While I'd say this is currently true, answers on SO tend to stick around for a long time, and the 'marked as duplicate'-heavy behavior means that even if the best practice to solving a given problem changes, the top SO google result won't. While this isn't a _huge_ problem right now, with the way the site is set up I only see it getting worse, to the point that some years down the line I imagine many SO answers will either cease to be relevant or push people down the wrong path when solving a problem.
The only wiki answers I've seen have been hugely popular with dozens of answers and hundreds of votes. Maybe it happens outside of that, but it's definitely not how it "usually" works, IME.
Most of the questions I see on SO have maybe 0-5 answers, and the average number of votes is something like 2. If you're lucky there's an accepted answer and/or an answer with a clear majority of votes. I've even seen cases where the accepted answer is wrong, while a more upvoted answer is correct.
Organizational and writing skill don't necessarily correlate well to programming skill. And when one adds gamification to that allowing for material curated by the masses without the corresponding community investment by those who _do_ have that skill set... well.
I've done it on a few other answers too, if the top answer seems clunky or inelegant, I tend to dig a bit deeper to see if there's a better way.
Also, scrolling down a bit in the answers you usually find new ways of doing something as answer 3 or 4. Just upvote it. Or edit the answer into the top answer.
And don't think those editions are as high quality as the ones that come by the official channel, where literally anybody in the world with the needed knowledge can reply, and there is a robust process for vetting.
Besides, despite any theoretical possibility, people mostly do not edit old SO answers. There are exceptions, but outdated answer mostly stay there.
I Google incredibly basic stuff like how to read files, how to read a web page (something that changes every 5 months in .Net), hell, even basic jQuery functionality.
The difference is that I have the knowledge deep down, I just use SO and Google as a prompt. I know the correct answer as soon as I see it. I get exposed to a lot of basic answers on SO, and generally they're right.
SO is generally good and is not suffering the problem you or the OP claim.
You're absolutely wrong in that case. Particularly answers being "good" or "correct" is not part of the mission statement; and the rules reject fixes of questions by default for not being "in the spirit of the poster".
All SO aims for is having the most questions/answers. Nothing else.
> All SO aims for is having the most questions/answers. Nothing else.
Is demonstrably false. If that were the only aim, questions would never be closed and there would be no need for a moderation system. They fact that they do have moderation system and do police questions and answers is the proof that quality is part of the equation.
It does nothing to affect the quality of answers that look ok, but are actually problematic. And in those cases "spirit of poster" always trumps correctness.
Nope you can't.
- You'll be blocked if your edit is not "in the spirit of the originally post(er)".
If you only have to add a separate and standalone paragraph at the end on a newer subfeature (like your asp example), it's generally fine.
If you have to rewrite one of the main paragraph, the reviewers will systematically reject your edit.
Typically, that means that there is no way to update an obsolete post when the tools or best practices have changed, because it's considered major changes diverging from the original author.
- Combine that with the aggressive policy of closing and locking down questions, and it's impossible to post a new separate answer on most of the questions.
For the first time in a while, I've picked up a language I have not used for almost 15 years.
Google is helping immensely, and I've been shocked at how often (>95%) of the time the first result is SO, and that answer is perfect. So much so I've stopped searching google, I've just got a lot of SO tabs open.
It might not be perfect, though I'd say it's the best by a massive margin.
I don't know how many times I've scanned down a SO page hoping, hoping, hoping that someone managed to smuggle in the answer to my question before the red guards noticed and put a stop to it.
Google simply wants to find the best answer it can to my question, while StackOverflow would rather try to find reasons to prevent it being answered.
I don't want a forum where answering my question is a last resort, and their highest priority is finding a justification for not only not answering, but actually preventing anyone else who was willing to answer from answering.
What we need is a developer forum that is totally open to questions, but with the amazing addition of Google-like search to do ranked prioritization of all answers instead of old-fashioned red guards to prevent answers they didn't like from ever existing, whether you needed those answers or not. You could ask anything you wanted, as long as you were polite, and the worst that could happen would be that your question would be re-tagged (ex: that's actually a DELPHI question, not PASCAL language, because it's not about the language.)
