Re: [Evolution-hackers] calendar implementation (Partly OT)

[stephen farrell <sfarrell almaden ibm com>]

CAP is vaporware.  Why not just enable the IMAP store to save calendar
events?  Accepting an appointment would just be a matter of adding a
header to the mail that contains it.  Creating your own appointment
would be done by putting the message directly into the IMAP folder.
There are performance implications (IMAP servers don't know the
semantics of events, so can't be queried for all events within a time
range), but they can be worked around in the short term by leveraging
the local cache and in the long term by extending IMAP to support more
search semantics.

IMAP supports the basic requirements (network storage of messages with
sharing support), and of course there are many robust free imap
servers out there.  FBURL support doesn't come for free, but could be
shoe-horned in with a simple cgi script.

I know this has been talked about before and "someone dismissed
it"... but if HTTP PUT and webdav can suffice in the real world for
many people to publish calendars, then surely a protocol that supports
shared folders, storing events in multiple folders, could be made to
work.

Also, the innovation potential is large -- email provides a huge
wealth of state information for people.  It's common for people to
send themselves todo items, documents they don't want to lose or they
want to access remotely, etc.  If appointments and mail were tightly
coupled, then it would be easy to do things like create a todo item
*inside* an interesting thread of messages, transform an email that
says "let's meet next tuesday" into an appointment, or annotate email
by using the same primitives as threading (namely "references").  Why
just be a clone outlook?

Here's the classic "email as habitat":
http://www.sims.berkeley.edu/~nicolas/documents/Ducheneaut-Bellotti-Email_as_habitat.pdf

And this guy seems to have some good ideas too:
http://www.imedia.mie.utoronto.ca/~jacekg/pubs.html

JP Rosevear <jpr ximian com> writes:

> On Thu, 2003-04-10 at 08:57, Ron Smits wrote:
>> Grin
>> 
>> I read that page, and it is more then a year old. So I was hoping some
>> more had happened. It hasn't
>> 
>> If you 'just point it to a published file' your info is always out of
>> date. The freebusy url I mentioned in my previous posting is already out
>> of date.
>> 
>> I was hoping to have a application server running somewhere that
>> evolution can talk to. Alas I have been searching the net for more then
>> a day now and I cannot really find anything except real good ideas, the
>> rfc's and that's it.
>> 
>> I would really like to work on an open source project geared towards an
>> standards based calendar server. But after reading a lot of info I would
>> not even know where to begin protocolwise? BEEP looks nice, but very
>> theoretically. HTTP might be much easier to use but it seems the IETF
>> people do not want that.
>
> The ietf standards people are building CAP (Calendar Access Protocol).
> http://www.ietf.org/internet-drafts/draft-ietf-calsch-cap-10.txt
>
> There are some projects started to support this I believe but I'm not up
> to speed on them.
>
> -JP
> -- 
> JP Rosevear <jpr ximian com>
> Ximian, Inc.
>
> _______________________________________________
> evolution-hackers maillist  -  evolution-hackers lists ximian com
> http://lists.ximian.com/mailman/listinfo/evolution-hackers