Removing useful but not immediately on-topic comments

Question

I recently deleted a few comments (mods, you can see them here) which had been flagged as off-topic. Briefly, the OP had asked a question, a community member answered, and then a separate discussion took place in the comments about a different academia-related topic. I removed the comments that weren't immediately related to the question. The answerer took issue with my removing the comments, as he had put work into finding the answer.

My question here is: did I act inappropriately in removing those comments?

Answer 1

Yes—and no.

Yes, you acted inappropriately in the sense that the discussion was still very much ongoing at the time. I would not delete informative comments so quickly—you have to allow people the chance to read them at the time that the discussion is going on. Otherwise, you could remove useful information which might actually lead to improved questions or answers.

However, if you had waited a few days before deleting the comments, then I would suggest that no, such behavior is appropriate. In the long run, Stack Exchange sites are not just about answering people's questions, but providing long-term curating of the answers. Extended side discussions that are not germane to the topic at hand detracts from that mission and should be removed when no longer pertinent to the discussion at hand.

Of course, we don't want to cut off relevant communications among users of this site; such "off-topic" discussions can always take place in chat rooms.

Answer 2

I'm coming late to this party, but I am puzzled by the desire to remove comments. If a comment thread starts getting excessively involved, the system automatically encourages people to move to chat. If they still don't want to, is that so terrible ?

I agree that a comment might reference an edit that once made makes the comment moot. But in such cases I've often highlighted the edit as an update, so that the comment continues to make sense.

I've also deleted my own comments when I felt they were no longer necessary. But I wouldn't delete others comments unless they crossed the line into abusive/obnoxious/spamming behavior.

Answer 3

I can see aeismail and Pete's answers, but in my opinion we need to keep comments short or the site becomes a chatroom (which we also have, of course!). In the example given, the number & length of comments mean that many newcomers will probably not scroll down enough to see answers below Pete's.

And the right time for clearing comment is the current time, because a) most people view the question when it's still active, b) we may not get around to doing it later.

In short, in my opinion: having a prolonged comment discussion should only be very temporary, and should only lead to improving the question or existing answers. If the comments come to a useful conclusion, it should be incorporated into the answer. In all cases, the comments should be removed as soon as possible, by their authors.

Answer 4

Let me respond to Charles' answer.

1. Thank you for including the link to the source of your excerpted passage about comments. That makes clear that it does not originate with this site and is not directly espoused by the users of this site: it is a message from some of the creators and adminsitrators of the hardware that explains how they intend it to be used.

   However, any particular SE site is a group of people allied around building information and answering questions in a certain focused area. Like all communities, we have the right and obligation to make our own norms. If these norms diverge too far from those of the SE administrators...well, we can cross that bridge if/when we came to it. We came very close on MathOverflow.net, but after a long period of time, the SE administrators agreed to some very site-specific features and philosophies. Note that this was done in part because of the MO community being clear that certain things were necessary in order to ensure their long term commitment to the site.

   I disagree that this passage should be used prescriptively across all SE sites. Some people at SE would want it to be: that's their perspective, which obviously carries a good deal of weight...as does ours, since both parties are equally necessary in order for the site to exist at all. But from a factual perspective it is clear that this "ephemeral" notion of comments is not the one which is practiced on MathOverflow.net, and to a lesser extent not on math.SE. Just to give an example, on MO the primacy of comments is so extreme that there are many questions which are only answered in comments. (In my opinion this is too extreme, and I have sometimes left community-wiki answers to such questions just to make sure that questions get answered in the technical sense. But I'm describing how things are at the moment, not how any one person wants them to be.)

   What follows most strongly from the previous paragraph is

2. There is certainly no uniform agreement across all SE sites on the precise purpose and usage of comments. On any given site this is something that evolves culturally: all users participate in it, and respond to what they feel is "inappropriate" participation by others. No one person on academia.SE gets to say what comments are for and how they should be used. Thus I am disappointed that Charles tried to do this in his answer:

   Hence, the proper form of communication on this site should have been: just-learning should have a left comment to Pete L. Clark asking to join a chat conversation to discuss about the top journals, and this comment should have been deleted once the conversation had started.

