As we observe the operation of the site we notice situations that appear to negatively impact the goals and purpose of it. For example downvotes have been disabled because they caused several situations where new users were discouraged, communities were split with half upvoting and half downvoting questions for no other reason than to support/disagree a point of view rather than the content. This behavior is well know and documented on other social sites.
Post closings seem to be the next example of good intentions occasionally going the other way. This applies to me as well, I have closed many posts that I later regretted, in fact in hindsight I can't see the value of any post closing that I've performed. So why do we even need them?
The concept of post closing has been imported from StackOverflow but the reasons for having that in SO are radically different. We need to recognize that the content produced at SO covers a radically different domain. The answer to How to sort a list in Python will not change from one year to the other. It is perfectly valid to use the answer to the old question and close every other such question as a duplicate.
Bioinformatics is different. It is a rapidly advancing field where the correct/recommended answers may change within a short period of time. The rationale of trying to maintain a single set of answers does not apply. Most importantly old and potentially highly voted answers may actually contain somewhat obsoleted advice. Fact is we don't go back to old answers and correct them, nor do people usually visit old questions to add new, updated answers (it happens but not it is not common).
So I propose to remove post closing. Only post deletion remains and should be applied as follows:
- If a post has no answers or comments then the original poster (author) may delete the post (this removes the post with no trace and no option of restoring the post)
- If a post is radically off topic with no possible bio-science connection then moderators should comment on it (to notify the author) then delete the post (this action can be reversed)
- If a post is spam then moderators should delete the post and suspend the user (both actions can be reversed)
Actions for posts that previously may have been closed:
- If the post is a duplicate question (identical to a previously asked and answered question) paste the link to the duplicate into an answer and let that be the answer.
- If the post appears to be slightly off topic: python programming, plotting with R, installation of a software tool, how to get started in science, please put a link to the more appropriate forum/resource into a comment or answer
I think what we need is not to disable post closing, but to set guidelines for admins or to restrict the permission to a few core admins. There are many admins on this site (how many?) and they (well, I should say 'we', as I am one of them) have different standards on closing posts. We are essentially closing posts based on the union of individual rules, which is the main cause of over-deleting posts IMO.
I prefer to enable downvoting. We get a better view of the quality of a question/answer with both up and down votes. As to your point "for no other reason than to support/disagree a point of view rather than the content", disabling downvoting does not help to improve the situation as we can still upvote to support "a point of view rather than the content". That is the limitation of the voting system, not the limitation of downvoting alone. The two major Q&A sites both have downvote enabled. I do not see why we are different.
To avoid hurting the feeling of new users, we can disable negative votes.
Btw, to improve the general question quality, I think it is very important to suggest related questions and suggest potential duplicates as we see from stackoverflow. We may consider to use some existing libraries, or if none of them are good, we can implement our own. I guess even the most naive approach such as keyword counts may do better than not showing related questions at all.
In my view, downvotes are a harsh but fair way of indicating that a question/answer requires improvement. And a quick indicator of whether one should spend (waste?) time on a question.
The number of moderators/admins can be seen towards the bottom of the about page: http://www.biostars.org/about/
I personally think (but I have no direct proof other than a hunch) that the creators of SO regret the existence of down votes. For example if one tries down-voting on SO the first thing they will get is a popup message from SO trying to discourage them from doing so. Instead SO tries to get them to do improve the question in some other way. To me that is the best indicator that they too greatly dislike the feature and they believe that it is not useful. But they are bound by the past - changing how their site works is not nearly as easy as it sounds. Even if the vast majority of people wouldn't mind those that do would be very vocal and overall would cause too much disturbance.
The more skilled one is with computing the thicker skin they have - having the computer expose every single mistake makes us all used to the concept of being absolutely wrong and we know that is just a way of life. For someone new this is different. Any negative action is perceived as amplified manyfold, plus it is a person doing it not a computer. It is not as much hurt feelings but being discouraged, depressed by and turned off of science. My overall feeling is that not getting an answer is already punishment enough - the post does not need to also have a downvote or a closing to make that point. The person is already down - with these actions we're just adding to their problems.
Adding title suggestion is definitely priority on the todo list.
I do not think the creators of SO regret about down voting. Down voting should be used with caution in general. It is a good behavior to leave comments about why you down vote. However, I would argue downvoting is still an essential feature. Say we have two software packages, A and B, for the same task. Many use A but they don't like A; in contrast, fewer use B (because, for example, it is from a small group) but everyone who has tried B likes it. Without downvoting, A gets more votes simply because it has a larger user base, but B is actually the right choice here. Downvoting will help to balance the two sides, to some extend. For a question with some votes, I don't see how downvoting is a bad thing to have. For a question with few votes, not giving negative votes should address most of your concerns wrt "our feeling".
odd things can happen if one starts add rules of when a downvote is allowed, may lead to other headaches:
Say we don't allow negative scores and only posts with upvotes can be downvoted. Suppose you upvote and I downvote, now the post is at zero. But what should now happen if you want to remove your upvote because you've changed your mind: either the post will become negative or the score does not change or the the removal is not allowed. Each is option is suboptimal.
