Comment by weinzierl | Hacker News Reader

Comment by weinzierl | Hacker News Reader

> A while ago I started writing a similar post, but it got longer and longer without coming to any conclusion. I’m writing this one with a timebox of one hour, and then I’ll post whatever I’ve got. (I may then reformat it later.)

While this is meta it pretty much sums up the situation regarding my feelings regarding Stack Overflow.

I was an early user of Stack Overflow, not quite beta but almost. At one point I even received a snail mail letter from Jeff Atwood, thanking me for my contributions. I read most of the early discussions about policy and, being a veteran of various internet forums at that time, it seemed to me the most sensible, most reasonable approach you could take to running a question-answer site. I went to CodeKen in London in 2011 where I met Jon and a lot of other excellent folks that made Stack Overflow happen as a community back then.

Today, I rarely consult Stack Overflow, and when I sometimes find a useful answer via DuckDuckGo it's almost certainly closed. I even joked that closing is a sign of quality at Stack Overflow. My last contribution is probably years ago.

The strange thing is that there is no point I "left", nothing I could point at and say, that's when Stack Overflow took the wrong turn. I think we just drifted apart. And for the conclusion I guess for me it's mostly: "The way to hell is paved with good intentions".

Is this a sign of strength or weakness though? Like Wikipedia, any given topic on SO will approach “completeness”. I use stackoverflow at least as much now as in its prime - but these days just like a Wikipedia for code. I don’t consider it a flaw.

I’m sure there are topics I don’t use (the latest ML framework or the js framework du jour) where the community is still growing the database. That I use it as a mostly completed reference work just means my field/tech is mature in terms of language/frameworks

It's really hard to consider an answer to a moving target to be "complete". If the question was answered for Rails 3 but you're using Rails 5 and have the same exact question, chances are that answer isn't going to work for you. Now you have to chance that if you post the question again, is someone even paying attention? Is someone going to close your question as a duplicate? Is someone going to give you the wrong answer?

Unless you're working with a dead technology, by the time the information is "complete", it's outdated. And I would consider much of the information on SO to be just that: outdated.

It’s an interesting problem that they need to tackle. Perhaps answers should have a lifetime? Perhaps the moderation queue should include reviewing (and e.g. tagging) questions marked “rails” to be “rails-3”?

I work with the same tech I did 10 years ago - so obviously I wouldn’t want the old stuff to go away and I probably often find answers that relate to too new tech more often than the opposite

We hadn’t quite figured out that opinions and discussion questions were not a good fit in the early days.

You mean decided.

The sibling comment is good enough and gets the point across. But I want to elaborate because I've ended up in basically the same place as the GP.

I know you've written about SO not being a good place for opinion-based questions before, and perhaps in theory you are correct, but I (at least) am not satisfied with the implementation, and tend to link that at some level to a flaw in the theory.

What I mean: some of SO's most [in]famous questions are opinion questions where the askers are requesting advice from "expert" answerers to help them navigate a new topic. They're not just infamous because they may be controversial and "hard to moderate". They're famous primarily because the way in which the question was asked drove many users (hundreds even thousands of upvotes with millions of views) to the site and helped a lot of people through their journey working on a new technology or in a new problem domain.

SO said "nahwp, we don't want questions without clear cut answers". Not necessarily SO's fault, but nonetheless their problem: the community took that as, "err on the side of annoying pedantic asshole and close, downvote, and otherwise ostracize any user asking any question that might not be a good fit for the site under the new regime".

In my experience, it feels like people are trying to find a way justify being annoying instead of find a way to provide meaningful advice. I've asked questions that, if you took 30 seconds to think about instead of a quick search for a marginally related title (which I already did in the 4 hours prior to asking my question) and immediate vote to close as duplicate, or an almost bot-like flag as "too broad" based on word choice heuristics, are not bad questions, and often after hours of flagging and arguing with the original users who are stubborn because they don't want to be wrong (especially publicly) I can make my case. But by that point it's too late and too many times I've had to do that to fend off users trying to do good for the site over helping me with a problem. And I'm not even asking opinion questions...

By making one small decision, you created two problems.

1. You actively choose to diminish the value to practical real users of the site by setting a long term vision of becoming some bastion of objective knowledge. We have those; they're encyclopedias.

2. You unintentionally validated the pedantic/authoritarian desire for and abuse of power that has deterred many of your original users.

Honestly most people can handle snarky comments and opinionated edged answers. Everyone (users of SO) has a tiny (and sometimes large) "programmer ego" inside of them. On SO you could responsibly exercise it. Not anymore. What I can't handle is nonsensical authoritarian and meaningless abuse of questions because they don't immediately and explicitly adhere to some SO agenda. The former can be mildly annoying at times but tolerable and hell even fun on occasion, the latter is frustrating and infuriating and something I cannot tolerate.

Sorry that really turned into a rant, didn't it?

I don't really have an opinion on the fit.

But people still ask those questions and the questions apparently get indexed. In my experience the result is that when coming from a search engine, even for non-fit topics, it is quite common that closed Stackoverflow posts occurr several times on the first result page.

It could be that I search for strange things, but at least for me the combination of closed but still indexed posts are not helpful, especially when the question didn't get any useful answers at all before being closed.

I don't mind closing topics, but I would prefer that closed questions with no even remotely useful answers wasn't indexed at all. Now, that might not be entirely under your control, but the issue remains, or so it seems to me.

Thank you for the letter and the stickers, I have one of them on my bicycle to this day.

> We hadn’t quite figured out that opinions and discussion questions were not a good fit in the early days.

This is something I had to learn as well, I guess one of my most upvoted questions falls well in that territory [1]. I just think that Stack Overflow was best when you were its primary moderator and it's arch enemy was Experts Exchange [2]. That battle is won, but the war goes on. Nowadays when the first search result is Quora I have very much the same feeling I had regarding ExpertsExchange back in the day and I can't help thinking that there is lost opportunity for Stack Overflow and Stack Exchange.

[1] https://stackoverflow.com/questions/961942/what-is-the-worst...

[2] https://blog.codinghorror.com/whos-your-arch-enemy/

Reply on news.ycombinator.com