----- Original Message -----
> Date: Tue, 20 Nov 2012 11:39:17 -0800
> From: David Lawrence <[log in to unmask]>
> Subject: OpenURL linking but from the content provider's point of view
> I have some experience with the library side of link resolver code.
> However, we want to implement OpenURL hooks on our open access
> literature database and I can not find where to begin.
> SafetyLit is a free service of San Diego State University in
> cooperation with the World Health Organization. We already provide embedded
> metadata in both COinS and unAPI formats to allow its capture by Mendeley, Papers,
> Zotero, etc. Over the past few months, I have emailed or talked with
> many people and read everything I can get my hands on about this but I'm
> clearly not finding the right people or information sources.
> Please help me to find references to examples of the code that is
> required on the literature database server that will enable library link
> resolvers to recognize the SafetyLit.org metadata and allow appropriate linking
> to full text.
Unless I'm misunderstanding what you're after, link resolvers don't really "recognize metadata" from databases -- library link resolvers are sent metadata in an OpenURL when a user clicks a link that points to her resolver. Unless you plan on maintaining accounts for libraries and implementing IP address recognition so that each library's users will see a link to their institutional resolver when using your database (which is what the commercial databases do), the best approach might be to take advantage of the OCLC OpenURL resolver registry -- instead of having to generate OpenURLs that include institution-specific base URLs, you generate OpenURLs that point to worldcat, and the registry reformulates the OpenURL to point to the resolver associated with the user's IP address. This is the approach used by the Bibliography of Asian Studies at U Michigan; you might want to contact someone there for advice. Of course this approach only works for institutions whose resolvers are registered with OCLC and when users either have an institutional IP address or are going through an institution's proxy server.
I imagine the code you use to generate the actual OpenURLs would likely be similar to the code you're using to generate COinS et. al.
Robert E. Duncan
Integrated Technologies Librarian
David Bishop Skillman Library
Easton, PA 18042
[log in to unmask]