What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 134 Stack Exchange communities.

I've seen the questions asking whether 48 hours is not too long, and why one can't just offer a bounty immediately. I would like to ask the community, at this point, a few years down the line, whether you don't think that 48 hours is too long to wait for a bounty.

In the last few days, I've had two cases where I needed answers quickly (I still do), and it was apparent that it wasn't going to be answered by the normal process. I tried editing the questions frequently, and just ended up with a question with 6 upvotes, but no answers.

Is 24 hours not a more appropriate length of time for a bounty? I understand why it must not be immediate, but why does it have to be two days? That's a really long time to be stuck on a problem.

I guess one would hope that within those two days, the person will solve his own problem. But in cases where that's not as easy, and you really need the experts to look at it thoroughly, two days is an excruciatingly long time.

share|improve this question
    
I think the idea is that, within 48 hours, it's still very likely that people will come across your question that haven't seen it yet, and will be able to answer it. –  Cody Gray May 3 '11 at 10:56
6  
And just in case you don't know: voting on MSO often indicates that people disagree with the proposal. Though this is not tagged [feature-request], my downvote still indicates that I like to keep the 48 hours. SE is not a help forum, and I think allowing for earlier bounties has the risk of turning it into one. –  Arjan May 3 '11 at 11:05
    
@Cody but why should one be prevented from placing a bounty because of that? (I realize you're not advocating the argument, just assuming that it is the reasoning behind it - I just don't think it's good reasoning.) –  Pëkka May 3 '11 at 11:32
1  
@Pekka: Yeah, your edit is exactly what I was going to respond. I don't really have opinions on the bounty system. Mainly because I don't care enough about reputation for the existence of a bounty to actually compel me to answer a question. I don't watch the "Featured" tab. If I know the answer to something (or it looks like an interesting question), I answer it. Otherwise, I let others try to answer it. –  Cody Gray May 3 '11 at 11:49

3 Answers 3

Yes please!

If people want to place a bounty on a question that they really really want/need an answer for, why force them to wait? It's their own damn reputation they're spending.

The bounty system is riddled with too many limitations anyway. Let's get rid of some of them.

share|improve this answer
9  
+1 Bounties are often going to be spent when the user is desperate for an answer. Saying "sorry, you're not allowed to be desperate for 48 hours" seems cruel. –  Blowski May 3 '11 at 15:17

No, it shouldn't be.

My biggest objection to this is also my biggest objection to bounties period - they encourage spending over improvement.

That's a really long time to be stuck on a problem.

Nothing is stopping someone from answering it during that time, except the nature of the question itself. Have you responded to the comments looking for clarification? Edited in details on what your own research has turned up while you were waiting? Worked to make the description and / or code more concise to encourage more people to read it? Made sure the title actually describes the problem you're trying to solve?

If, in two days of effort, you can neither solve the problem yourself nor find anyone else willing to help, then fine - slap a bounty on it, sweeten the deal that you've already done everything to make as attractive as possible. But if you're sitting on your hands for two days waiting out the waiting period... Then you're wasting your time for nothing.

This isn't a "pay to play" system. One of the original goals for the creation of Stack Overflow was to create a place for the completely free and open exchange of information - in stark contrast to the prevailing Q&A system of the time which asked you to "grind for points so you can get your answers"... So if your first impulse after a day of not getting answered is "I need to SPEND SOME VALUABLE POINTS to get an answer then", we're failing in that goal.

If we're failing, let's fix whatever is leading to the failure. Not give up and make it easier to pay.


Returning to this question after reading comments, other answers, and this recent request for the ability to close bountied questions reminded me that the two-day waiting period serves another purpose: giving the community a chance to review and potentially close your question before a bounty offer makes that impossible without moderator intervention.

In response to other suggestions:

  • Predicating this on reputation (or jmort's suggested minimum offer) just gives the most "buying power" to the folks for whom reputation means the least. I could offer dozens of 500-point bounties on SO without losing a single privilege - where's the motivation to reserve them for questions that actually need them?

  • Requiring approval from high-rep users just creates another queue for folks to mind. If it takes two days for your request to make it through the queue, you haven't really gained anything - you've added complexity to the system for nothing.

Random observation: the tumbleweed badge is awarded to questions that sit around, ignored, for a week...

Related:

share|improve this answer
    
The irony of answering a question on making bounties easier in response to a bounty being put on it isn't lost on me... Sadly, Paul could've just edited the tags and bumped it without spending anything and improved the question while doing so - I don't check the "Featured" list on MSO anyway. –  Shogging through the snow May 19 '12 at 18:33
4  
"Nothing is stopping someone from answering it during that time, except the nature of the question itself." That presupposes a fact not in evidence: that the reason the question isn't being answered is because it's a poorly phrased question. Sometimes this is true. But sometimes it isn't. I see no reason to argue against this just because it sometimes happens. –  Nicol Bolas May 19 '12 at 19:22
2  
"If, in two days of effort" That assumes that Stack Overflow is your first stop when dealing with a problem. In general, if I'm asking a question, it is because I cannot otherwise make progress. I've exhausted all other options and SO (or a forum or wherever) are all I've got left. Yes, for some people SO is the first resort. But quite frankly, it shouldn't be. And again, it is not a fact in evidence that everyone uses SO that way. –  Nicol Bolas May 19 '12 at 19:24
2  
"So if your first impulse after a day of not getting answered is" What alternative do you have? If you've already done your diligence in asking a proper question, in working on it yourself before bothering us with it, and so forth... why should you have to wait two days to post a bounty? And if you haven't done diligence... guess what? Having to wait that extra day isn't going to make someone do what they should have done. –  Nicol Bolas May 19 '12 at 19:27
2  
@Nicol: Why should you have to post a bounty at all? Why should that even be an option? Yes, I'm making a lot of assumptions here - but they're all predicated on the idea that you're using bounties for questions where you've exhausted every other avenue available to you for getting an answer. "Let folks pay to get answers quicker" was never the intention, and IMHO is an extremely poisonous idea. –  Shogging through the snow May 19 '12 at 19:58
2  
@Shog There is a large period of time between when a question stops getting much traffic, and when you can post a bounty. Constantly writing a rewriting a question (or fiddling with tags etc) does not provide value once you've solved the initial problems (which I try to do before I submit the question). –  Paul Biggar May 19 '12 at 23:45
    
