Live data from Hacker News

Total monthly number of StackOverflow questions over time

data.stackexchange.com

621–630 of 1001 posts

Re: Total monthly number of StackOverflow questions over time

#621
post #456

Earlier quoted context omitted.

> I disagree with most comments that the brusque moderation is the cause of SO's problems, though it certainly didn't help. SO has had poor moderation from the beginning. I was an early SO user and I don’t agree with this. The moderation was always there, but from my perspective it wasn’t until the site really pushed into branching out and expanding Stack Exchange across many topics to become a Quora style competitor…

There was definitely a bit of Pournelle's Iron Law of Bureaucracy [0] at work. I worked there during a lot of the drama you allude to, and... It sucked, for everyone. But also... For most of SO's history, the vast majority of visitors (and this questions, answers) came in via Google. Not "search engines"; Google. This was pretty much baked in right at the start, and it effectively served as the site's primary user in…

Lots of moderation issues are also UI issues.

I suspect it’s the same issue for whatever is the “meta” in a competitive video game.

Optimization based on the available affordances ?

Re: Total monthly number of StackOverflow questions over time

#622

I once published a method for finding the closest distance between an ellipse and a point on SO: https://stackoverflow.com/questions/22959698/distance-from-g... I consider it the most beautiful piece of code I've ever written and perhaps my one minor contribution to human knowledge. It uses a method I invented, is just a few lines, and converges in very few iterations. People used to reach out to me all the time with…

Has anyone tried building a modern Stack Overflow that's actually designed for AI-first developers? The core idea: question gets asked → immediately shows answers from 3 different AI models. Users get instant value. Then humans show up to verify, break it down, or add production context. But flip the reputation system: instead of reputation for answers, you get it for catching what's wrong or verifying what works. "This breaks with X" or "verified in production" becomes the valuable contribution. Keep federation in mind from day one (did:web, did:plc) so it's not another closed platform. Stack Overflow's magic was making experts feel needed. They still do—just differently now.

Re: Total monthly number of StackOverflow questions over time

#623

Earlier quoted context omitted.

I had a similar beautiful experience where an experienced programmer answered one of my elementary JavaScript typing questions when I was just starting to learn programming. He didn't need to, but he gave the most comprehensive answer possible attacking the question from various angles. He taught me the value of deeply understanding theoretical and historical aspects of computing to understand why some parts of progr…

Had a similar experience. Asked a question about a new language feature in java 8 (parallell streams), and one of the language designers (Goetz) answered my question about the intention of how to use it. An LLM couldn't have done the same. Someone would have to ask the question and someone answer it for indexing by the LLM. If we all just ask questions in closed chats, lots of new questions will go unanswered as thos…

Would you share the link to the answer?

Re: Total monthly number of StackOverflow questions over time

#624
post #525

Earlier quoted context omitted.

I wonder if there could be something like a Wikipedia for programming. A bit like what the book Design Patterns did in 1994, collecting everyone's useful solutions, but on a much larger scale. Everyone shares the best strategies and algorithms for everything, and updates them when new ones come about, and we finally stop reinventing the wheel for every new project. To some extent that was Stack Overflow, and it's als…

> To some extent that was Stack Overflow Yup, that was always very much the plan, from the earliest days. Shame it soured a bit, but since the content is all freely reusable, maybe something can be built atop the ashes?

There is https://grokipedia.com which encourages you to suggest an article and you may submit improvements to an existing article.

Re: Total monthly number of StackOverflow questions over time

#626

Actual analysts here that have looked at this graph like... a lot, so let me contextualize certain themes that tend to crop up from these: - The reduction of questions over time is asymptomatic of SO. When you have a library of every question asked, at some point, you asked most of the easy questions. Have a novel question becomes hard. - This graph is using the Posts table, not PostsWithDeleted. So, it only tells yo…

OP here: I had the same thought, but noticed a very similar trend in both [0]; I think this graph is more interesting because you'd expect the number of new users to be growing [1], but this seems to have very little effect on deleted questions or even answers

[0]: https://data.stackexchange.com/stackoverflow/query/1927371#g...

[1]: https://data.stackexchange.com/stackoverflow/query/1927375#g...

The second graph here ([1]) is especially interesting because the total montly number of new users seems completely unrelated to number of posts, until you filter for a rep > 1 which has a close to identical trend

Re: Total monthly number of StackOverflow questions over time

#627

I once published a method for finding the closest distance between an ellipse and a point on SO: https://stackoverflow.com/questions/22959698/distance-from-g... I consider it the most beautiful piece of code I've ever written and perhaps my one minor contribution to human knowledge. It uses a method I invented, is just a few lines, and converges in very few iterations. People used to reach out to me all the time with…

Naive question maybe but how haven’t the models been trained on your answer if it’s on SO?

Models are NOT search engines.

Even if LLMs were trained on the answer, that doesn't mean they'll ever recommend it. Regardless of how accurate it may be. LLMs are black box next token predictors and that's part of the issue.

Re: Total monthly number of StackOverflow questions over time

#629

Earlier quoted context omitted.

There was, obviously, only one main reason: LLMs. Anything else makes no sense. Even if the moderation was "horrible" (which sounds to me like a horrible exaggeration), there was nothing which came close to being as good as SO. There was no replacement. People will use the best available platform, even if you insist in describing it as "horrible". It's was not horrible compared to the alternatives, web forums like Re…

Look at the data - it had already been on the downslide for years before LLMs became a meaningful alternative. AI was the killing blow, but there was undoubtedly other factors.

The decline was much slower, not the following exponential decline that can only have been caused by LLMs.

Re: Total monthly number of StackOverflow questions over time

#630
post #585

Earlier quoted context omitted.

My genuine impression is that communities moved from forums to discord. Maybe that's why they are harder to find

And discord is a terrible tool for knowledge collection imo. Their search is ok, but then I find myself digging through long and disjointed message threads, if replies/threading are even used at all by the participants.

Not to mention, it's not indexed by search engines. It's the "deep web".
Post reply on HN