I am afraid that David will delete comments of 'tHooft, and others, on threads where he has participated. I don't like to see any of 'tHooft's writing go anywhere, at least not for a long time, so please can he hold his horses? Same goes for other comment threads, which are actively going on

The trigger-happy deletion has caused some consternation to users in the past. It is only reasonable for stale discussions, not for active things that are still being sorted out.

• Maybe a physics SE blog could help to resolve such issues... Therein Prof. tHooft or anybody else could write a post about something interesting and discussions in the comments below it could go on as long as the participiants like it and the comments could be kept alive – Dilaton Aug 22 '12 at 11:04
• @Dilaton: This is not a satisfactory solution, as it doesn't let people actively reply to criticism of their technical papers in a way that people will read. It takes a long time to sort out right and wrong in technical things which are not well established, in my experience, it unfortunately sometimes takes ten years (although this is rare, and I'm not expecting comments to sit that long), but it can't be done in a few days. I have not seen so much deletion on other sites, even in long discussions. – Ron Maimon Aug 22 '12 at 16:23
• Btw, what exactly is the comment feature for, if not for commenting? Upvoting comments feature are accompanied by the text 'this comment adds something useful to the answer'. What rule is broken? – TROLLHUNTER Aug 22 '12 at 19:47
• @Holowitz it's fine to take the useful information out of the comments and put it in the answer. Copying an entire comment discussion into the answer verbatim would be defeating the point, though. Comments are for suggesting improvements to the post or for seeking clarification. Once they've served that purpose, they don't need to be there anymore. – David Z Aug 22 '12 at 19:57
• By the way, the issue with "It is only reasonable for stale discussions, not for active things that are still being sorted out" is that the sorting out of active things should be taking place in the chat room, not in the comments. – David Z Aug 22 '12 at 20:57
• @DavidZaslavsky: Were that it was so easy--- when someone makes a technical claim on a work of yours, it is impossible to reply in a distant place, you have to use comments. – Ron Maimon Aug 23 '12 at 0:38
• @Ron sure, but doing so properly is as easy as e.g. "I don't agree because of X; let's discuss this in Physics Chat." There's no need to have the entire conversation in the comment section. – David Z Aug 23 '12 at 0:49
• @DavidZaslavsky: If chat was not restricted in space, and if one was notified of replies, this would be fine. But chat is like comments with all the space restriction and where you don't get notified, and it is designed to eliminate discussion, not foster it. – Ron Maimon Aug 23 '12 at 0:51
• That is absolutely not correct. There are no space restrictions in chat; yes, messages are limited in size but the length is longer than comments, and you can type as many as you want with no issues. And the chat system is absolutely designed to foster discussion. – David Z Aug 23 '12 at 1:53
• "The trigger-happy deletion has caused some consternation to users in the past." Huh? Don't you mean just you? And it's certainly not trigger happy. Your comments with Ben Crowell are still there after all this time. – Physiks lover Aug 24 '12 at 13:28
• I have to say I often hesitate to move discussions to chat because it's a much worse experience than comments. Primarily this is because of the lack of LaTeX, but also you get no information about whether you've crossed the character limit; if you come up with a new point a few days later you find you can't add it because the chat session has been frozen (why, why, why?); and while chat sessions are in theory preserved forever, I find it almost impossible to find old ones. Finally, it's an extra click away, meaning that other people are less likely to read it and weigh in. – Nathaniel Aug 26 '12 at 15:34
• I agree with @Nathaniel. I would also argue that I often find the comments threads on posts more enlightening than the posts themselves. – Jerry Schirmer Aug 26 '12 at 22:05
• ... and as the name and the present implementation (without LaTex, etc) says, to me it seems that chat is more adapted to chat about not too complicated (physics and other) issues of temporal interest than to do serious physics and increase the valuable content of the site which should be easily accesible to everybody. – Dilaton Aug 27 '12 at 22:38
• @DavidZaslavsky if you try to reply to a chat session that's more than a few days old, you get an error message saying something like "it is no longer possible to reply because the room has been frozen". Or at least, that's what's happened to me once or twice. I'm pretty busy right now, but at some point soon I'll try to find time to (a) reproduce this, and (b) post on meta.SE about these issues. – Nathaniel Aug 29 '12 at 11:08
• This collection of comments above is almost verification of how useful comments can be to contentious posts/questions... I lol'd. For those that disagree they seemed to enforce the idea by posting multiple comments in response, as if it would be useful... – kηives Aug 31 '12 at 18:08

I think the "deleting" feature, however it works, is one of the biggest "bugs" on Stack Exchange. In addition to the current controversy, in the past it has deleted very valuable references from the answers to some of my questions. Now I cannot recover them. I am highly displeased, to say the least. If there was an edit trail, or something, it might be tolerable.
This kind of delete without memory should be reserved for abusive comments and the worst kind of spam, in my opinion. It should never be applied to debates with significant physical content. Move, yes if you must. Even hide, if absolutely necessary. Delete without memory, only in the most flagrant cases.
My very passionate two cents worth.

