De Novo uses numbered links for individual posts, which is one of the Movable Type options. Another is to use the basename, which is automatically generated when you type into the "Title" space for a post. For example, the basename for this post is "problem_with_na," so if I used named links, the URL for the post would be http://www.blogdenovo.org/2007/03/problem_with_na.html. As the MT User Manual points out,
If, in creating the entry, the basename field is left unedited, Movable Type will check for basename collisions with other entries. In case of a possible collision, an underscore and incremented number will be appended to the basename (e.g. title1, title2, etc) and the new basename will be displayed after the entry is saved.(emphasis added). I hadn't realized this until I was looking at a De Novo post from last year, when I clicked on a link and got a dead page. Puzzled, I searched from Professor Bainbridge's index page, and found the post I'd linked; because I had linked it before he added information to it, my link was to the original "columbia_hires_" post name, which later had changed to "update_columbia," thus breaking my link.
If you choose to edit the basename, collision checking is turned off. If the basename you choose collides with another entry, it could cause one to overwrite the other on the published weblog. Also, if you edit the basename of a published entry, it will very likely change the published filename, hence breaking incoming links to the entry. Both of these actions are not recommended unless you know what you are doing.
Link breaks are always annoying, but when they're just on a blawg I can ignore them. When I'm hunting down a source because I'm checking an article's citations before it gets published in my journal, on the other hand, finding the correct URL is no longer optional. As pointed out not only in Adam Liptak's recent Times piece ("The assembled judges pleaded with the law professors to write about actual cases and doctrines, in quick, plain and accessible articles. ... On blogs like the Volokh Conspiracy and Balkinization, law professors analyze legal developments with skill and flair almost immediately after they happen"), but agreed by Ninth Circuit judges Reinhardt and Kozinski (when they recently spoke at Columbia), blogs are becoming a source for citations. They seem more likely than many legal journals to provide what judges want: precedent-based analysis instead of brilliantly original notions of jurisprudence. I had to agree with Kozinski that even in my limited experience, the most useful article is one that summarizes what's happening in a narrow field. One memo I wrote last year began, "There is an excellent survey of the deliberative process privilege in a recent law review article, Michael N. Kennedy, Comment, Escaping the Fishbowl: A Proposal to Fortify the Deliberative Process Privilege, 99 Nw. U. L.Rev. 1769 (2005)."
SPEAKING OF LAW JOURNALS: We're going to have to add http://law.bepress.com/ to our list of sites to search before assuming that one's Note topic is original. While most people seem to have embraced SSRN, the ExpressO Preprint Series allows one to find people who haven't bothered with SSRN but who definitely are submitting their article for publication, even if it hasn't been picked up yet.