- From: Sergio Fernández <sergio.fernandez@redlink.co>
- Date: Thu, 17 Mar 2016 17:07:06 +0100
- To: phil.barker@hw.ac.uk
- Cc: Dan Brickley <danbri@google.com>, "public-schema-course-extend@w3.org" <public-schema-course-extend@w3.org>
- Message-ID: <CAOfJQJ0zdpXxURxS1b_JrCSrzrJb+d0rYPXrNmO9b-q71dbGiw@mail.gmail.com>
Hi, I've quickly check the Course proposal, and definitely there is common problem with the EventSeries we outlined in #446. But also it opens a different problem about instantiating an course; e.g., "Algebra II" and "Algebra II in 2015/16", where the semantics is different of an event in a series of event. Because I'd not say that the "Concert of Arcade Fire in Bilbao" is an instance of "Concert of Arcade Fire", but part of a tour. On Thu, Mar 17, 2016 at 2:25 PM, Phil Barker <phil.barker@hw.ac.uk> wrote: > On 17/03/16 12:34, Dan Brickley wrote: > >> Phil et al., does this make any sense, e.g. perhaps treating >> CourseInstance as a named subtype of EventSeries? (or at least sharing >> structure/approach with it) >> > I wouldn't be too happy about subtyping unless we can have a series of one > (for single day events), and you mean multiple inheritance (we will be > making use of many other properties of Event). And if you mean multiple > inheritance then, given EventSeries < Series < CreativeWork, we would be > back with CourseInstance as a sub class of both Event and CreativeWork, > which has its own problems. > I wouldn't go for that path either... The proposal for EventSeries with the example of the different editions > every summer of the Salzburger Festspiele seems more like it is dealing > with a relationship similar to that between a Course and the > CourseInstances (running different 'editions' of the MSc in Basket Weaving > every year). It doesn't deal with subevents of those editions happening at > regular times, which is what I think we have for CourseInstances. > Recurrence is definitely also an open issue. I didn't followed the latest discussion, but have we reached any agreement on that? A common structure makes a lot of sense for things that happen at regular > time slots (either series of related events or repeats of the same thing) Yes, that's an aspect we have to work out, trying to provide a simple modelling covering a broader use case. Thanks for including me in the loop. Cheers, -- Sergio Fernández Partner Technology Manager Redlink GmbH m: +43 6602747925 e: sergio.fernandez@redlink.co w: http://redlink.co
Received on Thursday, 17 March 2016 16:08:15 UTC