Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Fall of Stack Overflow

observablehq.com

Post reply on HN
Print view View original

41–50 of 808 posts

Re: The Fall of Stack Overflow

#41

Post by grecy »

ajsnigrutin wrote: post #33

"How to do A?" Comment: "You shouldn't do A, it's better to do B" Closed, duplicate of "How to do C". Yeah sure, in the beginning, there were many more basic questions (how to increase number by 1, how to get division remainder, how to check if file exists,...), and if you're coding in Perl, you can still find all the answers... if you're working in python, you'll find answers for python2, some for python3, some spec…

Echos my experience perfectly. It's actually impossible to ask a question and get an answer, also impossible to ask a clarifying question on an existing question/answer.

That makes it much, much less useful than it otherwise could be.

Re: The Fall of Stack Overflow

#42

Post by rtpg »

dylan604 wrote: post #12
hirundo wrote: post #2

This is a large decline. It does not correlate well with the rise of LLMs, which have shifted my own habits away from Stack Overflow recently. Looking through the graphs, the most dramatic shift was the large drop in new visits in the first half of last year, by two thirds or more. That seems late to be a Covid effect. There were some political controversies, but it would surprise me if they had such a large effect.…

You trust LLM responses? I feel they are not worth the paper they are written on

"ask LLM, then look up the language/library docs to see if the idea makes sense" has worked relatively well for me.

Re: The Fall of Stack Overflow

#43

Post by maxlin »

Even though StackOverflow in the common use case has been taken over by ChatGPT, I sincerely hope it keeps operating, stays strict (even if it causes collateral) and keeps ban on LLM-generated content.

The wheels of this kind of stuff turn slowly, but obviously ChatGPT was trained partly with data only gainable from a healthy StackOverflow-kind of site with users actively asking unique questions and enough people answering those unique questions with well-though-out answers. The shittyfuture outcome is that StackOverflow goes out of business and LLM's stagnate on this front, while being capable of answering in fuwwy speawk when prwompted, would still be limited to / biased towards older versions of libraries, software, languages, tools etc.

Re: The Fall of Stack Overflow

#44

Post by morkalork »

ajsnigrutin wrote: post #33

"How to do A?" Comment: "You shouldn't do A, it's better to do B" Closed, duplicate of "How to do C". Yeah sure, in the beginning, there were many more basic questions (how to increase number by 1, how to get division remainder, how to check if file exists,...), and if you're coding in Perl, you can still find all the answers... if you're working in python, you'll find answers for python2, some for python3, some spec…

Perl is a good example because it's stagnant. Googling Python or Java? If the result is up to date, it's a low quality content farm. If it isn't from a content farm, it's a page from 2008 that is no longer relevant at all!

Re: The Fall of Stack Overflow

#45

Post by minimaxir »

MontyCarloHall wrote: post #5

Where were these data sourced from? As far as I know, StackOverflow does not publicize internal analytics with such granularity. If these figures are real, are they leaked?

Stack Overflow used to release their data archives quarterly on BigQuery. Looking at the BQ datasets, they were last updated Nov 2022, which doesn't have the latest 2023 info in the submission.

Re: The Fall of Stack Overflow

#47

Post by julianeon »

krackers wrote: post #28
fishtoaster wrote: post #20

Interesting that this seems to have started around Spring 2021 when posts + votes started tailing off, followed by traffic starting to decline around spring of 2022. I can think of a few theories that I don't think hold water: 1. The rise of ChatGPT to answer many questions that StackOverflow would previously have been used for. This seems unlikely, since the timing doesn't really work out. 2. The perennial complaint…

SEO games resulting in SO clones ranking higher than SO itself?

As much as SEO experts would love to do that for their clients, I don't see any way they could've realistically crossed that moat. If you had a programming question, and you went to Google to answer it and found some answers - would you choose Stack Overflow, or PythonIsSuperFun.com? I think the behaviour of Google searching for programming issues decreased overall, and I don't think Stack Overflow suffered more than any other site. Disclosure: I've nearly abandoned SO for ChatGPT myself.

Re: The Fall of Stack Overflow

#48

Post by ajsnigrutin »

morkalork wrote: post #44
ajsnigrutin wrote: post #33

"How to do A?" Comment: "You shouldn't do A, it's better to do B" Closed, duplicate of "How to do C". Yeah sure, in the beginning, there were many more basic questions (how to increase number by 1, how to get division remainder, how to check if file exists,...), and if you're coding in Perl, you can still find all the answers... if you're working in python, you'll find answers for python2, some for python3, some spec…

Perl is a good example because it's stagnant. Googling Python or Java? If the result is up to date, it's a low quality content farm. If it isn't from a content farm, it's a page from 2008 that is no longer relevant at all!

> stagnant

I would call that "stable" :) Or "good enough, that it doesn't need constant fixing" :)

Re: The Fall of Stack Overflow

#49

Post by l0b0 »

NotYourLawyer wrote: post #7

All internet forums tend toward shit as time goes on.

I agree, but I'm afraid the downvotes are inevitable. SO for a long time had by a very long shot the best signal-to-noise ratio of any free-as-in-beer Q&A channel. Basically any live comms channels (IRC, Matrix, and so on) disincentivise clear, thoughtful responses in favour of streams of consciousness. And forums is where information goes to die, as every question turns into a huge back-and-forth until the question is sufficiently refined to receive a useful answer, which is usually extremely specific ("run this command") rather than generally applicable ("you need to frobnicate the foon, for example by running this command").

Unfortunately anything popular is a target, and once they stopped trying to innovate to keep ahead of systems gaming they were going to lose eventually. It's amazing it took this long.

Re: The Fall of Stack Overflow

#50

Post by ourmandave »

smeagull wrote: post #21
readingnews wrote: post #3

I was promoted to one of those "mod" type people who reviews new answers, first posts, etc, a few years back. I can tell you that I quit in the last 6 months. All new answers are answers on questions from first time posters, answering questions that have good solid answers from 8+ years ago (mostly 10 years), and it is pretty obvious, they are just doing it to get points. The question/answer is not something that has…

> answering questions that have good solid answers from 8+ years ago (mostly 10 years), I hit upon those useless answers all the time. I have no need to know about solutions that might have worked a decade ago and are obsolete. Stack Overflow is useless now.

Yeah, except for those of us working on old software that still runs the business.
Post reply on HN