Stack overflow is almost dead
Stack overflow is almost dead
Today, Stack overflow has almost as few questions asked per month, as when it launched back in 2009. A recap of its slow, then rapid, downfall.
Topic already covered here https://stackoverflow.com/questions/5301389/how-to-use-exception-stack-trace-after-process-dead
Closing thread due to similar question already being answered.
😭
Do not resurrect a closed thread with unsupported emojis. User has been warned.
Well... https://lemmy.ml/post/30395990
That's too perfect. You win.
No, no, this question deserves the RTFM response, not the asked and answered. Unless you linked to an RTFM answer, in which case I'll allow it.
To be honest. (although I am guilty using chatgpt way too often) I have never not found a question + an answer to a similar problem on stackoverflow.
The realm is saturated. 90 % of the common questions are answered. Complex problems which are not yet asked and answered are probably too difficult to formulate on stackoverflow.
It should be kept at what it is. An enormous repository of knowledge.
This is a huge reason for the question decline! All the easy stuff has been answered, the knowledge is already there. But people are so used to infinite growth that anything contrary = death lol
People also blame ai, but if people are going to ai to ask the common already answered questions then… good! They’d just get hurt feelings when their question was closed as a dupe
Yeah, the article seems to assume AI is the cause without attempting to rule out other factors. Plus the graph shows a steady decline starting years before ChatGPT appeared.
People also blame ai, but if people are going to ai to ask the common already answered questions then… good!
exactly!
While I am indeed worried about the "wasted" energy (thats a whole other topic), thats pretty much why AI is good for.
I disagree. I still easily find new questions to ask, for example this one which is a nice demonstration of why StackOverflow is dying. Or this one (which also received 4 downvotes).
Even so, I definitely go to ChatGPT first now. Now that we finally have an alternative to the toxic downvotes/closing, why would I go there unless I absolutely need to?
That is a good example. If you aren't already aware you might be interested in ntfs-3g, the GPL-2.0 licensed ntfs driver - the function ntfs_upcase_table_build in libntfs-3g/unistr.c may be of some assistance
This is a really good point. I joined stackoverflow after graduating university a few years ago, and found it really hard to participate. You need karma to be able to vote on stuff or add comments, but the only unanswered questions are often basically unanswerable. I did find some success with adding answers that were better than previous ones, but it was limited, because at that point the site was already declining and there was no one left to upvote my contributions.
I'd wager that most new questions are in cutting-edge tech. Older tech will be saturated.
It's because all the questions have already been answered. How many times can you answer how to reverse an array in javascript?
... ... How do you reverse an array in Javascript?
marked as duplicate. closed.
array.reverse() should do the trick 😆
I haven't used it a lot even before AI.
This is exactly what I came here to say. They are militantly against duplicates. Doesn't that mean on a long enough timeline the number of new questions have to eventually reach zero?
I use stack overflow every day and have for years. I have never once had to ask a question.
A lot of people seem to be celebrating this, but I personally think this is a net negative for programming. Are people actually replacing SO with talking to LLMs? If not, where are they going?
I've seen an uptick in people using places like discord to get help. But that's not easily searchable and not in the same format that it is in stackoverflow. SO was meant to organize these answers to make asking questions easier. Now it seems like we're walking away from that, and I can't quite understand why. Is it really because SO is "toxic"?
I'm finding most of what I'd ask on SO can be asked on the tools GitHub issues. If a product doesn't offer a support form or GitHub issues it doesn't get used for me.
To the surprise of absolutely no one. Tends to happen when you cultivate one of the most tixic online spaces on the net. I've never asked a question on SO, but just the verbiage used to accost people just trying to learn is just insane. Mods don't really care about post content as long as its not perceived as "hostile," so you can be generally as passive aggressive and shitty as you want. It's just...weird.
You can find especially viperis content when you find a question which has been answered, but someone is just like "Well, this isn't the way that I do it!" etc, and then go on a tirade about how the question was asked poorly and the answer doesn't completely answer the question.
Shit is just wild.
I use SO daily and never seen anything like you describe there. All I see is that incorrect answers are down voted. I don't know, maybe I just don't pay attention to the "verbiage". I look at the code sample and move on. In the end, it's not a forum. I'm not there to read opinions.
I find the concept of downvoting very toxic and discouraging. It can potentially prevent people to express different views, something a discussion and our personal development is thriving on. It can be well seen on Reddit and even on Lemmy, where people with different views get sometimes heavily downvoted. It is something I consider to be close to “cancel culture” - a majority decides not to like your opinion, so it tries to silence you by voting you “out”. I would really love to see that Lemmy removes this feature and just allows to upvote - so you can upvote a comment or not, but you cannot downvote a comment.
