JournalTOCs Blog

News and Opinions about current awareness on new research

Do we need a “best practice” for generating RSS’s URLs for IR search results?

without comments

When Jenny suggested me to use the advanced search interface options provided by ePrints-based-software IRs to retrieve only the “Submitted articles”, in RSS format, it struck me that the URLs of those RSS feeds produced from IRs that were using the same software (ePrints) differ so much among themselves and are quite long and complex.

For example this is the RSS for Warwick IR “Submitted articles”:

https://wrap.warwick.ac.uk/cgi/search/advanced/
export_wrap_RSS.rss?exp=0|1|-date/creators_name/title|archive|-|type:type:ANY:EQ:submitted_journal_article|-
|eprint_status:eprint_status:ALL:EQ:archive|metadata_visibility:
metadata_visibility:ALL:EX:show&output=RSS&
_action_export=1&screen=Public::EPrintSearch

this is one is the RSS for Birmingham IR “Submitted articles”:

https://eprints.bham.ac.uk/cgi/search/advanced/export_eprints3_RSS.rss?
exp=0|1|-date/creators_name/title|archive|-|ispublished:ispublished:
ANY:EQ:submitted|type:type:ANY:EQ:article|-|eprint_status:eprint_status:
ALL:EQ:archive|metadata_visibility:metadata_visibility:ALL:EX:show&
output=RSS&_action_export=1&screen=Public::EPrintSearch

and this one is the RSS for the Cambridge IR “Submitted articles”:

https://publications.eng.cam.ac.uk/cgi/search/advanced/export_cued_RSS.
rss?exp=0|1|-date/creators_name/title|archive|-|ispublished:ispublished:ANY:
EQ:submitted|refereed:refereed:ANY:EQ:FALSE|type:type:ANY:EQ:article|-
|eprint_status:eprint_status:ALL:EQ:archive|metadata_visibility:metadata_visibility:ALL:
EX:show&output=RSS&_action_export=1&screen=Public::EPrintSearch

Reusing those URLs may not be a problem once you have “discovered” and “recorded” them “manually”, but I wonder whether a “best practice” for generating URLs for those types of RSS feeds is missing here. A more “interoperable approach” may enable their reusability via APIs, in particular RESTful APIs. For the Second Use Case that our project is trying to “solve”, it would make our task easier if those URLs look like these examples:

RSS for Warwick IR “Submitted articles”:

https://wrap.warwick.ac.uk/journalArticles/submitted/rss

RSS for Birmingham IR “Submitted articles”:

https://eprints.bham.ac.uk/journalArticles/submitted/rss

RSS for the Cambridge IR “Submitted articles”:

https://publications.eng.cam.ac.uk/journalArticles/submitted/rss

Written by Santiago Chumbe

July 24th, 2009 at 3:03 pm