Update (25/4/09). Since the launch, the number of pages on the Tricki has doubled (from 104 to 208), and is increasing fast.
Main post. If you have visited the Tricki recently, then you will already know that it has gone live. I’ve delayed posting about it until we were sure that everything was fully transferred: if you visit the prelive site you are now automatically redirected to the proper site, which you can also get to by clicking here. The URL is http://www.tricki.org.
A few small points to note here. In response to comments, we have introduced some new features. One is a feature for marking an article as a stub. Our working definition of a stub is that it should have no substantial mathematical content, and should not link forwards to any articles with substantial mathematical content. (That is, a parent of a non-stub is always a non-stub.) The thought behind this is that there are two directed graphs of interest: one with all articles, whether written or unwritten, and the other the set of all ancestors of articles with interesting content. The stub feature allows one to explore either of these trees with ease, because if an article is marked as a stub, then all links to that article are clearly marked as well, with a little leaf symbol.
A second change is that in the Forums section we have a section called “Article requests”. If you think there ought to be an article on a certain topic, but are not in a position to write it (because you are short of time, or don’t know enough about the topic, or just don’t feel like it), then it will be very valuable to the Tricki enterprise if you write, in enough detail to make clear what you have in mind, a contribution to the forum in which you ask for somebody to write it for you. More generally, pretty well any kind of feedback is helpful, and has already made a difference for the better (as the two changes just described indicate).
A less direct way of requesting articles is one that I myself have been doing a lot: I just create links to non-existent articles. I should probably change most of these dead links into stubs, to give more idea of what I hope these articles will contain. Another way of helping the Tricki without a huge investment of time is to add more such links. For example, if you see a navigation page (a page that consists mainly of links to articles on some given topic) and feel that the list of links is incomplete in some obvious way, then by all means add to it. Again, it’s probably better to create the articles in question as stubs rather than giving just a title: an obvious way of doing this is to write just the quick description and not the main body of the article.
A medium-quick way of contributing is to add examples to articles that are a bit short of them. This should be much easier than creating an article from scratch, because the technique in question will already have been explained by somebody else.
But while I hope that there will be many small contributions of this kind, the success or failure of the Tricki will in the end depend on whether we get enough new articles with substantial content, and also on whether we can keep control of the organization of the articles. On the first point , let me put it like this: if every mathematician who has visited the site so far were to contribute one “core” article, we would have a much bigger and more varied collection of articles than we have at the moment, and would already get to the stage where no single person was likely to read every article on the site. In other words, it would get to the critical stage where one could find something new and interesting whenever one visited the site. And surely all mathematicians have at least one Tricki article in them — one not completely obvious technique that they know and find useful.
As for the organization, if you come across an article and have the reaction, “This is an interesting and useful article, but the people who would benefit from it are most unlikely to realize that it is here,” then please do something about it. There are many things that you could do. You could create a new navigation page with links to that article and related ones (either already existing or to be created at some point in the future). You could add a link to an existing navigation page. Or you could start a new forum topic by saying something like, “There is an article on X but it seems hard to find. Does anyone have any good ideas about how to categorize it?”
If you don’t really believe that the Tricki would be all that useful to mathematicians, then these remarks are not aimed at you. But if you like the idea of the Tricki, then what I am asking for is a few hours of your time: if enough people are prepared to do a bit of work on it, then each person should get out far more than they put in. (In other words, it’s a typical prisoner’s-dilemma type of situation.)
This post is starting to ramble a bit, because I’ve just thought of something I should have mentioned earlier. Another quick way for some people to contribute is to modify things they have already written. There are informal accounts of bits of mathematics dotted all over the internet: I am hoping that many of them will migrate over to the Tricki, where they can fit into a more organized structure and benefit from the different search methods that are available.
Added later: if you find that you have technical problems, then the quickest way of getting them dealt with is probably to send an email to webmaster at tricki dot org. That way you’ll reach whichever of Alex and Olof is next available. (Of course, “at” and “dot” should be replaced by the appropriate symbols.)