After a bit of a break from doing anything on this, I’ve now got an RSS for individual journal titles up and running. To use it you need to have the ISSN for the journal you are going to track. For example, if you wanted to be updated each time an issue of the ‘The New Scientist’ came in, you would point your reader at:

http://library.rhul.ac.uk/cgi-bin/rss-serials.cgi?issn=02624079

There are some slight problems with it:

• The title of the feed has the ISSN rather than the journal title
• The ISSN in the title appears without leading zeros

These are both problems caused by my lack of skill with xslt – the issn is fed into the xslt as a variable, and seems to lose it’s leading zeros at some point. I’m also not clever enough to get the xslt to pick out the journal title by using the issn, and make this the ‘title’ of the feed.

I could probably solve the second problem if I changed the way I’m doing this, but the current method is (I think) quite neat, so I’m reluctant to change. All I’m doing is taking the existing ‘New Journals’ feed (http://library.rhul.ac.uk/rss-feeds/recentlibraryserialissues.xml), and putting it through an xslt that picks out only those records containing the ISSN provided. This is so easy to do (using a perl script), that frankly it is incredible I’ve only just got round to it.

I’m wondering about creating an RSS for search results – but I’m not entirely sure how this would work? Would the search be ‘re-done’ each time the RSS reader looked for the latest file? This would certainly be the easiest way to produce a quick feed. I wonder how useful this is? Perhaps email is still the best method for this kind of information – I can’t imagine people really want to search more than (say) once a week on our catalogue – it just doesn’t update that quickly.