Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.
1 to 17 of 17
In this Meta thread on study groups,
I suggested something which I should have guessed (or confirmed) did not work. This was to use the search features in meta to find information about a meta.mathoverflow user that was posted in their profile. Since it doesn't work, I suppose this might be a meta feature request made available to registered meta users. Something similar on MathOverflow would be great, since the public view of User pages would be the domain of the search (but we can't change the software, whadyatalk,whadyatalk).
So the actual goal is a specialization of the title goal of this post: How do I use Google to search the MathOverflow user pages for something (e.g. users who might be interested in study groups)? Anton (Hopefully Dr. Geraschenko by now?) has kindly provided some examples in the past, perhaps he can answer the specific question and remind us of the variations. I also welcome answers from other MathOverflow/Internet savants/masters.
Gerhard "He Doesn't Know The Territory" Paseman, 2011.06.10
That seems to be more of a google question.
What more do you need beyond?
Thank you for your response, Peter.
What I want (not so much as need) more is something that specifies the publicly viewable pages of users of MathOverflow, both registered and non-registered. I have not tried your solutions, but I suspect they will search all the pages of MathOverflow, and not just the User pages. I am hoping Anton or another moderator will say: specify site:mathoverflow.net/proprietarydirectory/users or specialtag:MOusers or download the database to your favorite Linux system and run the following grep command, or since the DB is in XML, you can use awk in the following way. Or something like that.
I will admit to some laziness on my part, as I can probably experiment and come up with a way. The question (and suggested user interface) seemed simple enough to me that many might want to know some more specifics. I invite you to fine tune your answer so that it works on User pages.
Gerhard "Ask Me About System Design" Paseman, 2011.06.10
Gerhard, as you might have guessed, I didn't exactly spend hours on my answer...
Nevertheless, it should be simple. From what I can guess, adding site:mathoverflow.net/users/ to your google search should restrict your search to user pages. I don't think there are any other user pages on MO.
Testing shows ~84.000 hits when searching for nothing (which sounds about right).
Unfortunately, "study group" returns 0 hits.
Peter, obvious an answer as it may be, I think you have just provided something useful for many. I call upon the moderators either to add this to the faq or tips and tricks page, or tell me where in the MathOverflow documentation it already exists.
And thank you, Peter, for testing it out and reporting the results.
Kevin, that's right. Since that is what I asked, his answer fits the bill nicely. Further, since it nicely complements other searches listed on the Tips And Tricks page, I have asked for it to be included.
Gerhard "Ask Me About System Design" Paseman, 2011.06.11
In my opinion, the logic behind putting this on the Tips and Tricks page is the same or very similar to the logic used to include the other search examples. If that doesn't work for you, I do not know what else to suggest. In any case, I have done what I feel was needed for this, and I thank Peter again for his help and David for his support. Since it was your sci.math post over a year ago that brought me here, I thank you too Kevin, logic notwithstanding.
Sorry, I don't quite understand this discussion (except perhaps if Tips&Tricks was changed within the last day).
There is a point in Tips&Tricks (16) that suggests to use site:http://mathoverflow.net for searching (of which grp is aware).
Now, grp suggest a small addition that is certainly not less MO specific, but rather more, as after all it relies on the particular structure of the site; if user pages were under /user rather then /users or, I don't know, depending on the day of cretation in /usersYYYYYMMDD then a different command would be needed.
And, he is told it is too unspecific !?
(If he would be told it is too specific and very few people will ever want to do this and thus it migh not be worth adding--this is not my opinion just for the sake of argument--, then I would at least understand the line of argument.)
an(ony?)_mo_user: As pure speculation, I can imagine a person who is either extremely savvy at using Google searches and thinks that such an item as site:/mathoverflow.net/users would naturally occur to anyone, so why bother to call it out, or another type of person who is not so savvy but neither sees a need for such a specific search, as it is not among his interests, so why bother to call it out . I will not conjecture as to Kevin's position or motivations (and I don't see him mentioning specificity or unspecificity as a criticism),
but I am not doing this for him; I am doing it for myself and for the other poster who wanted to add a feature to MathOverflow to "improve" in his view the networking capabilities of the site. For those who want to use MathOverflow as a base of operations, I think following a suggested format in the info part of the user page would do much to enhance the networking capabilities, especially since a search will focus on just the user pages. I might suggest such a format later, but for now I am content with knowing a way to limit my search so that I can easily make a list of other users on MathOverflow who have some background in, say, universal algebra. That is trickier to do with a general Google search that does not use site:mathoverflow.net/users.
Gerhard "Likes Searching To Be Easier" Paseman, 2011.06.11
I don't really see any harm in adding a short comment about doing directory-specific searches when using Google. Unfortunately, my moderator powers aren't enough to change pages. Also, "networking capabilities" are not a particularly high priority here.
