# Historical lock for preserving old, low quality questions

The question Compute $$\lim\limits_{n \to \infty }\sin \sin \dots\sin n$$ was recently closed (then reopened and historical-locked). I agree that, by modern standards, the question is not a good one. It is simply a problem statement question, and would rapidly be closed if it were posted today.

However, the question has proved to be quite useful to the site. It is highly upvoted, with a highly upvoted answer (for whatever upvotes are worth) and at the list of linked questions, there includes 20 other questions, most of which are duplicates. This seems to imply that the question of iterating the sine function comes up with some frequency, and that the older question has proved useful in the sense that it gives a place to direct new askers of the same question. Thus I believe that the question should be preserved for its "historical significance" to the site, as well as its continued utility.

Additionally, Martin Sleziak suggested in chat that some older low quality questions have become the target of links from offsite, and have therefore proved useful to the outside world. These questions, too, might benefit from a historical lock–they are poor questions, but useful and deserving of preservation.

I would like to suggest that we discuss a policy regarding the use of historical locks to preserve such questions. So that there is a point of view to discuss:

There are (typically quite old) questions on MSE that are of very low quality but have proved to be useful as duplicate targets or as reference material for other websites, such as Wikipedia. These questions should be preserved for their utility, but should be locked by moderators, as they are not examples of the kinds of questions that we want to see asked today.

• Some past discussions related to closures of (relatively) old questions: Is it good practice to analyse past questions by today standards? and Why close old questions with accepted answers using the “no context” reason?. Maybe also: Under what circumstances is it appropriate to delete a question that has received a good answer? I mention them mainly since some of the comments there clarify what are possible problems if such questions are simply left open. – Martin Sleziak Dec 5 '18 at 16:47
• I think this would be a great idea. The site has grown and changed a lot since the early days, and what's on-topic now can be a lot more stringent than the past. But there's a lot of value to many of the highly ranked questions that shouldn't be lost (and some beautiful expository writing too); I don't think it's harmful to keep a no-context question from 5 years ago as long as people don't think it's acceptable to ask the same thing today. – user296602 Dec 5 '18 at 16:55
• I know that dupe targets can't be deleted, but I'm not sure whether there's any tool support (probably unofficial, but some of the bigger stacks do have a moderate amount of unofficial tool support) to identify questions with significant traffic from other sites. Such tool support would be useful for implementation of the proposed policy. Do you know of any? (PS I see @MartinSleziak mentioned a Wikipedia tool for their outgoing links in the chat transcript you reference). – Peter Taylor Dec 5 '18 at 17:31
• I do think that the entire context of this post should be fully and clearly disclosed - namely that some of the highest rep participants of CRUDE built that rep via the exact behaviors that they now condemn. While I don't answer too many low quality questions myself, I can see why that would be frustrating to those who do. – Mark McClure Dec 5 '18 at 17:35
• @MarkMcClure I am not sure that reputation is really an issue here, since the questions are quite old. If such a question were to be deleted, there would be no change in reputation for most users: meta.stackoverflow.com/questions/320459/… . However, if reputation is an issue, locking prevents users from gaining additional reputation from these poor questions (which closure does not), while preserving the questions for the future. – Xander Henderson Dec 5 '18 at 17:58
• @MarkMcClure 7 years is a long time, and my impression is there was once a day when the vast majority of questions weren't garbage. A change in quality necessitates a change in behavior. – user98602 Dec 5 '18 at 19:07
• @Xander Locking prevents further evaluation of answers, be it by upvote or downvote, which is highly undesirable, e.g. if an answer has an error then there is nothing that can be done about it. A better solution would be to add a banner to the question if one insists on historical annotations on "quality". – Bill Dubuque Dec 5 '18 at 19:47
• @BillDubuque Frankly, I don't see future voting as being that important. In the example I cite, the answer has a net score of 164. This answer has been throughly evaluated. I am not sure that anything is lost by locking the question and preventing further voting. – Xander Henderson Dec 5 '18 at 20:56
• @Xander That's just one of many issues posed by locking. Many old questions serve as dupe targets. If you lock them then no further newer (better, deeper, etc) answers can be added (e.g. we are spotty in some fields so when new experts join it is important that they be able to add answers). Further, we need to be able to massage old dupe target answers so they can work as generally as possible - something I often do. Forcing a huge number of old pages to be static is not needed and is highly detrimental to knowledge evolution. – Bill Dubuque Dec 5 '18 at 21:13
• @gimusi So the answer to my question is "No", then? Let me also add that the questions I have in mind are those that are 5+ years old, from when the site was young and standards were very different. The standards in February of this year were not so different. – Xander Henderson Dec 7 '18 at 2:24
• @gimusi, since you're now clearly answering the question rather than seeking to clarify it, could you use the space for answers rather than the comments? – Peter Taylor Dec 7 '18 at 9:12
• @gimusi I have no idea what your point is. I have never had any interaction with that question until today, and that is certainly not the type of useful question that I would nominate for locking (I have nominated it for closure, and will happily vote-to-delete in a few days should it be closed). It appears to me that you have (in bad faith) gone through the answer history of a user that you have had arguments with in the past looking for answers to terrible questions. What is your point? – Xander Henderson Dec 11 '18 at 1:09
• "It appears to me that you have (in bad faith) gone through the answer history of a user that you have had arguments with in the past looking for answers to terrible questions." - any particular reason(s) behind this assertion? – user 170039 Dec 11 '18 at 5:12
• @gimusi Did you read beyond the title to the question? The point was not to preserve all old, low-quality questions. The point is that there are a small number of old questions that are, by today's standards, bad questions, but which have nevertheless proved to be useful. I do not see how the question you bring up is relevant. – Xander Henderson Dec 11 '18 at 14:28
• @XanderHenderson I've already expressed clearly my thought about that. I think we need simply same rules for all users and for all questions. There is nothing to add more than that from my side. Whenever moderators believe there is the needing to protect some particular question, they have the possibility to do that. We don't need nothing more than that. I respect others point of view but I would like others respect mine. – user Dec 11 '18 at 15:16

