Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.
1 to 22 of 22
Click on the 'Users' tab, and the resulting page should have just such a search box, if I recall correctly.
What Yemon said will search for registered users whose display name contains the word "Kevin". There are 11 of these. There are also four unregistered users whose display name contains the word "Kevin", and I'm afraid there isn't a good way for non-moderators to find them. On the users page, you have to type in the exact username for the user to appear at all (which is pretty weird behavior). One thing you can do is use the fact that Googlebot loves MO and that the username is part of the url in the user profile page. The following Google search returns 14 users, which is three better than the users page:
Anton, do you know why exactly google loves MO so much? it's shown up first in pretty much every search I've made which has it among its results!
I guess we get a lot of incoming links? SBS also got ridiculously high pagerank pretty quickly, and was clearly less of a hit than MO.
I think generally the academic world is treated nicely by pagerank. Personal pages of mathematics rank surprisingly high. We have plenty of connections, inbound and out, with this, and had a great surge of blogging links right at the beginning.
We also "play by the rules" --- the titles and URLs of our pages are real content, reflecting the actual page, and the Googlebot rewards this. We have absolutely zero spammy links, so the Googlebot trusts us. (Yes, I'm anthropomorphising, but in this case it's not so unreasonable, I think :-) The StackExchange software was very carefully written to make the Googlebot as happy as possible, and in part this is an example of the benefits of putting up with proprietary software that we don't control: it's awesome.
I think Scott hit most of the points I would mention. MO has great content, lots of incoming† links from high-pagerank sites, and SE is organized to make it easy for search engines to index. URLs contain the title of the question (search engine spiders tend to weight words in the URL when deciding what the page is about). There is also an up to date sitemap which tells the search spider which pages are changing and how much it should prioritize crawling different things (for some reason, it's not served to users, but you can fetch it with the command "wget -U Googlebot/2.1 http://mathoverflow.net/sitemap.xml"). According to Google's webmaster tools, they're crawling about 8000 pages per day.
†Scott said we also have lots of outbound links, but I think this is false from the point of view of a spider since SE automatically "nofollow"s links in posts. I wish it didn't (or at least I wish I could configure when it did). If you have meta.SE rep, please vote up this request.
@Mariano: My guess is that the main factor is actually Google search personalization. You'll notice that a similar thing isn't true if you search for StackOverflow-related material.
@Qiaochu, I've checked a few times while logged out, and we still do very well...
@Scott: Hmm. I don't completely understand how personalization works, but it seems to be tied to the Google toolbar as much as to your account. I did do the same check though, and you're right; we're doing pretty well!
@Scott: Did you remove all your google cookies in addition to logging out? I haven't checked if logging out removes all cookies, but I suspect it may not.
It doesn't, but I also wiped my cookies and MO results were still pretty high up there.
I get the feeling that a new professor could very well have his/her MO page accidentally eat up his/her personal page in terms of search priority.
logging out certainly doesn't remove google cookies. On the other hand, I'm very dubious that google personalises searches when you're logged out. I could be convinced, because if this is the case it would be easy to find evidence, but I'm sufficiently dubious that I'm not going to bother trying to find such evidence myself! :-)
@Scott: Personalized Search for everyone
I stand corrected! I see that there's a way to turn even that off, but still this is a little disappointing. I do like the idea of "objective reality". :-)
I don't think that's all that likely...The MathOverflow user page is on the first page for all the high-rep users I tried, but in all cases where the user has a personal page, that page was the first hit. Reid doesn't seem to have a personal webpage, but his Wikipedia page is higher; similarly, Qiaochu's blog beats his MO page.
The only people who seem to be in danger of this are people without professional webpages or blogs, like Harrison or Gjergji Zaimi, and I bet they could correct that quickly if they made university pages.
Both my webpage and my nLab page beat my MO page.
Harrison has a blog, but unfortunately he doesn't update too often. There are also a lot of people named Harrison Brown; searching for "Harrison Brown math" gives his MIT ESP page first, his MO page second, and his blog third.
I believe so. It's going to be a long time before MO has PageRank even close to most respectable institutions.