I can see why one might chafe at the rather bizarre commanding tone in comment number 5. Minor changes don't really need major posturing.
As Kevin Buzzard said, in the grand scheme of things this doesn't really matter, but since the point was missed previously I thought I might try to drive it home by analogy. I use a sledgehammer only because I am not sure what else works at this stage. Also, the person who provided the solution did not get the point on the first try, so I should cut us all some more slack.
Suppose I am trying to do some cool ninja search stuff on PhysicsOverflow. In particular, I know (or believe) that the comments (where the real meat of what I am interested in is) section is stored as publicly searchable web pages, and I want to search ONLY those pages, and not the cache of compiled pages which contains other information, e.g. questions, answers, and other dross. Since I find the site search function rather limited, I go to Google search to deal with the cached copies, since I don't care if the info is two days or two weeks old.
So I try to use site:physicsoverflow.org to start limiting the scope, and then I realize a problem: I don't actually know where the comments are located in Google space. They may be in a subdirectory whose name I need to learn. They may involve a tag or command that I need.
There may be other bits of information I need in order to refine my search, because I don't want to search all of physicsoverflow.org.
So I call up Google tech support: they say we don't know physicsoverflow from a hole in the ground, and since we have no topologists on this shift, we don't know it from n holes in the ground either. So scratch that line of inquiry.
So I try meta.physics.stackexchange.com: they say there are political problems, I should upgrade to version 2.0 anyway, and the information
is proprietary and dependent on how the sysadmin ran the installation script that set up what data is located where and how it should be viewed by the user, never mind how Google might see it. So that line is a bust too.
So I email the moderator at PhysicsOverflow. I'm hoping to hear back from him someday; seems his hand got caught in the large hadron collider and he is trying to retrieve enough parts of it for his high-energy physics dissertation (I TOLD him he should have gone into theory).
So I try a guy I know at another Overflow website. He's been helpful in the past, provided some similar examples, and worked with the
installation enough that he might be able to give me a clue. Plus, he might find this specialization useful as some people (OK, me) have been bugging him about a way to search comments. If I'm lucky, he'll reply soon (sooner than the PhysicsOverflow moderator), although I understand that such requests are a second (or lower) level priority for him too.
It's not that I wanted to be taught how to use Google search. It's that I wanted to be taught (or even just told) what site-specific features would enable such a specific search. At some point I would have tried what Peter did; I thank him for doing it before me. However, I do not yet know if what he proposes searches all user pages or just registered users (I suspect all user pages). I thought my first reply to Peter made it clear what I wanted; it seemed to make it clear to Peter. Perhaps that shows Kevin some logic. If not, I hope he finds it at least partly amusing.
(Yes, I know there are other ways to find out the desired information. That's why I had the experimentalist moderator get his hand
caught in the collider. I like to ask about previous experiences before trying it myself.)
If you are reading this Anton, I offer my (conditional) congratulations on your new degree. The conditional is based on whether it has been awarded, which I do not personally know. I have every confidence, however.
To Scott Carnahan: I am sorry if I come off as posturing or commanding. I would have said exuberant myself, but I'm on this side of the keyboard. If there is no one among the moderators with the interest or ability to add the suggested search example, it is unfortunate. I can also believe there are bigger issues to deal with regarding running of the site.
Gerhard "Someday Might See Grand Schemes" Paseman, 2011.06.13
This is an excellent trick (thanks Peter for pointing it out!). I've updated the tips page to include it. Please let me know if it can be improved in some way. I agree that this tip shouldn't be in the FAQ itself, but I think it's worth including on the tips page even though it's not MO-specific. After all, we have a help page for markdown even though lots of other sites use markdown, and lots of stuff on the how to ask page is about how to ask questions in general.
Thanks Gerhard for your congratulations. It was indeed awarded (though the physical paper won't exist for several months); they taught me the secret handshake and everything. My wife and I will be postdocs at Caltech next year. Thanks to Peter, I can easily find new friends.
Thank you for your reponse and enthusiasm Anton. I'm glad you won't turn into a stranger with respect to MathOverflow.
I have updated my registered user page with a suggested example of formatting social networking data. Mostly it is placing double underscores (see example) before an important tag and HTML tag (BREAK) after to get it to display nicely. I would place an example here but it makes the paragraph not display through markdown text.
If this catches on, I will ask the moderators if they would post the use of the format on the Tips and Tricks page. I want to wait until sufficient enthusiasm and conformity of usage has developed before troubling them with such modifications.
I am not wedded to the format. If anyone else wants to spearhead the efforts in getting such adopted, I am quite willing to turn it over to them and go along with their changes.
Gerhard "Ask Me About System Modifications" Paseman, 2011.06.13