[uf-discuss] [citation] url field

[Alf Eaton]

On 08 Dec 2006, at 18:05, Michael McCracken wrote:

> On 12/8/06, Alf Eaton <lists at hubmed.org> wrote:
>>
>> On 08 Dec 2006, at 17:26, Michael McCracken wrote:
>>
>> > On 12/8/06, Alf Eaton <lists at hubmed.org> wrote:
>> >>
>> >> On 04 Dec 2006, at 21:48, Michael McCracken wrote:
>>
>> >> If there are enough useful identifiers that aren't URIs (I think
>> >> there probably are), then there needs to be a UID field as well.
>> >
>> > Can you come up with some examples of IDs that aren't URIs?
>>
>> 1234567
>>
>> (actually that's an ID not a UID: some CMS's might only have internal
>> identifiers for articles - if you were using that to look up further
>> information about the article, the ID would be useful to know).
>
> I'm not sure why you'd mark up a CMS ID that wasn't also a URI as part
> of a microformat. Can you expand on your example?

I'm imagining that a journal produces an article that has a URL,  
obviously, but also has an identifier for the article specific to  
that journal/publisher/CMS (a7f6r9, say). If the journal had a  
service that exported further metadata about that article based on  
the identifier, then a processor scraping the page for citation data  
would want to know about that.

This is what the class="unapi-id" field in unAPI is for <http:// 
unapi.info/specs/>, so we could just keep using that, but it would be  
nice if the two were compatible.

alf.