• I exactly agree with this +1. It would be nice if SE could let the different SE sites handle the "comment issue" a bit more flexible and adapted to the specific needs of the local community or the topic of the site. For physics, using chatrooms to clarify physics issues has the real drawback (as far as I know) that LaTex does not work, the content can less efficiently be retrieved (searched for) than the comments that are clearly associated with the "regular" post they belong to, and chatrooms are practically hidden to people who do not participate in the discussion but could be interested etc – Dilaton Aug 27 '12 at 16:27
• I mean, to learn physics (and other natural sciences) one has to UNDERSTAND it instead of just learning facts by heart, and to achieve this a little bit of discussion and clarification is needed sometimes ... – Dilaton Aug 27 '12 at 16:45
• BTW does physics SE no longer encourage experts and active "grown up" researchers to join in here and contribute some high or research-level content? As I have observed on TP.SE these people probably really need more time to come up with good research-level questions and answers because they have to read cutting edge papers, do longer calculations (?), etc, as Ron says. Deleting too eagerly or carelessly important notes or useful hints people have put down in comments, while still working on a "regular post", takes the risk of annoying serious researchers and driving them away from physics SE. – Dilaton Aug 27 '12 at 22:20
• I suspect that you've heard this before and are not impressed, but the "official" position is something like "comments are second class citizens by design and should not be expected to remain indefinitely, anything really important that comes up in the comments should be edited in to either the question or an answer to insure its persistence". And you can hope that the moderators can recognize important stuff when they see it, but if the good stuff is buried in an extended back and forth the odds really go down. – dmckee --- ex-moderator kitten Aug 28 '12 at 1:51
• @Dilaton our position toward research-level contributions is the same as it has always been: they are of course welcome and encouraged, but we're not going to change the whole model of the site to accommodate them. (BTW your last sentence is exactly why people should not use comments for important notes.) – David Z Aug 28 '12 at 21:34

For example:

Me: "Why is A?"

Answer: "Ah, that is because of D!"

Me: "Aha! Thank you! That was my sticking point!"

So I find it greatly distressing to learn that some of these comments might be deleted. It just seems to be fighting against all that is good and useful about this site. It seems, in practice, anti-educational.

• The problem is that those insights should really be edited into the answers wherever possible, then the comments should be deleted. It defeats the point of question/answer if the answers aren't in the answers. Comments are there to help hash out the question/answer, but important info should be moved into the question/answer itself after it's discovered. – Ben Brocka Aug 28 '12 at 13:23
• @Ben Brocka -- I agree that that is what should happen, but the fact is that it doesn't. As a result, this is an unavoidable fact that will have to be lived with. It seems rather bullheaded to face such a situation and decide to delete comments anyways. Please be pragmatic. – user1247 Aug 28 '12 at 15:47
• Well if you see important information in comments, edit them into the relevant post – Ben Brocka Aug 28 '12 at 15:51
• Yep, what Ben said. This is the point of giving advance warning before deleting a long comment conversation. – David Z Aug 28 '12 at 16:02
• @Ben Brocka -- You mean edit into your own question? I don't think I'm able to edit someone else's answer. – user1247 Aug 28 '12 at 16:15
• @user1247 you can edit any question/answer on the site, it just needs to be approved (or it can be applied without approval with 2k rep) – Ben Brocka Aug 28 '12 at 16:21
• @Ben Brocka -- The 'edit' button is grayed-out for me. How do those like me (with < 2k rep) get an edit approved? – user1247 Aug 28 '12 at 16:36
• It's greyed out on Meta cause meta doesn't do suggested edits. Works on questions on the main site. Someone has to approve the edit is all (someone with 5k rep I think, or a mod) – Ben Brocka Aug 28 '12 at 16:42

The following comment got a lot of upvotes, so I'm posting it as an answer so it will be preserved:

I have to say I often hesitate to move discussions to chat because it's a much worse experience than comments. Primarily this is because of the lack of LaTeX, but also you get no information about whether you've crossed the character limit; if you come up with a new point a few days later you find you can't add it because the chat session has been frozen (why, why, why?); and while chat sessions are in theory preserved forever, I find it almost impossible to find old ones. Finally, it's an extra click away, meaning that other people are less likely to read it and weigh in.

