# Is there a way to find which MO questions/answers point to which others?

Every once in a while, an old question of mine (e.g., "$C^1$ isometric embedding of flat torus into $\mathbb{R}^3$"—approaching 5 yrs old) sees a minor flurry of votes. I presume this is because some other MO question (or MSE, or MESE, or ...) cited it.

Q. Is there a way to find which StackExchange questions point to one particular question/answer?

Perhaps there is a general mechanism to list which web pages point to one particular web page?

• Aside from citation, it can also be the result of clicking on something in the "Related" sidebar. – Todd Trimble Mar 25 '15 at 13:19
• Links to and from the question within MO are listed in the list of linked questions, but I'm not aware of similar link tracking within the entire SE network. I think SE monitors where viewers come from, but I don't know if they are willing to share that information about any particular post. – Joonas Ilmavirta Mar 25 '15 at 13:20
• The MSE search feature doesn't help much, but Googling site:math.stackexchange.com mathoverflow 31222 (with or without the site specification) gives a list of links. Do the search results look like what you wanted? – Joonas Ilmavirta Mar 25 '15 at 14:51
• @JoonasIlmavirta: Very clever, Joonas! Unrevealing in this instance. I guess the activity could have originated from any web page, so it becomes a general Internet question of finding all the pages that link to a given URL. Which is of course the basis of Google ranking. – Joseph O'Rourke Mar 25 '15 at 15:00
• Google search links:http://mathoverflow.net/q/ does find "all" the webpages that link there, but extending the search more specifically to links:http://mathoverflow.net/q/31222/6094 fails. – Joseph O'Rourke Mar 25 '15 at 23:10
• @JosephO'Rourke, one reason is that the last part of the link (6094) is your user id. Other forms of the link (copy from the address bar instead of sharing or just remove your user id) give some other results. – Joonas Ilmavirta Mar 26 '15 at 10:18
• If this has happened recently, it could be due to Nash winning (jointly) the Abel prize. In his presentation on the laureates' work Alex Bellos used images from the Hevea project, and if people used image search on screen caps they might find your question, which also uses them. – David Roberts Mar 28 '15 at 5:01
• @DavidRoberts: That is a brilliant conjecture! "nash embedding hevea" hits my question as the 1st image. Mystery solved! – Joseph O'Rourke Mar 28 '15 at 14:17

produces a list of posts on MathOverflow and Math.SE containing links to the question you mention. Actually you don't need both the "q" and "questions" form for the final search term, since in fact all the links provided happen to use "questions". But if you want to use this method for other questions, this may not be the case.

It appears that you are asking two separate questions. You express curiosity about where flurries of activity come from and also ask for a list of questions that link to yours. These are separate things.

Even if you did build a list of all sites and other SE questions that link to one of yours, it would not have enough information to determine when those links were followed, and thus could not answer your question of where a sudden flurry of recent views / votes came from. SE has the ability to track real-time referrer information, and they probably do track it, but that information is not in the public data dump. I suppose if you went to headquarters and got one of the employees drunk enough they might be convinced to organize and share data with you. I wouldn't count on it.

You will, unfortunately, not be able to satisfy your fundamental curiosity here.

As for lists of questions that link to yours; the easiest mechanism available to us users is one that lists linked questions within a given site, as in nicael's answer (which now seems to be deleted; it was to visit http://mathoverflow.net/questions/linked/{postid}, e.g. https://mathoverflow.net/questions/linked/31222).

Cross site links are not readily available. You could download the weekly public data dump and analyze comments and post text on all questions on all sites to look for links to your questions. The in-site links that nicael mentions are available directly in the PostLinks table. This task requires a good bit of effort. Even then, this does not answer your question about flurries of recent activity.

Links from outside are not available in SE's public databases.

Search engines like Google, etc., have information about links between sites all across the internet. I am fairly certain this information is not directly accessible to the public, although you can sometimes get some information by searching Google for the URL of your question to see what pages contain it (and get some sense of their age). This is probably the closest you can get to finding a source of activity, but of course this excludes many things (e.g. perhaps it was mentioned in a Facebook comment, in a chat room, on some mailing list somewhere, or on some unindexed page; who knows).

• You are right that I asked two separate questions, the 2nd of which may be the most interesting. The little-known links: prefix in Google searches is quite interesting. E.g., links:http://cs.smith.edu/~orourke/ -site:cs.smith.edu. – Joseph O'Rourke Mar 25 '15 at 23:48
• @JosephO'Rourke: It's link:, not links:; the latter has no meaning for Google Search. Also, the link: operator deliberately throws away most of the results, see here: moz.com/blog/google-link-command-busting-the-myths – Dmitri Pavlov Mar 27 '15 at 13:03
• @DmitriPavlov: Well, that certainly explains what appeared to be mysterious behavior! Thanks. – Joseph O'Rourke Mar 27 '15 at 15:56