People would then be free to answer or not, as they wished, with votes for answer quality and links to other answers that add to the quality measures. They could express opinions (horrors! opinions from people with experience that you lack! who would EVER want that) or quote the docs, or insert a link to another answer that (in their opinion only) probably answers your question, too, but they do so without preventing anyone else from going ahead and offering their own answer.
Google manages to sort through an entire web full of comments that are "possible duplicates" or "primarily opinion based" and it's amazingly useful. A mini version of that is what we need for software Q%A.
If Stack Overflow really wanted to make life easier for people, they would delete closed questions, or move them to a directory that is excluded from search engines. Closed questions are basically bait-and-switch for searchers; they show the question, but don't have an answer.
So whilst I'll often choose an SO site I can't use their search.
Try "KDE notifications from bash scripts" for example. Google will show that there are answers on SO and Unix.stackexchange - searching at AskUbuntu (I use Ubuntu with KDE on top) will return zero results and won't even suggest results from other SO sites. SO itself has the best result but seems the least appropriate of the silos to place that search in.
Then there's having to register at each of the different sites on the network. The siloing of the site is PITA IMO. The only reason to do it this way I can think of is to separate computing resources, it seems there are probably better ways around that issue too? Why not a unified site with hierarchical tags such that you can choose a tag as a preferential search term?
I have 10+ memberships at Stack Exchange network most of them math/programming related and it seems so silly.
Sure, keep the multiple brands if that is what your investors want but make it searchable and taggable across sites.
Often I will see a question posted on main SO that would have been perfect by the Stack Exchange rules on some other stack exchange site.
Instead it just gets closed and whoever asked must feel quite dispirited of asking again.
with all due respect but SO search isn't quite stellar.
Point is, it's good resource to free-ride but not actively participate.
FWIW it wasn't always like this. I have a lot of points on SO mostly acquired about 5 years ago. Back then SO was a much friendlier, and much more interesting place to participate in. I haven't actively participated in about 3 years or so.
Anyone who has had to work with others in tech knows the kind of crap you get. They don't give you the stack trace, they say 'I get an error' without giving the error message, they just say 'it doesn't work', it's not clear what they want, they post a screenshot of half a stack trace, etc...
Again, it's a constant trade off where you are balancing the asker's needs with the thousand people who find this via Google, and the people answering the question.
I can often stick with second best in class as long as it doesn't have a habit of biting you for no apparent reason all the time.
> People have just forgotten that the lack of that carefully curated environment created the complete mess that was all the forums we used to have before.
People keep saying that and I keep wondering if we either never visited the same forums or if I'm somewhat blind to all the vile bullying that must have been going on.
Edit: broaden answer with first quote and answer.
It wasn't bullying so much as low-quality content. It is quite likely you didn't have this problem, as I didn't very often either. If you know where to go, help has always been available for many languages/runtimes/environments/etc., and generally that help still is available today. If you know the right IRC channels, the right forums, the right mailing lists, the right other resources, you can get great, high-quality help.
However, prior to SO if you just poked your query into Google you ran a decent risk of getting something really crappy, laden with ads, locked behind obtaining accounts (to get your email for spam) if not outright paywalls, or expertsexchange. And this was more true for beginners; I didn't encounter this stuff while looking up an error message resulting from an obscure combination of hardware under Linux, I encountered this stuff while trying to refresh my memory about some particular Javascript intricacy.
Yes, it really was bad for a lot of people. SO is still an improvement inasmuch as while the community may rub some people the wrong way, at least the answers are still there for free, uncluttered, and generally of high quality for the most-commonly searched things.
I'm not saying there's any generic forum better than SO out there. But they are not that far behind anymore.
The years immediately pre-SO were the height of the 5 page adsense spam site. Put anything into google and you ran a decent risk of getting something really crappy, laden with ads.
Compare with the complete shitshow you can still see on some of the survivors of the old era, like tomshardware, Msdn, and others.
On the other hand, I've had a lot of success with SO, both asking and answering, and while there are occasions where I find questions closed as duplicates or off-topic, they are pretty rare. I frequent the Java, Docker, Kubernetes, jclouds and similar categories, and they are generally very good.