In your example in the end I think the stronger, larger group will win out no matter what the rules are, there is always a workaround.
On negative votes, you way is too complicated. The right way is to keep the exact vote in database. You display negative vote as zero when generating the webpages we see. As to my example, the stronger and larger group will probably choose A and we will be propagating a suboptimal choice. Even if we ultimately choose B, having downvoting enabled helps to give a more accurate picture, doesn't it? Another related (but slightly differerent) example is to vote for favorite programming languages. The language with the largest user base will be the winner, but that language may be hated by more. You cannot solve the problem without downvote. In my view, at a Q&A site, a voting system without downvote is fundamentally flawed. That is why the major Q&A sites support downvote.
but if a user votes on a post that others have downvoted already then the displayed score will not increase and stay zero
wouldn't that feel confusing and/or counterintuitive, perhaps even feel like a programming bug: "I voted but the score is still zero"
You can show a tooltip to explain the situation. Even if we show negative votes, I do not see this hurts too much. Those who get negative votes are less likely to return no matter how you encourage them. The merit of downvote outweighs.
I think there may be some benefit to closing posts in order to control what type of content you and others want on the site. That is the goal of the posting guidelines after all, which have been discussed numerous times, but often people don't follow the guidelines and post a one-line question that can easily be answered with google, for example. Or, they don't demonstrate an actual question or if they have tried anything. Quite often you see "What is your question?" or "What have you tried?" on a post. I don't think that is funny and I also don't think the questions that get those responses are of interest or helpful, in my opinion. It seems to me that closing posts is one way of making sure there are some guidelines that everyone follows, but maybe there is a better way to go about it. Perhaps a better system would be where you vote to close, and after a certain number of votes a post may be closed. Even with the admins, there is always differing opinions on this issue, so maybe a voting system could help resolve these disagreements.
I agree. I guess the issue is how important signal-to-noise is for question quality, and how to promote it (if indeed we agree that it's important). I personally don't mind downvotes as much, but I acknowledge that they can come across as punitive and discouraging. I think in some cases, however, that's a proper response for questions where minimal effort was taken to form a well-posed question or asking something that is trivially "Googleable", as an (admittedly negative) inducement to try harder.
If only it were easier to get users to always live by the 'Ten Simple Rules for Getting Help from Online Scientific Communities.'
But you see when we close a post the OP can't make it better anymore, and even if they did they would have to wait for someone to open it before it could be answered. That's really what I dislike the most about it.
Would it be useful to have a mechanism for temporary closing that would lift if the OP edits (hopefully improving) their question? So maybe it would show as closed to everyone else, but the OP could improve it and have it automatically reincluded.
I agree that eliminating the chance for improvements is bad, but I think it's also not good when there are a lot of poor-quality questions clogging things up. This is just my possibly wrong impression, but I feel like many questions are abandoned after posting even when other users respond asking for clarification or improvement. Those orphan questions would be better candidates for closure.
how about if someone asks a bad question, others chime in and ask for clarifications and for a followup and if that does not happen in some reasonable amount of time maybe it is best if we just delete the whole question.
I too feel that abandoned and un-answerable questions add nothing and should be penalized.
I think overall I have to agree with Istvan on both issues of downvoting and post closing. In the thread above, there are given some real concrete examples where (rarely on this forum) downvoting or closing a post has a positive effect. But, in my opinion the ecosystem of BioStar is very different from SO. For one thing, we have a much much smaller user base (both creators and consumers). It is (or should be) more of a community of colleagues talking to each other than a forum of a few domain experts helping masses at large. We should make it as welcoming as possible for new comers. We are very dependent on fresh blood coming in to deal with the constantly changing landscape in which relatively few people have the knowledge that others need. We don't have a critical mass like wikipedia that would allow us to be very strict on posting rules and still continue to grow. We need to be more flexible and understanding of newbs so as not to scare them away. For these reasons, I think the downsides of downvoting outweigh the benefits. Same thing for closing all but the most flagrantly off-topic posts.
Educated undergraduates and most postgraduates can ask naive but good questions. Such questions are less likely to be downvoted or closed. In the old biostar, most questions getting downvoted were impropriate questions where the questioners did not bother to spend a minute to even try to find the answer by themselves while expecting a long answer requring 10+ minutes to write. Most of these questioners either asked once or kept asking low-quality questions even after others suggested improving the questions. They wouldn't learn. Such behaviors should be discouraged for good.