   Please don't phrase your opinions as telling me and others what is the "proper form of communication". That is for all of us to judge. For my part I do not use the chat features of the site. I do sometimes contact people via email off the site, and since my email address is available they are free to contact me. When I make comments, I have chosen to do so. One difference is: comments are public; emails are private; chat is somewhere in between (I think; anyway it is not available to those who are reading the content that prompted the comments).

   Can comments get out of hand? Yes, of course. It has happened to me on many sites, including on this one. In the sites math.SE and mathoverflow that I have participated in (and, I don't want to make a measuring contest about this, but I do have very extensive participation on these two sites, as anyone can check; in particular, I believe that I can claim to have as well-formed idea of what I want to use comments for as almost anyone), sometimes comments spiral out of hand by either (i) getting personal or (ii) getting confrontational / impolite. In such situations it is great to have moderator intervention. My understanding is that moderators spend a lot of time (on certain sites, anyway) intervening in such matters, and that seems like the best use of their time.

3. Concerning idea that comments which range off-topic from the question/answer should be deleted, I respond: it depends what is meant by off-topic. If it ranges outside of the scope of the site, then yes, it is a good time to curtail the conversation. However, if it just switches from one on-topic issue to another: well, that's what happens when professionals are having a profitable interaction. To me that is exactly the sort of positive interaction that sites like this are supposed to be encouraging. If you disagree, that's your right, and it's your right to try to act on it. But there are various ways to do that. You could act on it by leaving a comment saying, "Hey guys, maybe you'd like to move this discussion to...." In the case at hand that would have been a very appropriate and positive contribution: probably the OP should have asked a new question. But my point is that a short exchange in the comments is a positive move in the direction of such a new question. Flagging the comments and then deleting them is an incredibly negative move to make. This brings me to:

4. I get personally as well as professionally annoyed when my speech is deleted without a record made.

   I am an academic mathematician, and though in many ways the latter is more definitive of my professional identity, in other ways the former prevails. This is such a time: valuing others' speech and writings is a sacred principle of academia. Academics agree that censorship is bad more strongly than they agree on almost anything: this was passionately argued for in Milton's Areopagitica in 1644 and has been well-accepted in the Western academic world more or less ever since. To me there is almost nothing which is more offensive than simply removing my text. (Seriously: if you want to respond to my comment by saying, "Hey, you asshole, why are you wasting my time and cluttering my page. Nobody cares!" then not only am I not that much annoyed by that...but it is much more likely to get me to remove or reformulate my comment. When you delete my comment, I think you are almost denying my personhood and wonder why I should even be associating with you.) This idea is well-understood on the two math sites mentioned above. It would be pretty ironic if this academic principle were felt less strongly on academia.SE....but let's see how people feel about it, I suppose.

5. We do need to push back in contemporary fora when our concepts of personal and academic freedom are being comprised. It is clear to me that the founders of SE do not value these freedoms as strongly as I do or most academics do. Those who have known me for a while know that several years ago (SE cofounder) Jeff Atwood deleted two of my comments on meta.math.SE. I said then what I said now: if that continues to happen, I must leave. Jeff Atwood was at the time very uncompromising, and I did take a break from contributing to that site...during which time I honestly think that my action was felt and taken into account. Jeff Atwood quickly became much more reasonable in his efforts as a "policeman" (his word; I think it's silly) on math.SE....and in fact he is no longer associated with the company at all.

6. Charles also writes:

   No user is forced to contribute, and no user gains from contribution, apart from the collective gain of getting great answers to great questions. It is perfectly fine to question decisions made by mods, and to offer new solutions to solve problems, but someone threatening to stop contributing if they don't get it the way they want is not helpful.