It may be down to the choice of topic? Perhaps PHP and .NET have more first-time users, and a lower quality of questions? I don't know; although since the SO data is all freely available [1] it might make a nice research project to discover [2] what causes these questions to be closed!
[1] https://archive.org/details/stackexchange [2] https://data.stackexchange.com/stackoverflow/query/540597/ov...
What Experts Exchange had going for it was a lot of corporate-domain programming knowledge, stuff like COBOL and IBM Visual Age that are basically unknown on Stack Overflow.
You certainly never visited c.l.p.m back in the day, or you'd know exactly how bad the alternative could get.
I wonder how many people Abigail alone drove away from perl?
It wasn't, the issue was the content was useless. To find anything you had to wade through pages of off-topic stuff, and try to get a sense of the value of the given answers. It was terrible. That and the paywalling and other stuff that used to go on.
I don't know, I've always felt that the apparent decline in usefulness came from the strict curation (and I do insist on the word "apparent"). In the past the site felt more natural to me as a big hodge-podge of various technology related topics. Questions came from various aspects of the tech world, whether they related to sys admins (e.g. linux configs, installations, scripting, etc), super users (e.g. text editors and other tools), programming languages (e.g. how to use a library or language feature), comp sci (e.g. algorithms), databases, or even networking. To me it all kinda related to how we all use the set everyday. Then came what I can only describe as over-categorization. Instead of relying on the machine to search stuff for us, they split the content into specific websites and changed their standard of what's relevant or less so (which is why you have closed questions that thousands of users still find useful). Google figured this out years ago with gmail, don't waste time organizing, just search. Case in point, how many people actually search any of the stackexchange websites directly? I'd bet that most users get there from google. And it's not unusual to be searching for a topic that intersects with any two or more of these communities. But none of them will refer to any other to tell you that the topic has a related thread in a sibling community. Only google gives you a clue, because it looks at it as the afore-mentioned hodge-podge.
It's not to say that the site didn't need any curation, I just think it somehow became a bit too systematic and oblivious of our instinct as human users. I also have the strong suspicion that people who feel that SO is declining are way off, but I doubt that the rise (if any) of its relevancy has much to do with the heavy handed curation.
> The consequence thereof is that many good questions not only get closed before anyone is able to respond, but that many of them end up vanishing into oblivion for eternity after merely 9 days.
Yes, any cutoff time has the downside that you lose posts. But it's not like SO makes rules for no reason. Where's the discussion of the problem that that rule is solving? How do we know they're not worse than this problem?
For anything involving Linux, the Arch wiki is light-years ahead. It's not even a close competition.
Also, the way in which Arch is as a system makes it less likely you have that kind of problem. Because you install and configure everything yourself, there is less 'magic' that is likely to go wrong in a way you don't know how to begin fixing it.
- The SO was great because answers for noob questions where hard to find. At this point of time (thanks to SO and their knowledge database) there are so many other resources to get these noob questions answered or they are already answered. And some of these other resources are updated to reflect the most recent technology changes while SO answers are static.
- Now people start asking hard stuff - but engineers (I intentionally do not say developers) who know answers on these questions are not interested in answering.
- People who got reputation and they are moderator (they made SO successful) do not know what they do not know.
I can't speak for others, but in my case it is not for lack of interest. I don't have enough 'reputation' to be allowed to answer questions and am too busy to find earning it a good use of my time.
You don't need reputation to ask or answer questions, so this is FUD.
So probably mistaken, but hardly FUD (fear, uncertainty and doubt).
I suspect part of the reason they're not interested in answering is because a difficult question requires a lot more work to likely get less points as a result. Meanwhile, answering a ton of easy questions can be done quickly and gets more points over on Stack Overflow in the same time frame.
That and yeah, lots of people who could answer your question would rather be doing something that pays the bills or helps them instead.
I have to admit I have fallen into this myself on both sides.
The (now) rare times I browse the site to find questions I can answer, when I run across one like this, its usually by a low rep user, so I think "hmm, I'll spend fifteen or twenty minutes sharing this, get maybe one or two upvotes, and if I'm lucky the original person will accept the answer" and usually deem it not worthwhile. Partly this is because the correct answer is often "don't do it that way, you should do x then you can do z and completely avoid your problem instead" which understandably people don't like to hear.
