News for hackers. Stuff that matters.
Re:StackOverflow now gets fewer than 1000 questions a month by zahlman on Thursday October 29, 2026 @02:00AM (#50285037)
Your question was neither deleted nor locked. It was closed as a duplicate, which immediately provides a link to existing useful information and helps keep the site organized by putting the best answers in one place.

My extensive experience has been that the overwhelming majority of the time that people complain "this isn't a duplicate", it actually quite obviously is. (And there were even more irritating complaints on the meta site at times; all the way up to the level of "you shouldn't ever close duplicates at all because I think you shouldn't", demonstrating no interest in even trying to understand the reasons for doing so.) Of course, people do make mistakes; there was an entire process for fixing this, but most people could not be bothered to engage with it productively (or even civilly). But feel free to give examples of things that you feel were wrongly closed from your own experience.

Re:StackOverflow now gets fewer than 1000 questions a month by kstrauser on Thursday October 29, 2026 @02:12AM (#50285107)
It's astonishing how vastly different our experiences were. I trawled the new Python questions frequently, looking for fun, novel things to answer. It was dismaying how often they'd been wrongly marked as dupes, often because the original answer was ancient and obsolete. It would often be egregiously bad, like:

Q: How do Python's dict preserve insertion order?

A: [duplicate of "Why does a Python dict's keys have a different ordering each run?"]

It was clear that the mod/curator/whatever spent no time attempting to understand what was actually being asked, and linked it to something that was completely incorrect as an answer to the real question.

I saw this constantly, so frequently that I'd be hard pressed to answer with specifics. If you asked me "when did you last eat a fruit bat", I can tell you the month and year because that stood out. If you asked me "when did you last eat spaghetti", I'd have a hard time saying.

Re:StackOverflow now gets fewer than 1000 questions a month by giancarlostoro on Thursday October 29, 2026 @11:53AM (#50291069)
Honestly, if the "duplicate" question is older than 5 or 10 years they should force the mod to prove its resolved the same way before marking it duplicate.
Re:StackOverflow now gets fewer than 1000 questions a month by braiamp on Thursday October 29, 2026 @01:45PM (#50293070)
Printing a string in python changed exactly twice. This is a 17 years old question about printing.

The top answer is still up to date

https://stackoverflow.com/q/493386/792066

Re:StackOverflow now gets fewer than 1000 questions a month by giancarlostoro on Thursday October 29, 2026 @03:04PM (#50294217)
Sure, but that's not the case for every question. I've found questions spanning various languages, .NET is more damning because the mod didn't apparently know that .NET Framework and .NET Core are distinctively different, so the problem might "look" related, but its not .NET Framework and .NET Core were NOT compatible at all. .NET 5+ became a .NET Core rebrand, and .NET Framework is basically abandonware. You cannot import .NET Framework libraries into .NET (Core / 5+) projects, but the confusion comes from libraries that migrated to Core and beyond, so a problem looks "familiar" but its completely different runtimes and in some cases underlying third-party code.
Re:StackOverflow now gets fewer than 1000 questions a month by braiamp on Thursday October 29, 2026 @08:08PM (#50298086)
Which mod? Show the question
Re:StackOverflow now gets fewer than 1000 questions a month by zahlman on Thursday October 29, 2026 @05:29PM (#50296333)
The large majority of duplicate closures are done by a consensus of three people, and the large majority of the rest are done by people who have a badge indicating expertise with the subject matter in addition to the close-vote privilege. It is not done by "the mod"; there have historically been about a couple dozen moderators for a couple dozen million user accounts.

Generally, the people doing this are familiar with the material. Fast closures are motivated by an unconscious "ugh, not THIS question again". And the kinds of things that get asked over and over usually don't change very much.

There's a particular type of question asker who likes to say "what's the way to do this in [current year]?" as if to imply that it should be something different from before. But it very often isn't. And indeed, developers complain when languages/libraries/ecosystems change too quickly, which makes language designers/library authors more conservative about those changes.

