undermining the community that made the platform valuable.

Stack Exchange was built on an implicit social contract people contributed their expertise and thousands of hours of unpaid work because they believed they were building a collectively governed knowledge resource. The company owned the platform, but the community created its value.

After time, management became increasingly centralized and disconnected from contributors. Also the policing of questions based on geeky criteria and not what users wanted. The increasingly aggressive policing of questions, closing perfectly reasonable discussions, and incorrectly saying questions are duplicates made the platform unnecessarily hostile to everyone, qmd at least thats why i quit. Quality control gradually became useless gatekeeping.

So there were really more than one problems, corporate management treating contributors as free labor, and established contributors treating their moderation as authority over everyone else, there were bad tools closing questions. The platform had become extremely hard to use for posting questions, finding amswers, moderator work. For me it was useless. Incidentally the closing of “opinion” questions was the worst decision since it was the only thing that could make it distinct from AI produced answers.

For some of the events there is paper here with details https://arxiv.org/html/2512.08884v1

I had stopped going to it quite some time ago.

1. It was hard to ask any question. Sometimes it felt if a few characters match it will be closed off as dupe. Besides no thought was given to the fact that same question might change entirely from pov of the answers for a platform or SDK across, say, a decade (?).

2. I also realised reading the documentation (that is if available) and looking though github and people's blogs helped better.

3. But bigger takeway was that due to relative uselessnes (if not downright hostility) of SO I started just breaking down the problem and solving on my own, talk to colleagues, or look for solutions for broken down problems and then I think I used to come across SO in months and that too most of the time when a colleague will refer to something.

This was before LLMs or COVID.

> 1. SO question numbers dropped precipitously before AI came along.

This was single-handedly the work of arrogant self-righteous moderators who would close random useful questions as not useful just to prove they were doing something. A lot of us long timers got annoyed at selective enforcement of rules and never looked back. It's sad because it was actually a fun place to hangout. Too much moderation always kills the vibe.

I contend that shitty moderation was why it flatlined.

I enjoyed answering algorithm questions. Unfortunately, every so often, people who weren't into algorithms would show up. And somehow summoned each other. To close good questions.

It was clear that they didn't understand the topic. They couldn't recognize what a good question, or a good answer, was. I wound up going away, coming back every few years, go until my next bad experience, and leave again.

Each time, the population answering those questions was smaller.

I believe that a lot of good people left like I did. Then just didn't come back.

The issue was that Stack overflow never was enjoyable to use.

Asking questions was difficult and any difficulty questions I might ask would not get a response anyway.

Answering questions was a gamble, sometimes you would get rewarded, most of the time you would not.

If you look at the data on a log scale, and include other stack exchange sites, it's apparent that the serious decline is starting after ChatGPT launches: https://x.com/CraigGidney/status/2115664509042053377/photo/1

One way that stackoverflow does stand out from the other stackexchange sites is that its half life appears far shorter. On a log scale plot it has a steeper descent than the other stackexchange sites. I'm not sure what to make of that. Another interesting thing is that the math stackexchange initially declines more slowly than the others for a year or two and then decays pretty similarly. You can also see examples in some of the other sites of their decay beginning before chatgpt releases (e.g. drupal). You can also see some sites decaying a decade earlier (e.g. expressionengine). I think those outliers are from their subject being too specific and going out of style.

I got the cross-site data from data.stackexchange.com with this vibe-coded query:

    DECLARE @sql NVARCHAR(MAX) = N'
    SELECT
      site = [$site_name$],
      month = DATEADD(MONTH, DATEDIFF(MONTH, 0, CreationDate), 0),
      questions = COUNT(*)
    FROM PostsWithDeleted
    WHERE PostTypeId = 1
    GROUP BY DATEADD(MONTH, DATEDIFF(MONTH, 0, CreationDate), 0)
    ';

    EXEC dbo.sede_ineachdb
      @SQLCommand = @sql,
      @IncludeMainSites = 1,
      @IncludeMetaSites = 0,
      @IncludeMainMeta = 0;

Agreed, I found SO and Quora and similar sites with general or specific question/answer based communities weak long before AI.

AI replaced Google Search in many ways more directly on arrival than those.

SO to me was weak as it had outdated answers to outdated questions (yet related to search queries) that nobody maintained well.. and the community more focused on patronising to push their point of view than being essentially objective in many ways.

that was a total killer to me and I started preferring to come back to books, manuals and researching topics myself than short cuts like SO.. and that research is heavily optimised during AI/LLM era, yes, but that's just one of many nails to SO's coffin.

A downward trend in the number of questions doesn't imply the site was in decline, since there's only so many questions you can ask about a given technology, so would-be question askers find the answer via search engine rather than having to create a new question on the site.

Tangental, but I tried plotting number of questions by month including deleted questions. It shows similar trends to questions by month (so nothing particularly interesting): https://imgur.com/a/NLLB3OK

Query: https://data.stackexchange.com/stackoverflow/query/edit/1969...

https://data.stackexchange.com/stackoverflow/query/1948078/c... shows the situation even better.

And, StackExchange has adjusted stale/inactive question deletion scripts as a result of this change: https://meta.stackexchange.com/questions/419988/we-ve-rework...

In my experience, SO became useless many years before AI showed up on the horizon. I stopped actively participating about 2019 and about 2022 gave up on it completely.

I like that you can see a Covid bump in there and then it immediately craters lol

Often overlooked but Discord(and similar communities) are also a big reason why people no longer ask ques. on SO.

No surprises here. However I recently tried to register global account on Stack Exchange network, just to get to some organic conversations and meet people... and I couldn't. Gateway errors, "try again in a few minutes", etc. I did quick research and it seems that I'm not the only one. Maybe I just run into one of the bad days.

Where else were people getting answers and help from between 2018 and AI getting good in the mean time? As I remember it there wasn't really an alternative.

Maybe it was more consolidation of tech and better docs over time.

Yes that was my impression too. SO already died before the rise of AI.

> SO had been flatlined since 2014 and declining since 2017.

I'm fairly certain this early decline came from the fact that many common questions had already been answered, and therefore were available via Google, so hardly anybody asked them again. In the beginning there were just way more unanswered questions.