> As moderation and curation restricted (changing the close reasons to more and more specific things - "it's not on that list, so you can't close it") meant that the content that was not as well thought out but did match the rules became more and more prevalent and overwhelmed the ability for the "spolskyites" to close since so many of the atwoodians have left.

Just to make sure: I always got the impression that Atwood was the one who wanted to keep things strictly on mission and Spolsky was the one more interested in growing a community. Yes? I do get the impression that there was a serious ideological conflict there; between the "library of detailed, high-quality answers" and the, well, "to every question" (without a proper understanding of what should count as a distinct, useful question that can have a high-quality answer). But also, the reputation gamification was incredibly poorly thought out for the "library" goal (https://meta.stackexchange.com/questions/387356/the-stack-ex...). And I suspect they both shared blame in that.

A lot of it was also ignored for too long because of the assumption that a) the site would just die if it clamped down on everything from the start; b) the site would naturally attract experts with good taste in questions (including maybe even the ability to pose good https://en.wikipedia.org/wiki/Dorothy_Dixer questions) before the beginners ever cleared the barrier of trying to phrase a proper question instead of using a forum.

(Nowadays, there are still small forums all over the place. And many of them try to maintain some standards for the OP. And they're all plagued with neophytes who try to use the forum as if it were a chat room. The old adage about foolproofing rings true.)

Around 2014 is when the conflict really seems to have boiled over (as new question volume was peaking). Notably, that also seems to be when the dupe-hammer was introduced (https://meta.stackoverflow.com/questions/254589).

▲ shagie 4 days ago | parent [-]

Jeff was the author of https://stackoverflow.blog/2011/06/13/optimizing-for-pearls-... and was more focused on quality than community - his vision was the library.

Joel was indeed more community minded - though part of that community mindedness was also more expectations of community moderation than what the tooling was able to scale for.

And yes, they both were to blame for gamification - though part of that was the Web 2.0 ideals of the time and the hook to keep a person coming back to it. It was part of the question that was to be answered "how do you separate the core group from the general participants on a site?" ... and that brings me to "people need to read A Group Is Its Own Worst Enemy" ( https://news.ycombinator.com/item?id=23723205 ) to understand how it shaped Stack Overflow.

https://blog.codinghorror.com/its-clay-shirkys-internet-we-j... (2008)

https://web.archive.org/web/20110827205048/https://stackover... (Podcast #23 from 2011)

   Atwood: Maybe. But the cool thing about this is this is not just me, because that would be boring. It is actually me and Clay Shirky. You know, Clay Shirky is one of my heroes.

   Spolsky: Oh...

   Atwood: Yeah I know, it's awesome. So we get to talk about like building communities online and I get to talk about StackOverflow, you know, and all the lessons we've learned and, get to present with Clay. Obviously he's an expert so. That's one of the people that I have emailed actually, because I thought that would be good, because he is from New-York city as well. So we could A) show him the site and B) talk about the thing we are going to do together in March, because he needs to see the site to have some context. I mean I did meet him and talk to him about this earlier a few months ago, I think I mentioned it on the podcasts. But that was before we had sort of even going to beta, so there's really not a lot to show him. But I would love to show him in person. So we'll see if I'll hear back from him, I do not know.

https://meta.stackexchange.com/questions/105232/clay-shirkys... (2011)

2014 sounds about right for when it peaked... it was also when a lot of things hit the fan one after another. General stress, the decline of community moderation. The dup hammer was a way to try to reduce the amount of close votes needed - but in doing so it became "everything is a nail" when the dup hammer. It was used to close poor questions as dups of other questions ... and rather than making it easier to close questions that didn't fit well, corporate allowed the "everything is a dup" problem to fester.

That also then made Stack Overflow's search become worse. Consider https://meta.stackoverflow.com/a/262080 which provides itself as a timestamp of 2014...

    How much traffic do the questions that get duped to something bring? Especially the (currently) 410 questions linked to the Java NPE question.

That question now has 10,356 questions linked to it... and that's part of the "why search quality is going down" - because poor questions were getting linked and not deleted. Search went downhill, dupe hammer was over used because regular close votes took too long because community moderation was going down, which in turn caused people to be grumpy about "closed as dup" rather than "your question looks like it is about X, but lacks an MCVE to be able to verify that... so close it as a dup of X rather than needing 5 votes to get an MCVE close.. which would have been more helpful in guiding a user - but would mean people would start doing FGITW to answer it maybe and you'd get it as a dup of something else instead."

All sorts of problems around that time.

	▲	zahlman 4 days ago \| parent [-]
		Thanks; lots of great information here. Regarding duplicates and deletion you may be interested in my thoughts: https://meta.stackoverflow.com/questions/426214/when-is-it-a... ; https://meta.stackoverflow.com/questions/434215/where-do-the... ; https://meta.stackoverflow.com/questions/421677/closing-a-qu... seem relevant here, browsing through a search of my saved posts. Having duplicates should make the search better, by pointing people who phrase the same problem in different ways to the same place. But low-quality questions often don't produce something searchable for others, and they cover topics relevant to people who lack search skills.