- From: Niklas Lindström <lindstream@gmail.com>
- Date: Thu, 15 Apr 2010 20:47:51 +0200
- To: Sandro Hawke <sandro@w3.org>
- Cc: public-egov-ig@w3.org
Hi Sandro, all! I like where you're going with this. Unfortunately I was unable to attend this time, as I was caught up in another meeting (relating to open data of course..). For reference, I'd like to mention the notes I have at the COURT project site, on collecting data from datasets at [1] and [2]. This is basically an outline of the practise used to gather data in the Swedish legal information system ("RInfo"), where sources are currently described like in [3]. My interest lies a lot in the relationship between datasets and feeds. >From my perspective, there are some important details to iron out regarding how to describe and use distribution and update notifications. The references I provide speak of three different ways to link a dataset to a feed subscription document: via dady, directly using dcat:subscription, or (as in [3]) just iana:current. I think dcat:subscription may be the best way if it becomes a standard. Note also (as mentioned in [1]) that a distribution mechanism based on Atom feeds probably needs to use something stable like feed archives or complete feeds (described in RFC 5005 [4]), since collecting data from regular paged feeds cannot be guaranteed to be complete (due to variations of a feed page over time, which can cause a climbing mechanism to miss entry updates). (For reference, there's a vocabulary by Toby Inkster at [5] which defines these types of feeds, as subtypes of AtomOwl:s Feed class.) (There is also documentation in *swedish* describing how collection of Atom feeds is done in the mentioned Rinfo system, at [6]. You can try Google Translate on it, but I guarantee nothing legible. ;) ) [1]: <http://code.google.com/p/court/wiki/Timelines> [2]: <http://code.google.com/p/court/wiki/RDF_Practices#Data_Sources> [3]: <http://rinfo.lagrummet.se/sys/sources/rdf> [4]: <http://tools.ietf.org/html/rfc5005> [5]: <http://ontologi.es/atomix> [6]: <http://dev.lagrummet.se/dokumentation/system/atom-insamling.html> Best regards, Niklas Lindström On Thu, Apr 15, 2010 at 6:47 PM, Sandro Hawke <sandro@w3.org> wrote: > > See http://www.w3.org/egov/IG/meeting/Data_Catalog_Vocabulary/2010-04-15 > > IG members welcome to make corrections on the wiki; others e-mail me. > > -- Sandro > >
Received on Thursday, 15 April 2010 18:48:48 UTC