Large language models reduce public knowledge sharing on online Q&A platforms • Hacker News Board

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#201

Post by Ferret7446 » Sun, Oct 13, 2024, 5:57 PM UTC

Aurornis wrote: post #161

melenaboija wrote: post #38

It’s been a relief to find a platform where I can ask questions without the fear of being humiliated Half joking, but I am pretty tired of SO pedantry.

Many of the forums I enjoyed in the past have become heavily burdened by rules, processes, and expectations. They are frequented by people who spend hours every day reading everything and calling out any misstep. Some of them are so overburdened that navigating all of the rules and expectations becomes a skill in itself. A single innocent misstep turns simple questions into lectures about how you’ve violated the rule…

The reason those rules are created is because at some point something happened that necessitated that rule. (Not always of course, there are dictatorial mods.)

The fundamental problem is that communities/forums (in the general sense, e.g., market squares) don't scale, period. Because moderation and (transmission and error correction of) social mores don't scale.

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#202

Post by Ferret7446 » Sun, Oct 13, 2024, 6:01 PM UTC

wokwokwok wrote: post #109

beeboobaa3 wrote: post #101

Earlier quoted context omitted.

I'm sorry but the funny thing is, the only people I've ever seen complain about SO are people who don't know how to search.

Everyone has a pet theory about what’s wrong with SO; but here’s the truth: Whatever they’re doing, it isn’t working. Blame mods. Blame AI. Blame askers… whatever man. That is a sinking ship. If you don’t see people complain about SO, it’s because they aren’t using it, not because they’re using the search. Pretty hard to argue at this point that the problem is with the users being too shit to use the platform. That’s…

I submit that what SO is doing is working; it's just that SO is not what some people want it to be.

SO is not a pure Q&A site. It is essentially a wiki where the contents are formatted as Q&As, and asking questions is merely a method to contribute toward this wiki. This is why, e.g., duplicates are aggressively culled.

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#203

Post by Aurornis » Sun, Oct 13, 2024, 6:01 PM UTC

Ferret7446 wrote: post #201

Aurornis wrote: post #161

Earlier quoted context omitted.

Many of the forums I enjoyed in the past have become heavily burdened by rules, processes, and expectations. They are frequented by people who spend hours every day reading everything and calling out any misstep. Some of them are so overburdened that navigating all of the rules and expectations becomes a skill in itself. A single innocent misstep turns simple questions into lectures about how you’ve violated the rule…

The reason those rules are created is because at some point something happened that necessitated that rule. (Not always of course, there are dictatorial mods.) The fundamental problem is that communities/forums (in the general sense, e.g., market squares) don't scale, period. Because moderation and (transmission and error correction of) social mores don't scale.

> The reason those rules are created is because at some point something happened that necessitated that rule. (Not always of course, there are dictatorial mods.)

Maybe initially, but in the community I’m talking about rules are introduced to prevent situations that might offend someone. For example, the rule warning against using the word “blind” was introduced by someone who thought it was a good thing to do in case a person with vision issues maybe got offended by it at some point in the future.

It’s a small group of people introducing the rules. Introducing a new rule brings a lot of celebration for the person’s thoughtfulness and earns a lot of praise and thanks for making the community safer. It’s turned into a meta-game in itself, much like how I feel when I navigate Stack Overflow

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#204

Post by timhh » Sun, Oct 13, 2024, 6:02 PM UTC

hifromwork wrote: post #19

timhh wrote: post #8

Stackoverflow mods and power users being arseholes reduces the use of Stackoverflow. ChatGPT is just the first viable alternative.

>Stackoverflow mods and power users being arseholes reduces the use of Stackoverflow While they are certainly not perfect, they willingly spend their own spare time to help other peoples for free. I disagree with calling them arseholes.

The people I am referring to are not helping. At this point they are making SO worse.

The problems are two-fold:

1. Any community with volunteer moderators attracts the kind of people you don't want to be moderators. They enjoy rigidly enforcing the rules even if it makes no sense.

2. There are two ways to find questions and answer them: random new questions from the review queue, and from Google when you're searching for a problem you have. SO encourages the former, and unfortunately the vast majority of questions are awful. If you go and review questions like this you will go "downvote close, downvote close, downvote close". You're going to correctly close a load of trash questions that nobody cares about and a load of good questions you just don't understand.

I've started recording a list of questions I've asked that get idiotic downvotes or closed, so I can write a proper rant about it with concrete examples. Otherwise you get people dismissing the problem as imaginary.

These mods now hold SO hostage. SO is definitely aware of the problem but they can't instigate proper changes to fix it because the mods like this situation and they revolt if SO tries to remedy it.

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#205

Post by bee_rider » Sun, Oct 13, 2024, 6:02 PM UTC

internet101010 wrote: post #110

tlogan wrote: post #67

Earlier quoted context omitted.

The main issue with Stack Overflow (and similar public Q&A platforms) is that many contributors do not know what they do not know, leading to inaccurate answers. Additionally, these platforms tend to attract a fair amount of spam (self promotion etc) which can make it very hard to find high-quality responses.

Medium is even worse about this. It's more self-promotion than it is common help.

QA platforms and blogging platforms both seem to have finite lifespans. QA forums (Stack Overflow, Quora, Yahoo answers) do seem to last longer, need to be moderated pretty aggressively unless they are going to turn into homework help platforms.

