A few thoughts on urls and metadata:

1. I keep getting put off referencing web pages which have overlong urls. They look ridiculous when published on paper and are almost impossible to use if someone has a paper copy and are trying to type in all the characters from that. I have a great example from earlier, but won’t name and shame .. for the moment.

2. One of the systems I access collapsed this morning so can’t access that particular email system or various shared drives. Need to find the most up to date version of a report and found it by looking through files where I have added the date - see how much easier it is to find the right file (forget for a moment all the stuff about ‘modified file dates):

overview v0.1 .swx
overview v0.2 .swx
overview v0.3 5-Dec-05 .swx

That’s Tuesday and I know that I did the work then. The date becomes part of the retrieval process.

In some cases it must be worth putting the date is as part of a URL (I’m sure some organisations do this in some cases):

www.council.gov.uk/2004/events/circus.html

3. Is there work afoot (or maybe it’s been done?) on metadata integration into urls? It could even work in some cases by integrated the top level categorisation in some instances.