Re:StackOverflow now gets fewer than 1000 questions a month by zahlman on Thursday October 29, 2026 @05:24PM (#50296261)
> It's astonishing how vastly different our experiences were.

Yes, it is. Especially considering that I too was mainly "trawling the new Python questions" for similar reasons, and being dismayed at how many dupes there were and how hard it was to find an actually good version of that question to use as the duplicate target for something that was constantly asked.

> I saw this constantly, so frequently that I'd be hard pressed to answer with specifics.

Without specifics I really can't engage with that, sorry. I would encourage you to check out the [reopen-closed] tag on the meta site, in chronological order (or take a random sample) and see what kinds of things people think should be reopened and how off-base they are (and also see how it is explained that things are in fact duplicates).

As it happens, I have a single "saved" (site bookmarking feature) question on the meta site with that tag, saved because I wrote what I think is a useful reference answer to the general question of why duplicates get closed so quickly: https://meta.stackoverflow.com/questions/422581

> often because the original answer was ancient and obsolete

This does not make a question "not a duplicate". The fact that there are different ways to solve the problem now, doesn't make it not the same problem. It is, instead, a reason to write a new answer, but we still want all the high-quality answers in the same place.

From time to time, a new "canonical" gets chosen for FAQs, because the community recognizes that it's better quality. "Up-to-date" is one consideration in that equation. But a lot of the time, people argue that an old Q&A should be disregarded or replaced simply because it is old, and this is often just not true. For example, the question 'What does if __name__ == "__main__": do?' (https://stackoverflow.com/questions/419163) was posted in 2009; as it turns out, it still does the same thing in Python 3.15 and the reasons for using (or not using) it are the same as well.

Re:StackOverflow now gets fewer than 1000 questions a month by kstrauser on Thursday October 29, 2026 @05:49PM (#50296595)
I’ll be frank: I no longer care enough about SO to bother with researching it. The results of all the rules lawyering and purity tests ruined the fun of it for me — and according to the graphs, everyone else, too. It just wasn’t enjoyable anymore. I like answering questions and teaching and explaining. I don’t like spending more than 0% of my time dicking around with gatekeepers.

No amount of “you’re holding it wrong” could convince me that the millions of people who wanted to use it the way it was broadly described were truly doing it wrong. The payoff for slavishly doing things the “right” moderator-approved way just wasn’t worth the effort.

Re:StackOverflow now gets fewer than 1000 questions a month by zahlman on Friday October 30, 2026 @06:09AM (#50301879)
There have been countless other places throughout the history of the public Internet which could be used in that way, and were explicitly designed for that. The "broad description" was mostly telephone game, but a good part of it came from making assumptions based on, as far as I can tell, nothing more than the existence of a submission form.

Where a community exists that is already organized around a common goal, following previous leadership, newcomers should not be permitted to decide that actually the community should/does work according to what they want. No matter how many of them there are.

Re:StackOverflow now gets fewer than 1000 questions a month by pdw on Thursday October 29, 2026 @03:50AM (#50285682)
In my experience, any questions about an obscure or rare problem was likely to be marked as a duplicate if it sounded even vaguely similar to a common problem. My impression was that some moderators just were unwilling to admit they didn't know the answer to a question.
Re:StackOverflow now gets fewer than 1000 questions a month by giancarlostoro on Thursday October 29, 2026 @10:23AM (#50289600)
> Your question was neither deleted nor locked. It was closed as a duplicate

Fair, I got my wording wrong, its been a while since I've landed on the site... Sadly, I used to love reading SO for fun and to learn random tidbits.

I've run into way too many "this is a duplicate of" and its referencing something from 2003 in 2025 where the runtime has changed, and a similar looking issue was confused for the actual issue, the resolution in the linked "duplicate" doesn't apply or work. In my opinion, if you're going to call something a duplicate, prove that the solution there works before you lock it down.

Re:StackOverflow now gets fewer than 1000 questions a month by zahlman on Friday October 30, 2026 @06:33AM (#50302078)
Sure. But there were countless times that people insisted to me that their question was not a duplicate and that "the solution there" would not work, having not tried it (spoiler: it would work).