Hi Mateusz,
Mateusz Biliński wrote:
* there were a few attempts to build a specification. I list them
below in the order (I think) they've appeared:
The oldest (last update:2001) approach to specify a standard for
whiteboard (I don't suppose this is used. but I'm not sure):
http://www.xmpp.org/extensions/inbox/whiteboard2.html
I believe the authors of the proto-XEP still use something based on it.
XEP-0113: Simple Whiteboarding (this one's deferred)
http://www.xmpp.org/extensions/xep-0113.html
An SVG Based Whiteboard Format (this could be used with the next one
-sxde - to build a standarized whiteboard?):
http://www.xmpp.org/extensions/inbox/whiteboard.html
Correct. This just defines how the whiteboard should be represented in XML.
However, no one actually uses this particular format at the moment. The
only implementation that uses SXE (Psi) uses a regular SVG 1.1 document
right now (no pages, etc).
Shared XML Document Editing (a generalized approach, I suppose this
one was substituted by the next one):
http://www.xmpp.org/extensions/inbox/sxde.html
Correct.
Shared XML Editing (This is by far the most recent approach - this is
going to a fundamental standard for all applications of shared XML
editing. This is the one that team is currently working on.):
http://www.xmpp.org/extensions/inbox/sxe.html
It's the most recent approach and Psi currently implements this version
almost exactly except for the negotiation part, which doesn't yet use
Jingle.
It _could_ serve as the fundamental protocol for all applications that
need to share an XML document. However, so far the Council hasn't
accepted any of the proto-XEPs so who knows what _will_ happen...
* stated above there are two types of discussions around:
- whether the 'shared XML editing' isn't too generalized approach
for whiteboard (but AFAIK this SXE would only be a common _base part_
for all applications that would use shared editing - or am I
wrong?)
- how to solve concurrency problems, especially when users attempt
to edit the same object in parallel
I think this would be it. Maybe I have missed something - if so,
please point this out.
Finally here are my two questions:
- what is currently _the best way_ to implement whiteboard (in this
case in Gajim). Should it be based on SX(D)E? Or maybe it should be
based on currently working implementations in other clients so that it
could interoperate with them? (I guess the first one is more sensible
in current situation)
- what are the chances that this project will be chosen by XSF for
GSoC2008. What I am worried about is whether currently there is any
sense in implementing this if there is no official specification for
whiteboard. What I mean is that I would really like to do this, but
what is XSF opinion about this.
My current proposal would be to implement SX(D)E in Gajim as
whiteboard and point out pitfalls that I fall in to during coding
which could help build a specification draft. But I suppose Jonnas
Govenius is working on this in Psi, so this falls back again to the
question: is this project sensible for GSoC in XSF opinion.
I suppose these last questions are a little off topic for the list but
I'm very curious about the answers myself. In fact, I'd like to work on
this stuff again for GSoC but I have a feeling the Council will want to
give the chance to someone new (like you!)...
Joonas
