ORO Goes Naked With New ePrints Server

A few weeks ago, the OU Open Repository Online (“ORO”) had an upgrade to the new eprints server (breaking the screen scraping Visualising CoAuthors in Open Repository Online Papers demos I’d put together, sigh…).

I had a quick look at the time, and was pleased to see quite a bit of RSS support, as the FAQ describes:

Can I set up RSS feeds from ORO?
RSS feeds can be generated using search results.

To create a feed using a search on ORO:

Enter the search terms and click search. RSS icons will be displayed at the top of the search results. Right click the icon and click on Copy Shortcut. You can then paste the string into your RSS reader.

It is also possible to set up three types of RSS feed, by OU author, by department and by the latest 20 additions to ORO.

To create a feed by OU author start with the following URL:

http://oro.open.ac.uk/cgi/latest_tool?
mode=person&value=author&output=RSS

Please note the capital “RSS” at the end of the string

Substitute author for the author’s OUCU and paste the new string into your RSS reader.

To create a feed by department start with this URL:

http://oro.open.ac.uk/cgi/latest_tool?
mode=faculty&value=math-math&output=RSS

Please note the capital “RSS” at the end of the string

This displays all research that relates to Maths (represented by the code “math-math”). To extract the other department codes used by ORO, go to the following URL:
http://oro.open.ac.uk/view/faculty_dept/faculty_dept.html
locate your department and note the URL (this will appear in the bottom left corner of the screen when you hover over the link). The departmental code is situated between “http://oro.open.ac.uk/view/faculty_dept/” and “.html”, e.g. “cobe”, “arts-musi”, etc. Copy the department code into the relevant part of the string and paste the string into an RSS reader.

To create a feed of the latest 20 additions to ORO use this URL:
http://oro.open.ac.uk/cgi/latest_tool&output=RSS

This feed can also be generated by right clicking on the RSS icons in the top right corner of the screen and choosing copy shortcut

The previous version of eprints offered an OAI-PMH endpoint, which I haven’t found on the new setup, but there is lots of export and XML goodness for each resource lodged with the repository – at last, it’s gettin’ nekkid with its data, as a quick View Source of the HTML splash page for a resource shows:

Output formats include an ASCII, BibTeX, EndNote, Refer, Reference Manager and HTML Citations; a Dublin Core description of the resource; an EP3 XML format; METS and MODS (whatever they are?!); and an OpenURL ContextObject description.

The URLs to each export format are regularly efined and keyed by the numerical resource identifier, (which also keys the URL to the resource’s HTML splash page).

The splash page also embodies a resource description meta data in the head (although the HTML display elements in the body of the page don’t appear to be marked up with microformats, formal or ad hoc).

This meta data availability makes it easy to create a page scraping Yahoo Searchmonkey app, as I’ll show in a later post…

Author: Tony Hirst

I'm a Senior Lecturer at The Open University, with an interest in #opendata policy and practice, as well as general web tinkering...

4 thoughts on “ORO Goes Naked With New ePrints Server”

  1. Hmmm – yes, and you could also bring the service to its knees by running extremely inefficient queries against it. Also, if the database structure changes in a future version of the s/w, anyone who has done this would have to re-write queries.

    At the end of the day there are good reasons to provide a well documented API that ideally is stable or maintains backwards compatibility rather than giving access to the underlying db itself.

Comments are closed.