[Standards-JIG] avatars

Richard Dobson richard at dobson-i.net
Fri Jul 2 17:39:06 UTC 2004

> I've been watching this off and on for the past few days, and I'm not
> really understanding where the discussion is going, primarily in
> relation to avatars.  What I took back from my last RFC on the avatar
> wiki was that we needed to figure out OOB, and I would prefer it to be
> in SI.  So then the only question is do we need a generic way to say "I
> know where you can get a chunk of data" in a generic way?  I'm not sure
> how much that actually gains since, using avatars as an example, we
> have an extremely simple wrapper for sipub to just give it a bit more
> context.  Sipub is doing the most of the work and is easy to have as a
> common impl.  So I guess I need clarification on how this is not
> already mostly generic?  It would be nice if some others would comment
> in here too, seems to be a 2 person conversation at this point.

My problem with using sipub is that it is far too verbose and provides far
more information than is really needed certainly for using it with XHTML in
the way XHTML inband objects works, also I would argue that the same is true
for avatars, all that is strictly needed in the initial exchange is a
filehash, all the rest is a complete waste of bandwidth if the user already
has the file in question, which as far as avatars are concerned is a
distinct possiblity for the reasons I have already explained in previous
emails, also sipub doesnt even have a filehash which means that if the
filename is slightly different even tho it is exactly the same file it will
need to be transferred again, filehashes have shown their worth in the P2P
file sharing systems and IMO we should definately be using them in this
instance. Also btw incase you havent read the spec XHTML inband objects does
use SI as it uses the file transfer standard to transfer the file if the
user does not already have it.


More information about the Standards mailing list