I asked a question on there about Apache Poi. Then no one answered it so I found a solution and answered it myself. Must've stayed relevant because I fielded a few questions about it for years.
Then they took my account away, I think maybe because I didn't confirm my identity after a big breach? Then I looked for my Q/A and it was attributed to someone else. I was hot about it for a minute and then realized I didn't care and was finally free from being the expert in that one niche thing I've never done since.
I think it'll make a comeback eventually. LLMs will get progressively less useful as a replacement as its' training data stales. Without refreshed data it's going to be just as irrelevant as the years go on. Where will it get data about new programming languages or solutions to problems in new software? LLM knowledge will be stuck in 2025 unless new training material is given to it.
Well. I doubt that very much. Take as an analogy the success of the chess AI which was left training itself - compared to being trained...
Your analogy simply does not hold here. If you're having an AI train itself to play chess, then you have adversarial reinforcement learning. The AI plays itself (or another model), and reward metrics tell it how well it's doing. Chess has the following:
- A very limited set of clearly defined, rigid rules.
- One single end objective: put the other king in checkmate before yours is or, if you can't, go for a draw.
- Reasonable metrics for how you're doing and an ability to reasonably predict how you'll be doing later.
Here's where generative AI is different: when you're doing adversarial training with a generative deep learning model, you want one model to be a generator and the other to be a classifier. The classifier should be given some amount of human-made material and some amount of generator-made material and try to distinguish it. The classifier's goal is to be correct, and the generator's goal is for the classifier to pick completely randomly (i.e. it just picks on a coin flip). As you train, you gradually get both to be very, very good at their jobs. But you have to have human-made material to train the classifier, and if the classifier doesn't improve, then the generator never does either.
Imagine teaching a 2nd grader the difference between a horse and a zebra having never shown them either before, and you hold up pictures asking if they contain a horse or a zebra. Except the entire time you just keep holding up pictures of zebras and expecting the child to learn what a horse looks like. That's what you're describing for the classifier.
Until someone releases an open LLM in the sense that every prompt/question is published on a forum like site
Good. StackOverflow is toxic, while I was in school I would ask questions that were “obvious” I guess. I’d get told that I’m dumb (didn’t get those words but it was implied) when trying to ask for clarification. Then I got banned from posting anymore questions due to downvotes. Like imo how can you learn if people shun you for asking questions?
Reddits programming community was more welcoming and kinder than the stuck up folk on SO.
It's mainly a different model, but I totally sympathize that it's the opposite of welcoming or encouraging.
SO recognizes that many, many questions are really just rephrasings of the same underlying question, and the aim is to find and provide the best answer to those. It explicitly does not want to repeatedly answer the same question, and given how few people find out how it works before simply asking, they have to be pretty ruthless about it. The result is that usually the most active and fleshed out questions and answers are very informative. So there's a big upside in trade for those downsides. Answers are meant to be durable, ~singular, and authoritative.
Reddit is basically halfway between that, and Discord. Discord is the polar opposite, questions and answers are naturally ephemeral, duplication happens constantly, and quality of responses is all over the map.
I greatly prefer the StackOverflow model, and - to be very clear - I have never once asked (to say nothing of answering) a question of my own there, lmao.
looks like an opportunity for the fediverse
In my opinion SO has been replaced by forges discussions/issues. Whenever I have an issue with some library or piece of software, I will always check their repo to see if someone is going through the same thing or how they solved it.
When I tried to engage with SO it was a pain in the ass so I just stopped answering/asking.
My favourite closed question is how to do Case-insensitive string comparison in C++ - closed as opinion based!
ChatGPT helped, but this is why you died StackOverflow.
The question asks for "the best" way to do it (making it opinion based) and forbids a potential solution without explaining why (it's clearly some kind of assignment, but that doesn't matter here). And it has plenty of answers both using Boost and in pure C++, so I'm not sure why that wasn't enough for you. Just because it's closed doesn't mean the answers already provided are bad.
By that measure basically every StackOverflow question asking how to do something is opinion based - the very nature of the site is questions asking for the best solutions. The "opinion based" rules is NOT meant to prevent questions like this. This is the kind of useless pedantry that killed StackOverflow.
I think it comes from a fundamental disconnect. You have something like:
- People ask questions like "which programming language is best" or "what's the best game engine" or "should I use tabs or spaces"?
- Someone decided they didn't want StackOverflow being used to debate these things, so they made a rule against opinion-based questions.
- People later come along and blindly apply the rule to ban anything that is worded as if it is an opinion, even if it's a perfectly suitable question.
not sure why that wasn’t enough for you
I never said it wasn't; just that it shouldn't have been closed.
Just because it’s closed doesn’t mean the answers already provided are bad.
Again, I never said otherwise. The point is it shouldn't have been closed.
Its been a dead end a long time.
There's no one thirstier than someone with a coding question.