What is that?
SO is great to search for snippets and tl;drs, though. Reference manuals fail to contain tl;drs way too often, and that was fixed, thanks SO.
http://meta.stackexchange.com/questions/19665/the-help-vampi...
A common variation is someone who asks a reasonable question, gets an answer, then repeatedly modifies/extends the original question, or who asks a long series of follow up questions n comments. This variation is described here:
It's a sliding scale. Most people just don't know how to ask a good question initially, but some just don't care. In general, I'd say the 'help vampire' problem is where the asker is looking for the answerer's time, not their help. They want someone to do free work for them, without putting in effort themselves.
I believe this is the origin: http://slash7.com/2006/12/22/vampires/
1. Earn some reputation first in order to see how stuff works from the other side. When you join a new company, you don't come the first day and tell everyone how they're doing it wrong before learning about the project, about the tradeoffs etc. Same here.
Having rep above certain threshold gives you additional privileges (commenting, editing etc.).
Also, when you have a rep over few hundred, I'm more likely to even read your question. Most questions from 0-rep users are very basic same-stuff-again / low-quality questions that are not challenging to answer.
Moreover, when you have some points, you can put a bounty on your question, which means you really want to have a good answer for that and can make a sacrifice, and makes the question stand out (on a separate bounty-enabled-questions page). This gives more incentive to others to look at your question. There's nothing more soul crushing on SO than taking time to write an elaborate answer and getting 0 upvotes, because the guy forgot to upvote or whatever because he doesn't care.
2. It's a misconception that you need to "respond very fast" to the questions to get the rep. Everyone has a different strategy, and well I spent some time a few years ago using this strategy to get some initial reputation. But I stopped some 5 years ago. I almost never play the "fastest gun in the west". I google things that I encounter during my normal daily work, and keep the tab open. If the best answer is not satisfactory, and later I come up with something better, I write it down as a new answer, and perhaps if I found similar problem in multiple SO questions, I put a comment with a cross-link to my answer in other thread.
You really need very little effort to get SO rep as a part of your normal daily work, yet everything I talk to seem amazed "wow how come you have so much rep on Stack!". Just create a damn account, be always logged in, and don't keep your solutions for yourself.
3. Try to put up an exhaustive question, write down a SMALL, reproducible scenario, explore some options yourself (ask yourself "what did I try?"). A lot of newcomers put very vague questions, very little information, it's not fun to ask someone a ten clarifying questions.
4. I felt really bad when I got my first downvote but it was indeed a bad answer, to a different question than the OP asked. I learnt the lesson and I don't read in haste anymore. It's good to be humbled sometimes.
Regarding "closing as duplicate", it's true some mods are too eager, but if it makes sense, I believe dupe questions should be closed - duplication is not good (DRY) and best code is no code, if possible.
For example having a question incorrectly closed as duplicate is one of the big problems, yet you only address it dismissively in a single sentence.
In general, you make reasonable points - although it's clear SO could improve here. The new asker experience obviously has problems. Not worth sacrificing question quality to improve on, but I have no doubt there will be ways to help without doing so. (And they seem to be working on it).
The 'interesting threads' are exactly the problem. Those are discussions and they don't suit the format. If you want to talk about something, find a forum. SO does not have threads, it has questions and answers, and that is why it's useful.
Turns out people hate wading through pages of outdated discussion. People are forgetting what it was like before SO. Stop expecting SO to be all things to all people - if you want a discussion, go register at a programming forum. SO does what it is trying to do well.
Sure, it's not ideal - SO could do better in terms of helping people understand the site's goals and enforcing the rules in a less hostile way, but they are working on that (a lot of the more hostile rule enforcement tropes are banned and filtered against), and it's a hard problem.
SO isn't dying any time soon, and the content is still good. If you want to kill it, please go ahead and solve the issue of explaining to users how to contribute quality content and getting them to take that in instantly.
I used to contribute a lot, which tapered off as I had other things filling my time. Towards the end though, the help vampires were getting to me, and I understand why people are harsh towards new users in some situations, it's an easy trap to fall into. Trying to fix that problem by stopping the curation of the content is insane, however. That's a fast route to going from some people being turned away to having no decent content.