On Tue, 11 Nov 2025 at 10:58, Eric Lease Morgan <
[log in to unmask]> wrote:

>
> Thus, for a good time, I decided to apply OAI-PMH to the whole of the
> Journal for the purposes of collection building, indexing, and
> preservation. I started out with the OAI-PMH data URL. [2] But that link
> will break, sort of. Instead one wants to append the Identify verb to the
> URL.


The link you provide is the OAI-PMH baseURL which is the protocol-specified
identifier for the repository.

Alas the OAI-PMH protocol is very old (it's a litter-mate of RSS) and
predates the collective realisation that URLs used as identifiers are less
than optimally useful if they're not dereferencable.

At least the OJS implementation fails in such a way that when you open the
baseURL in a web browser, you can easily get to the content.


> Why should you care? Well, there are bunches o' open access journals. Some
> of them may fit into your library's collection development policy. You too
> can collect, preserve, and index this sort of journal content. I'd be more
> than happy to share my software, but there is a caveat: the journal in
> question must be published using OJS (Open Journal System). [6]
>

For the recent OpenAccess week, I wrote a wikidata query that found all New
Zealand diamond open access journals:
https://www.wikidata.org/wiki/User:Stuartyeates/DiamondOpenAccess  It uses
seven different definitions of "New Zealand journal" because there was some
debate here about that.

OAI-PMH is really cool. It is a shame we -- us librarians -- do not take
> advantage of it to a greater degree. Still, fun with librarianship.
>

 I think that actually we do use it a lot. DOAJ is built on it, for example.

One issue that is arising is that some of the more comprehensive metadata
formats (i.e. that exported to crossref.org, including full reference list
and authority control for contributors, etc) are not being included in
OAI-PMH but default to proprietary APIs.

cheers
stuart
--
...let us be heard from red core to black sky