[OAI-implementers] repository auto-discovery

[Herbert Van De Sompel]

> On 11/19/06, Herbert Van de Sompel <herbertv at lanl.gov> wrote:
>
>> If I am not mistaken, I think a major immediate aim that you have is to
>> make your journal article metadata Zotero-extractable from your HTML
>> pages.  In this case you might want to check out the COinS work
>> <http://ocoins.info/>, by some of the same people who did unAPI.  If I
>> am not mistaken Zotero eats COinS. The COinS stuff also integrates
>> nicely with browser plug-ins to point to OpenURL Resolvers.
>>
>> the problem with coins is, that a URL may just be too short for all the
> metadata - in particular if the abstract is part of it (see
> http://support.microsoft.com/kb/208427 ); and OpenURL resolvers are a
> non-issue in the case where you're already surfing the fulltext of an
> article.

well, if I remember correctly OpenURL Resolvers can do more than provide
links to full-text. so I don't think they are a non-issue, honestly.  but
I agree that in many cases the full-text _is_ an endpoint. not in all
cases, however.

> i do think unAPI is a good fit, in particular because it can also be used
> for non-xml formats like json; but as Ed pointed out, adding a link
> element
> to point to the oai pmh baseUrl would be a quick win.

Thanks for your input, Robert.  I think we got your point.  As I
mentioned, I trust this will be discussed in the early days of ORE.

cheers

herbert