Blogging platforms are the worst though. Medium looked pretty OK when it first came out. But now it is just a platform for self-promotion. Substack is like 75% of the way through that transition IMO.

People who do interesting things spend most of their time doing the thing. So, non-practicing bloggers and other influencers will naturally overwhelm the people who actually have anything interesting to report.

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#206

Post by bee_rider » Sun, Oct 13, 2024, 6:03 PM UTC

n_ary wrote: post #144

internet101010 wrote: post #110

Earlier quoted context omitted.

Medium is even worse about this. It's more self-promotion than it is common help.

Begin rant. I don’t want to be that guy saying this, but 99% of the top results on google from Medium related to anything technical is literally the reworded/reframed version of the official quick start guide. There are some very rare gems, but it is hard to find those among the above mentioned ocean of reworded quick starts disguised as “how to X”, “fixing Y”. Almost reminds me of the SEO junks when you search “how…

Somebody who just summarizes tutorials can write like 10 medium posts, in the time it takes an actual practitioner to do something legitimately interesting.

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#207

Post by Ekaros » Sun, Oct 13, 2024, 6:03 PM UTC

elicksaur wrote: post #155

PhilipRoman wrote: post #55

Earlier quoted context omitted.

I haven't really found stackoverflow to be that humiliating (compared to some IRC rooms or forums), basic questions get asked and answered all the time. But the worst part is when you want to do something off the beaten path. Q: how do I do thing X in C? A: Why do you need to know this? The C standard doesn't say anything about X. The answer will depend on your compiler and platform. Are you sure you want to do X ins…

On the other hand, I find it to be a fatal flaw that LLMs can’t say, “Hey you probably don’t actually want to do it that way.”

I always wonder about that. Very often it seems that you need to be able to LLM that they are wrong. And then they happily correct themselves. But if you do not know that the answer is wrong how can you get correct answer?

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#208

Post by mvdtnz » Sun, Oct 13, 2024, 6:04 PM UTC

wholinator2 wrote: post #117

wokwokwok wrote: post #109

Earlier quoted context omitted.

Everyone has a pet theory about what’s wrong with SO; but here’s the truth: Whatever they’re doing, it isn’t working. Blame mods. Blame AI. Blame askers… whatever man. That is a sinking ship. If you don’t see people complain about SO, it’s because they aren’t using it, not because they’re using the search. Pretty hard to argue at this point that the problem is with the users being too shit to use the platform. That’s…

We're talking about stackoverflow right? The website is a veritable gold mine of carefully answered queries. Sure, some people are shit, but how often are you unable to get at least some progress on a question from it? I find it useful in 90-95% of queries, i find the answers useful in 99% of queries that match my question. The thing is amazing! I Google search a problem, and there's 5 threads of people with comparab…

Nobody is criticising the content that is on the site. The problem is an incredibly hostile user base that will berate you if you don't ask your question in the exact right way, or if you ask a question that implies a violation of some kind of best practice (for which you don't provide context because it's irrelevant to the question).

As for the AI, it can only erode the quality of the content on SO.

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#209

Post by CogitoCogito » Sun, Oct 13, 2024, 6:05 PM UTC

jancsika wrote: post #181

haolez wrote: post #72

Earlier quoted context omitted.

The first time that I asked a question on #cpp @Freenode was a unique experience for my younger self. My message contained greetings and the question in the same message. I was banned immediately and the response from the mods was: - do not greet; we don't have time for that bullshit - do not use natural language questions; submit a test case and we will understand what you mean through your code - do not abbreviate…

This being HN, I'd love to hear from one of the many IRC channel mods who literally typed (I'd guess copy/pasted) this kind of text into their chat room topics and auto-responders. If you're out there-- how does it feel to know that what you meant as a efficient course-correction for newcomers was instead a social shaming that cut so deep that the message you wrote is still burned verbatim into their memory after all…

> was instead a social shaming that cut so deep that the message you wrote is still burned verbatim into their memory after all these years?

Maybe that was the point?

Re: Large language models reduce public knowledge sharing on online Q&A platforms

#210

Post by Ferret7446 » Sun, Oct 13, 2024, 6:05 PM UTC

irunmyownemail wrote: post #164

stickfigure wrote: post #133

Earlier quoted context omitted.

> The problem is eventually what are LLMs going’s to draw from? Published documentation. I'm going to make up a number but I'll defend it: 90% of the information content of stackoverflow is regurgitated from some manual somewhere. The problem is that the specific information you're looking for in the relevant documentation is often hard to find, and even when found is often hard to read. LLMs are fantastic at reading…

Published documentation has been and can be wrong. In the late 1990's and early 2000's when I still did a mix of Microsoft technologies and Java, I found several bad non-obvious errors in MSDN documentation. AI today would likely regurgitate it in a soft but seemingly mild but arguably authoritative sounding way. At least when discussing with real people after the arrows fly and the dust settles, we can figure out th…

Everything (and everyone for that matter) can be and has been wrong. What matter is if it is useful. And AI as it is now is pretty decent at finding ("regurgitating") information in large bodies of data much faster than humans and with enough accuracy to be "good enough" for most uses.

Nothing will ever replace your own critical thinking and judgment.

> At least when discussing with real people after the arrows fly and the dust settles, we can figure out the truth.

You can actually do that with AI now. I have been able to correct AI many times via a Socratic approach (where I didn't know the correct answer, but I knew the answer the AI gave me was wrong).