If the threads are useful the question-post usually will include a good nucleus that just never developed.

It thus usually should be possible to add some context in order to make the question-post acceptable.

I know that there is resistance to larger edits by users other than the post owner, but I think at least for questions that are old and of the type mentioned, we could make an exception.

If the post owner is still active we could even try to encourage and to guide them, somewhat side-stepping the issue.

And, if it is a long abandoned account I also do not see major down-sides from others editing the post. The edit could be done in such a way as to hint at the fact that the post was rewritten.

• Since you mentioned editing the question, I will add a link to previous discussion about this: Editing someone else's question to add context. (Of course, this topic was mentioned in other discussions too - but this one was specifically about editing.) – Martin Sleziak Dec 6 '18 at 7:12
• @quid I don't agree with that proposal, the editing should be limited to minor issues and/or typos. Exceptions in the the rules are never a good idea. Of course I'm in favour to encourage the original asker to improve the question. All the other cases should be handled according to the same rules. – user Dec 7 '18 at 9:13

Since these questions are apparently points of entry for new people finding this site, I think it is especially important to say why they are considered low-quality today. As you say, a few years ago they were considered acceptable, which means they probably also look acceptable to someone unfamiliar with the site.

Say for example I arrive at this site for the first time via the linked question. The question is clearly a math question, on a site that claims to be a math Q&A site. I really can't imagine why this question would be "not considered a good, on-topic question for this site". It is beautifully formatted, is clear, and has an appropriate title. I conclude that there must be a problem with the mathematics itself, but again, I don't see such a problem.

In the end I conclude I must not understand what this site is really about. It seems to be a math Q&A site that discourages math Qs, so I stay away.

To obviate this problem:

The person doing the historical close should type a short line explaining why it would be considered low quality today (perhaps with some standard reasons being selectable for convenience), and this line should be shown in the lock box.