On Tue, Nov 17, 2009 at 6:58 PM, Jodi Schneider <[log in to unmask]> wrote: > The first question is: what are they trying to accomplish by having DOIs? DOIs are just a form of Handle, which is a persistent URL schema. I don't think I need to explain what PURLs are designed to accomplish. > If they're looking for persistent identifiers, I don't understand (a > priori), why DOI is better, as an identifier scheme, than any other > 'persistent identifier scheme' (ARK [1], PURL, Handle, etc[2]). (Though I > really like CrossRef and the things they're doing.) The advantage is that DOIs over other PURLs are used only for citation purposes. As someone who works with a lot of students and faculty, I have observed that DOIs are becoming familiar to them as a definitive citation identifier. As more journals, publishing in an online environment, stop using page numbers in their citations and turn instead to article identifiers -- e.g., citations like this one: Neylon C, Wu S (2009) Article-Level Metrics and the Evolution of Scientific Impact. PLoS Biol 7(11): e1000242. doi:10.1371/journal.pbio.1000242 then DOIs become the most consistently recognizable identifier for constructing findable citations. So, you could use a PURL, but they wouldn't be understood to mean the same thing. Also, DOIs are not dependent on a single resolver -- i.e., you don't have to send them through http://dx.doi.org/ although that's largely been the case up to this point in time. PURLs tend to be server-specific. We don't have to think too far back to recall an instance when a PURL server failed, causing some temporary access problems. Hopefully, DOIs are less vulnerable to this -- although this certainly hasn't been tested. And, responding to Jonathan, who said: >investigating whether every cited article has a DOI and then making sure >to include it... is non-trivial labor. It certainly is if you have to go back and apply them to a backfile of published articles. However, with the Code4Lib Journal, I've been doing this all along in the articles I've edited. CrossRef has good tools for finding this information and when that fails, I go to the cited article itself. Some work, yes, but I figure that's part of my job as an editor. Tom