As Jerry Schirmer added, "I would also argue that I often find the comments threads on posts more enlightening than the posts themselves."

OK, on the advice of others in the SE network, I'll refrain from deleting those comment discussions in their entirety right away. Some of the comments posted have no useful physics content, and those will be deleted as normal, but the rest I'll leave alone at least until there has been ample time to incorporate any relevant physics into the questions and answers.

At the same time, I would like to remind everyone that the comment section isn't really meant for extended discussions like these. Ideally, the discussions should have been moved to a chat room before they got so long - it's as simple as posting Let's take this to [chat] after the first 6 or 7 messages. Discussions that take place in chat are always preserved forever. Let's try to keep that in mind for the future.

• Sensible attitudes, David. Please don't get controlled by anybody else, there are good reasons why you're a moderator. I seem to share your viewpoint on the role of questions, answers, and comments, and it would be unwise to distort these things just because the name of a user in them is 't Hooft whose contributions to this server are, I strongly believe, below those of an average user rather than above it. – Luboš Motl Aug 23 '12 at 13:36
• @Luboš I appreciate the reasonable comment :-) I've found that the viewpoint of Stack Exchange Inc. toward comments is apparently not as straightforward as you or I would like it to be, and in my actions as a moderator I have to put the official SE policy above my own personal view of how the site should be run (which is what leads me to post this answer). It does make things difficult at times. – David Z Aug 23 '12 at 18:38
• @LubošMotl: I think 'tHooft has contributed to the discussions here greatly, even though I disagree with the technical content of some of his answers and comments. It's like having Einstein on your websight, you can learn from the methods and ideas without agreeing with everything. Also, nobody is trying to control David, but it takes a little bit of judgement, and I used to be always on pins and needles that stuff that I'm thinking about will get deleted. – Ron Maimon Aug 24 '12 at 6:44

There is an easy solution to this. Why not display the most voted comment and have the reader click a link to unravel the rest of the comments?

If they persist long enough they will get included in a Data Dump.

• how is this an answer? – Ron Maimon Aug 22 '12 at 16:11
• @RonMaimon It's not very much, but with David's comments it adds up to the official position of Stack Exchange as I understand it. That particular question has been flagged for upwards of a week now, all the mods have looked at it and none of us have acted. But we also haven't cleared the flag. My reasoning is that the whole mess is iffy for a Stack Exchange site, but I'm making allowances for a VIP. For now. – dmckee --- ex-moderator kitten Aug 22 '12 at 16:14
• @dmkee: Why not just ignore the stupid policy and let comments sit for a year before deleting them? This is a reasonable length of time for things where people actually need to read technical papers and do research before modifying answers. – Ron Maimon Aug 22 '12 at 16:21
• @Ron because then those comments spend a year distracting people from the questions and answers. – David Z Aug 22 '12 at 19:24
• @DavidZaslavsky: Comments do not distract! They are small , in a small font, and they are hidden mostly from users coming in from search. – Ron Maimon Aug 23 '12 at 5:43
• None of that is true, though - they're only in a marginally smaller font, they are perfectly accessible to users coming in from search, and they represent a lot of text that those users can't be sure whether they have to read through or not. – David Z Aug 23 '12 at 6:27

As I have said, you're free to copy the comments to your own notes or to a separate web page, or to any other arena where you want to continue the discussion (subject to the Creative Commons license). That way you can keep his writing as long as you want.

The comment system doesn't exist to support extended discussions. That doesn't change no matter what credentials the poster has.

• Do you ever delete comments that are less than a month old? – Mitchell Porter Aug 22 '12 at 15:09
• @MitchellPorter: David deletes as soon as he can, usually a few days, and he selectively deletes interesting discussions I participate in. – Ron Maimon Aug 22 '12 at 16:11
• i think an allowance needs to be done, at least for a couple of months. – lurscher Aug 22 '12 at 18:18
• @Ron I don't selectively delete your discussions, it's just that you don't notice that it happens when you're not a part of it. – David Z Aug 22 '12 at 19:23
• @DavidZaslavsky: It might be that I selectively tend to participate in longer discussions, that then get deleted. But I noticed, for example, that some comments of mmc with a link I needed were gone (I trawled through several pages of his comments trying to find the link to the paper), and this is probably because they were deleted. The comments already have a space limit which prevents abuse pretty well, and nobody is complaining about distraction, so why fix it if it ain't broke? – Ron Maimon Aug 23 '12 at 0:41