   Whoa there: you are really devaluing your product. Users can gain from contributions to this site in the following other ways:

   1. They can get their questions answered in a timely manner by an expert audience. [Questions are contributions, and getting your question answered when you need it answered is much different than the collective gain of an agglomeration of answers.]

   2. Users can gain experience about how academia works in a worldwide, field-independent context, whereas most academics' (I mean me...) day-to-day life is mostly confined to their department in their university in their country. This is one of my main reasons for being active on this site: I have a lot of experience with my corner of academia, but I know very little about how HCI people do things or how people do things in India. Knowing this will be both directly and indirectly valuable to me: for instance it could make me a better department chair in the future.

      [In order to get the most benefit out of interacting with academics around the world and the academic spectrum....you need to actually let us interact a bit. The idea that anything other than "The answer to Question X is Y" is idle chit-chat is really not helpful in this respect.]

   3. Participation on globally active sites like this promotes me, my department and my university. My activity on MO and math.SE has done more for graduate admissions at UGA than anything else I have done or could do...I can really see the difference it's made.

   4. Dually to the previous point: participation the site puts students in contact with professors: maybe by participating on this site I meet a math student in a faraway land who turns out to be really promising. My contact with this student could be really helpful for her (and hers for me). Please think about that when you delete comments between me and other students and young mathematicians.

Answer 5

In complement to the other answers, from How do comments work?:

How do comments work?

Comments exist so that users can talk about questions and answers without posting new answers that do not actually answer their parent questions. Comments are often used to ask for clarification on, suggest corrections to and provide meta-information about posts.

Comments are intentionally short, having maximum length of 600 characters, and allow only limited markup. URLs in comments automatically become hyperlinks. Each user may post only one comment every 15 seconds.

Comments are disposable: unlike posts, there's no revision history, and they can be deleted without warning by their authors, by moderators, and in response to flags.

When should comments be deleted?

Comments are temporary "Post-It" notes left on a question or answer. You should not expect them to be around forever: Once a clarification has been made, an edit added to the post to include new information, or the issue in the comment is otherwise resolved, it is subject to deletion. In reality, many obsolete or chatty comments remain untouched due to the high volume of comments posted, but this does not mean that they can't or shouldn't be deleted in the future.

In the case mentioned here, the comments were not made to ask for a clarification or suggest a correction, but to ask a new question, quite different (the original question is on acknowledgment, the comment asks about top journals). Hence, the comment should be deleted, and the conversation should be made in the chat, or a new question should be asked (although I would suspect that the corresponding question would likely be closed, due to its opinion-based form). Note that the search does not work on comments, making this information basically inaccessible to other users interested in the topic. Either this information is useful for many users, and should be converted into accessible content, or it is not, in which case it should be deleted.

Hence, the proper form of communication on this site should have been: just-learning should have a left comment to Pete L. Clark asking to join a chat conversation to discuss about the top journals, and this comment should have been deleted once the conversation had started.

There has been some discussions on the past on the possibility of contacting directly another user (Contacting/Referring to another member of SE outside of a comment thread), and this has been declined, for the reason that Stack Exchange is by design avoiding social networking features.

Now, we can discuss about what should be StackExchange, and how we can improve it, and perhaps in this particular case, it could have been useful to have an option to automatically turn the conversation into a new chat room, instead of just deleting the comments. Perhaps it could have been smoother to first leave a message indicating that the comments are off-topic, and to delete then only when another solution has been found. But in the end, deleting them was the right decision with regards to the site.

--

On a side-note, I personally find particularly non-constructive the attitude "if this is the way it works, I will stop contributing". No user is forced to contribute, and no user gains from contribution, apart from the collective gain of getting great answers to great questions. It is perfectly fine to question decisions made by mods, and to offer new solutions to solve problems, but someone threatening to stop contributing if they don't get it the way they want is not helpful.