Welcome to Software Development on Codidact!
Will you help us build our independent community of developers helping developers? We're small and trying to grow. We welcome questions about all aspects of software development, from design to code to QA and more. Got questions? Got answers? Got code you'd like someone to review? Please join us.
What’s the point of Q&A sites anymore?
Lately, whenever I consider asking a question on a site like Software Codidact, I end up just asking an LLM chatbot instead. Just thinking of the effort of distilling a minimal working example, formatting the post correctly, and anticipating the downvotes, is enough to have an aversion to the site. So what’s the point of these Q&A communities if the friction to participate often outweighs the benefit, especially when a language model can give you an answer instantly?
In short: LLMs are trained based on what we already know. I'm oversimplifying massively, but essentially they produce a …
6mo ago
Although I was tempted to add a sarcastic comment suggesting you could just ask this Meta question of an LLM, I do think …
6mo ago
Asking the last five or so questions on SO (I am new here), I felt pretty much the same as you do. I've put a lot of eff …
2mo ago
There are always problems that LLMs can't answer, or they give trash answers. Then you check the traditional sources, do …
5mo ago
Just thinking of the effort of distilling a minimal working example, formatting the post correctly, and anticipating the …
6mo ago
5 answers
You are accessing this answer with a direct link, so it's being shown above all other answers regardless of its score. You can return to the normal view.
There are always problems that LLMs can't answer, or they give trash answers. Then you check the traditional sources, do your investigation, collect references - and summarize them in a Q&A here.
In my opinion, creating self-answered questions this way serves not only the knowledge of the humanity, but it is also a memo for yourself. Obviously it is a very generous thing to share the findings with your human and program visitors.
I believe the role of the original idea (humans asking and answering) will decrease. Obviously there is a lesser need for those answers created by humans, which can be answered also by machines.
At least, that is the case now. Odds are that the amount of the questions answerable by humans only, will decrease over time, but it is not sure. No one knows the future yet.
0 comment threads
In short: LLMs are trained based on what we already know. I'm oversimplifying massively, but essentially they produce a string of words which is likely to follow the prompt you provide based on extensive training on samples of language, code, etc. At their core, they are statistical models, and as the saying goes— lies, damned lies, and statistics. They are not general artificial intelligence; they carry no intrinsic understanding of their input or output, and are therefore not capable of knowing that their output is correct in the same way a human is.
They're useful for some things, in the same way data and statistics are. Here is an example of a model:
The points are my training data; the dotted line is my model. My model predicts that given an input of 11, the output will also be 11. In this way, a good model can make good predictions for what comes next; LLMs work essentially the same way, but on billions of input parameters and training data points. That said, what's to say that my graph doesn't show a sequence which maintains $x=y$ from 1 to 11, but then jumps to 13, then 24, then 35, etc? It's not possible for my model to know or predict this, in the same way that it's not possible for an LLM to predict with 100% accuracy the correct answer.
Sorry, slight tangent.
What does that mean for Q&A sites? An LLM may be able to give you a correct answer a lot of the time, particularly for predictable or repetitive tasks; computer code often follows a very well-defined syntax and API so is often predictable enough for an LLM to get it right.
What the LLM doesn't know is anything that it hasn't been trained on. Anything new. If I were to create a new programming language and not share a formal syntax or API with anyone, then—quite apart from the fact that it would be useless—LLMs would be unable to accurately predict what comes next.
That goes for anything that's new. For Q&A communities like this, perhaps that means a pivot away from asking simple questions looking for help debugging code, and towards asking complex questions looking for ideas that only humans can generate. What should I consider when structuring this large and complicated project? What are the trade-offs of this niche technical decision?
Just what is a foo, anyway?
Although I was tempted to add a sarcastic comment suggesting you could just ask this Meta question of an LLM, I do think it's a useful question, which gives the opportunity for answers that explain the problems with LLM output, and the benefits of asking on a Q&A site.
I think the existing answers sum these up well, but for anyone who doesn't want to trust biased humans, here's how a large language model, specifically Claude Haiku 4.5 via search engine DuckDuckGo's "Duck ai" responds to the body of your question:
You've identified a real tension, and your observation about friction is valid—but I think it points to something more nuanced than Q&A communities becoming obsolete.
The friction you're describing is partly intentional
The effort of creating a minimal working example, formatting clearly, and anticipating criticism isn't just bureaucratic overhead. It serves a function: it forces you to understand your problem more deeply. When you distill a question down to its essentials, you often solve it yourself. When you format it well, you're not just being polite—you're organizing your thinking in a way that makes the answer more useful to others and more useful to future-you.
An LLM will skip this step. It'll give you an answer quickly, but you might not internalize the underlying principle. You might copy-paste a solution without understanding why it works, which bites you later when the context changes slightly.
Q&A communities solve different problems than LLMs do
LLMs are great for:
- Quick syntax lookups
- Explaining concepts you're already familiar with
- Brainstorming or rubber-ducking
- Getting unstuck when you're blocked
Q&A communities are better for:
- Legitimately novel or edge-case problems where the LLM might confidently hallucinate. A human who's encountered the exact same issue can say "I've been there; here's what actually works."
