[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
SV: The SKiCal extensions - namespaces
Eric
At the IETF in San Diego there where quite a few discussion within the
applications area as to how to utilize namespaces and schemas. Personally
I have had email discussions with Graham Klyne on how to approach this
subject and Patrik Fältström and I once talked about it at a lunch.
The cluster-modularity that the namespace-schema mechanism offers is
undeniable.
The problem so far seems to be coordination. Coordination seems to be a big
problem. Why couldn't (for example) the date-time and recurrence rules of
2445 be coordinated with the date-time and recurrence rules of XML-Schemas?
http://www.w3.org/TR/xmlschema-2/
Frank Dawson has some input on that perhaps?
I just don't know where we could find the force to accomplish something like
this.
Since the semantics of SKiCal will also be published as a schema using XML
syntax - it would be just wonderful if your Sword-of-Alexander suggestion
was to gain ground, but also a bit perilous for us if it was to lead to
diversion in the current discussions.
Hopefully the ongoing SKiCal submissions are serving to enlighten. The
ensuing discussions certainly do so for me. I don't think it is simply so
that the SKiCal submissions as a group fall outside the basic 2445
framework. I think, for example, that the issue of dualism (the calendar
object as a pointer versus the calendar object as a sovereign entity) should
be resolved and many of the SKiCal properties address that problem.
I think that the RECOMMENDED, REQUIRED and PROHIBITED properties certainly
pertain to meetings as well as events. Actually I think all the properties
and property parameters at least to some degree fall with in the business
meeting model.
On the other hand in conjecture with Shannon's mail, there is a valid point
as to what extent automated discovery will be exercised on business
meetings! And consequently to what extent application designers wish to
offer this sort of extensibility.
I surely hope that others will speak up on this namespace issue and I only
wish there was a convenient method to cross post on this thread with the
other groups within which the issue is being considered.
Greg