Archive for March, 2009

Inviting Outlook users using open source systems

Wednesday, March 25th, 2009

I’m a happy bunny this morning with regards to calendaring. I’ve finally managed to solve why MS Outlook was ignoring the events sent with a timezone stamp. If I scheduled an event without specifying the time, then no time zone id is attached to the event so Outlook parses it quite happily. If I did set a time, say 10:00, then the timezone id for Europe/London is attached. I followed a suggestion on this thread on the ical4j forum.  Jari Oksanen’s page on iCal for Outlook suggests that Outlook does not like local timezones and asks whether Outlook can do with out them but that appears to be a very local solution. If, like us at JISCmail, you need to be able to service requests across a geographic range, then timezones are very important.

The full version of the vtimezone adds in RDATEs and EXDATEs for the various exceptions which Outlook does not appear to read, so you need to use a cut down version to ensure that all clients read the data (the ical4j package – download from Sourceforge – contains a several version of timezone headers which you  can either use with the package or as a basis for rolling your own).

One gotcha might be that the timezones for Outlook may not be 100% reliable and I’m assuming that this is due to the amount of fine-grained material that is missing from the vTimeZone header that you need to send. However, until the Outlook team provide better support for zTimeZones, the occasional error may well have to be lived with.

You must set a timezone so that clients can accept the invitation though but apperently this is on the list of things that the calDAV committee are looking at changing since it causes so many issues.

Depositing blogs – feeding repositories from blogging applications

Thursday, March 19th, 2009

I’ve recently been working on a plugin for WordPress to set up each post as RDF enabled using OAI_ORE and SWORD which I presented to the Oxon SWIG on Tuesday.

The Berlin Declaration of Open Access states the work should be free and also that it should be deposited in a repository. This seems to be about papers and articles but what about the use of blogs, wikis and even perhaps Twitter (might be a little stretch at the moment but I could see it being used)? That suggests a layer of data which could and, where practical, should be being archived in repositories as they are being used as open Laboratory notebooks with links to data.

The plug -in that I’m working on is designed to make blogs readable in RDF for the purposes of repository deposit.

At the moment, I have written a channel which lists all the blog’s posts (using the ?repository=site ) as well as individual post’s in RDF ( using ?repository=post&repository_id= postid). I’ve been using the SIOC exporter as the base model but I’m looking at using skos to get the categories and tags out of the WordPress (and trying to leverage folksonomy through that). Next will be to look at the comments and trackbacks and using the isReferencedBy to export incoming links.

I’ve put this onto KnowledgeForge as its own project.