@shog: I dont have to post a bounty, but it is sometimes important to get an answer quickly, which is why I use SO in the first place. –  Paul Biggar May 19 '12 at 23:46
2  
@Paul: again, bounties were never intended to allow ANYONE to pay for quick answers. The intention was to provide a way to make time-consuming questions - the sort that can't be answered quickly - a bit more appealing to answer. That's why they're active for up to a week... I'd definitely argue we don't do enough to make this clear in the UI, but the solution isn't to further muddy the waters by implying that if you haven't gotten an answer in a day the gravy train's passed you by and it's time to pay up. –  Shogging through the snow May 20 '12 at 21:37
1  
@Shog I still don't see the point in the time limit. Yes, bounties are for the hard questions, and sometimes there are hard questions that I'd love to have an answer to really, really fast. (Because usually, there's something that urgently needs doing that is being held up by the problem.) I usually know such a hard question when I ask it, because I'll have done a lot of fruitless research beforehand. That's why I turn to SO. So why are you making me arbitrarily wait for 48 hours until I can put money on the table? (...I love how I can blame you for everything now that you're an employee.) –  Pëkka May 20 '12 at 22:39
4  
... I say make it six hours. Or twelve if you must. I think that is enough to fulfill the goals that the time limit has. –  Pëkka May 20 '12 at 22:41
1  
@Shog9: I believe you, but there's a big gap between impatiently needing it now, and having to wait two days. A two day gap in fact. Why not make it one day? In my experience, if you don't get an answer in the first 6-12 hours, you're not getting one in the next 2 days either. –  Paul Biggar May 21 '12 at 1:35

I see the point in the 48 hour wait. We definitely don't want to see a featured tab that is filled with a series of low quality posts that are unclear, incomplete, or unanswerable simply because of missing details.

One of the things that I enjoy about the bounty questions is that there is generally lots and lots of good details in the questions, and the questions themselves generally produce really great, detailed, quality answers.

In this regard I strongly feel that the system isn't broken and is working as it should. If the threshold were to be lowered to 24 hours, would that lower the quality of these questions or create a series of frivolous bounty questions that really aren't worthy of a bounty? This is a very valid concern and one that should be considered carefully.

It would be a tragedy for the reputation system to suddenly be diluted by users who give normal, one or two sentence, hit and run answer on categories of questions that have traditionally been answered with great care and attention to detail.

How to make this happen without sacrificing quality?

However, StackOverflow is a system that is built on the concept that the more reputation you gain on this platform, the more responsibilities and trust you earn.

Thus, here are a few possible proposed solutions that may mitigate the very valid concerns of lowering the threshold to 24 hours, while still providing the community with the option to post a bounty at the 24 hour mark:


1 - Require that a user have at least 10k reputation (or 3k or 5k) in order to post a bounty prior to 48 hours.

These users know -- or should reasonably know -- what constitutes a good question and will more than likely only post bounties on good questions.


2 - Limit the posted bounty amount to no less than 500 reputation. If the question isn't worth a bounty of 500 reputation, then the asker can wait until the 48 hour mark.

Desperate people should theoretically be more willing to sacrifice more in exchange for their desperation. This will eliminate frivolous bounties posted by users who just need to avoid waiting until the last minute (because of lack of planning or preparation) and reserve this type of bounty for those unique situations where a bounty would really help.


3 - Require that a user convince a 3k or 10k user approve bounties posted prior to the 48 hour mark.

Again, this could help provide the necessary community controls, which work really well for closing and other moderation issues, to effectively prevent abuse of the bounty system. Posting a bounty prior to the 48 hour mark could for instance require a 3 to 5 vote approval by high rep users.

Additionally, another thing that could help ensure these users don't just approve any bounty is to consider removing 2 reputation points -- from the users who approved the bounty -- for every bounty that isn't awarded by the person posting the bounty. In other words, 3k and 10k users will only approve the bounty if the question is clear enough to where the question-asker will likely get a good, solid answer. At the same time, the reputation hit won't be so significant as to be more than a simple downvote.

NOTE: I don't think all three of these need necessarily be implemented in order to make this a success. Any combination of these 3 suggestions could be beneficial and allow for a special 24 hour threshold.

share|improve this answer
    
Seems like a very good middle ground! –  Paul Biggar May 25 '12 at 23:51
1  
@PaulBiggar - Instead of saying no, sometimes I think we can find solutions that work for everyone, or at least the majority, while still keeping in line with the StackExchange vision. Additionally, taking steps like this can allow SE to "ease into it" so that it's easier to back out of if the experiment fails. –  jmort253 May 25 '12 at 23:58

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .