Overhauled & improved Evergreen implementation of schema.org

Hey all:

Sorry for basically disappearing for the past couple of months. Anyway, I thought it would be worthwhile pointing out the improved Evergreen implementations of schema.org in microdata & RDFa Lite that I've posted on our test servers (heads-up: the servers are starved for resources, so broad searches in particular are likely to time out). Here's a couple of sample records:

Microdata: http://laurentian-test.concat.ca/eg/opac/record/345239
RDFa Lite: http://rdfa_lite_test.concat.ca/eg/opac/record/345239

Of note:

* Personal main/added entries are now Person objects; and where available, they have birthDate and deathDate properties.
* Corporate main/added entries are now Organization objects.
* Albums are now correctly typed as MusicAlbum instead of MusicRecording. Aside: WorldCat also seems to have made the same mistake. If it's possible to claim prior art on a mistake, then I think this one is all mine as the public commit is dated March 18, 2012! Fixing this enabled Ross Singer's cool Backbeat GreaseMonkey script to actually work :)
* Added a mapping for cartographic types to Map objects.

Note that this is all in a proposed branch for Evergreen, not yet committed. Definitely getting better, though.

Again, I welcome comments and extend my offer to model proposals for extensions (for example, holdings). I think it makes our extension proposals stronger if they're actually implemented by more than one system. 

On that note: is anyone aware of any other boring old integrated library systems or discovery layers that are trying to publish schema.org data? I see VuFind has an old Jira ticket at http://vufind.org/jira/browse/VUFIND-425 but they don't seem to be working on making it a default.

Dan

Received on Monday, 8 April 2013 18:48:19 UTC