For the future of biostar, attracting expertises is more important than encouraging those who don't respect others' works. To attract expertises, we should make them feel that biostar is a high-quality Q&A site and that they can benefit from it. We need to improve the overall quality. Although I agree that post closing is a little aggressive right now, I don't think we should get rid of post closing and downvoting. When used properly, these mechanisms will help to reduce bad questions but not alienate new comers who are willing to make contributions.
You make a good argument. I think the problem is that an inappropriate question is easy to observe. But, the effects of inappropriate negative voting are hidden from us and much harder to observe. Nevertheless they can have pervasive and undesired effect on the forum. I agree that attracting expertise is more important than saving the feelings of a few "bad actors". With question closing I think moderators can agree and follow a good policy which strikes a balance. But, with negative voting being open to all, we can't really enforce "proper use". In my opinion a much bigger problem than negative voting is insufficient positive voting. I would rather people take more effort to upvote and accept good questions/answers than spend time down-voting or not voting at all.
Asking for improvements and deleting if it does not happen after some time would be my favoured mechanism. It needs to be combined with a redirect to a guide of how to post good questions.
Here is also something that I just realized - as I happen to have some further insight into the last two post closings. Each were performed and agreed to by one or more male moderators (including me) and involved a female poster. This just occurred to me this morning and really gave me a lot of pause and caused some reevaluation of my own approach to everything.
Let me make it clear that I don't accuse anyone of sexism - after all we can't even tell the gender of the poster. But there is a different type of sexism one that still alienates the other gender. And perhaps promoting and relying too much on actions with negative connotations: downvoting, closing and the overall sentiment where every action needs to be judged and must be fixed or else are such actions. Look around - there are so few women participating on this site and I think that is probably true for StackOverflow as well.
Couldn't that also be because we are adopting rules that are unpleasant and alienating to women? Something to think about.
+1 on this and absolute agreement.
I think the use of closing/deleting questions has been (inappropriately?) implemented for questions that are difficult to understand and one concern I have in this regard is the proliferation of poorly worded questions from first time users (this is just an observation, not a rule).
Just trying to brainstorm ways that we can most efficiently handle what I see is a proliferation of "new, yet old" questions.
Should we also enlist a "start-page" for first time users, ideally just the about and FAQ tabs with a check box "I have read the following and understand how to post". Perhaps even a one page tutorial on how to get the most out of the forum that each new user should read? I think the idea of a "have you checked for a similar question" flag has come up before too. Should these (Ten Simple Rules for Getting Help from Online Scientific Communities and BioStar: An Online Question & Answer Resource for the Bioinformatics Community) be mandatory reading for all new posters?
good ideas, one way would be to add a mechanism where people posting their first question would get some extra tips/help and messages inserted into the form to help them guide through the process.
I know I am frequently the minority - but I do not think that helps. If it were me, I would be annoyed a little by requiring me to read through a lengthy article. I would simply skip this step just as I click "I agree" without reading the agreement when I regiester to most forums. Users who already know how to ask proper questions do not need to read the articles; users who don't know how to ask proper questions will not bother to read them.
I would also probably ignore a lengthy article. But, having some short posting tips for newbies probably wouldn't hurt. It can just be a few gentle reminders like: have you searched the forum for an answer to your question? and then an optional link to a guide on good posting habits. I don't really see the harm in (apparently) poor questions. I have seen many weak questions which get improved/corrected by a moderator and/or clarified by the poster and then go on to have quite good answers. Our goal as moderators should be to lead by example, respond to newcomers positively, in such a way that new users are encouraged to come in and hopefully help carry the effort forward themselves.
What about implementing a system where 2 mods are required to close a question? Or maybe even more than 2?
Or maybe a system where mods are able to downvote a question, and when the question becomes more than a certain negative value, mods are able to close the question.
well, it is something to consider. I wish the suggestions were towards simplifying the site not making things more complicated. Minimalism and removing features are like Ockham's razor, when in doubt they are almost always the better choice.
I have to agree with Istvan's comment: "Bioinformatics is different. It is a rapidly advancing field where the correct/recommended answers may change within a short period of time. The rationale of trying to maintain a single set of answers does not apply. Most importantly old and potentially highly voted answers may actually contain somewhat obsoleted advice. Fact is we don't go back to old answers and correct them, nor do people usually visit old questions to add new, updated answers (it happens but not it is not common)."
This to me is a very important difference between Biostar and SO, where a vast majority of answers are timeless.
I strongly agree with this and posted to this effect in the newsgroup last year
is it possible to use some kind of question revision to improve the question and admin consume a little time in human revision also we can use short sentence pop up to guide the user and give him tips to how to post a question?
When reading your post again, I do not see any real arguments. It is solely your wishes. Also, you do seem to miss the concept of post closing and what it is good for. If there is something that should be closed it is this discussion, desperately hitting the close button.... You seem to want to turn the board into a 'petting zoo'.
A poll has been initiated here: Poll: Should the "post closing" option exists? Vote Now!
Post closing has been added back