- Accountability and vetting. Downvotes and comments aren't just friction—they're a filter. A highly-upvoted answer has survived scrutiny. An LLM answer has only survived being plausible.
- Discovering that your question is wrong. Someone might comment "you're trying to solve this the hard way; here's a better approach." An LLM might not challenge your framing.
- Long-term value. A Stack Overflow answer from 2012 is still useful. An LLM conversation disappears.
The real decline isn't inevitable
You're right that friction matters—sites can become less useful if they're too unwelcoming. But that's a design problem, not a fundamental one. Some communities have actually reduced friction while maintaining quality (better templates, friendlier moderation, clearer guidelines).
The bigger issue is that Q&A sites thrived when they were the fastest way to get an answer. Now they're not. So they need to offer something else: depth, vetting, curation, community, or solving problems that LLMs actually can't solve well.
What this means for you
If you're just looking for a quick answer to a straightforward problem, the LLM is genuinely the better choice—that's not a failure of Q&A sites, it's a shift in what tools are best for what tasks. But if you're stuck on something genuinely tricky, or if you want to understand why something works, or if you want your solution to help others years from now, the friction of a Q&A community might actually be worth it.
The sites that will survive are probably the ones that lean into what they do better than LLMs: specialized knowledge, community judgment, and long-term curation. General-purpose Q&A might fade. Niche communities where people have real expertise and investment? Those might actually strengthen.
My thoughts
Although this highlights several of the problems with LLMs and benefits of asking on a Q&A site, I'd be wary of the conclusion:
If you're just looking for a quick answer to a straightforward problem, the LLM is genuinely the better choice
Note that LLM hallucination (confidently asserting something that is false) is not restricted to complex problems as claimed here. Even straightforward problems can result in false answers that sound plausible, even citing sources (that turn out to say the opposite if you read them).
This problem isn't unique to LLMs. Humans can be confidently wrong too. The difference with a Q&A site is that you can get answers from several humans, with voting and comments making confidently wrong answers less likely to convince you.
Asking an LLM is only safe in the same cases that asking an individual human is safe: When you feel qualified to assess the answer you receive. A Q&A site provides some degree of assessment, rather than just providing an answer.
0 comment threads
Asking the last five or so questions on SO (I am new here), I felt pretty much the same as you do. I've put a lot of effort in illustrating the problem and ask a concrete question. While others confirmed they were legit questions, they received downvotes without explanation, and as one commenter pointed out:
Eh, the replies are mostly about side stuff rather than what you're asking about.
So I asked my local Gemma 4 and got a good, well-balanced answer.
I also asked Gemma your question and the final point is:
The friction you feel is real, and it is a sign that these communities need to evolve—perhaps by becoming less about "fixing syntax" and more about "discussing architecture and nuance"—to remain relevant in the age of AI.
Or as ArtOfCode put it:
For Q&A communities like this, perhaps that means a pivot away from asking simple questions looking for help debugging code, and towards asking complex questions looking for ideas that only humans can generate.
I have actually become unsure about "asking complex questions looking for ideas" as I've received feedback like "too broad; inviting for opinionated answers" and close votes when I tried that on SO.
Just thinking of the effort of distilling a minimal working example, formatting the post correctly, and anticipating the downvotes, is enough to have an aversion to the site.
Frankly, this suggests your attitude is the problem. Many questions don't require an example at all. Reasonably formatting text you write is something you should have been used to since grade school, and is really no extra work. If you're old enough to post here, writing text that others can easily read should come naturally.
The real effort of asking any question, whether in a Q&A forum, elsewhere on line, or in person, is to think about the problem. What is it you are really stuck on? What are you asking others to do for you? What have you already done to try to solve the problem, and what does the failure of those things tell you about the problem? What context do others need to understand your problem? What information do you need to include to describe the problem succinctly, but completely enough so that others can help?
Yes, it takes some effort to properly ask a question, even if that's mostly organizing your thoughts. Note that this process is useful, even required, so solve the problem whether you are asking other people, bots, or a rubber ducky. In fact, that's the whole point of asking a rubber ducky, and why that method actually works sometimes.
If you're not willing to put in the effort, then you're not going to be successful on a Q&A site, but probably not elsewhere either. On a Q&A site there are real people on the other end which will downvote questions that appear to come from a lazy entitled ingrate. That's as it should be. If you find your questions getting downvoted regularly, then the problem is with you, not the Q&A site.
So what’s the point of these Q&A communities if the friction to participate often outweighs the benefit
It doesn't, to most users in most cases anyway. If you put in the effort as described above, there won't be any friction.
especially when a language model can give you an answer instantly?
Sure, it will give you an answer. Is it the right one, though? How do you know? Vetting an answer can be more work than generating a proper answer from scratch.
A bot versus real people on the other end of a Q&A site are very different resources with different strengths and weaknesses. If you are looking for a purely factual answer, something that a human would "look up", then a bot can be appropriate.
The Q&A site, on the other hand, gets you access to human experience, intuition, ability to make inferences and extrapolate from data, and the ability to not only give an answer, but explain why it's a good answer.
0 comment threads
2 comment threads