I've been thinking and reading about the IETF and W3C
lately. A lot. A brain-dump of my current thinking:
At Tuesday's
breakfast, I was told that a W3C chartered Atom working group
this month is not a possibility. Nor is next month a
possibility. Perhaps the following month is possible, but
there are no guarantees. By contrast, the
primary reason why the IESG has not already considered setting
up an
AtomPub working group is the comments
sent to the IESG by
Eric Miller of the W3C.
As far as I can tell, this difference in schedule is not
primarily due to a difference in speed or order in which things are
done between the two organizations. Instead it is an
indication of when each got started. The IETF was first
suggested
as a home last June, has been the default assumption ever
since, and
work on a charter started in earnest in February.
Most of the advantages I have seen put forward for pursuing
Atom under the auspices of the W3C apply equally to all recognized
standards bodies in general, and to the IETF in specific.
Furthermore, I have seen nothing put forward which would lead one
to the conclusion that the IETF would be a BAD home.
I am pleased with the clarity of the
W3C RF Licensing Requirements, and the fact that the W3C is
actively pursuing Atom is not lost on me. On the other hand,
I very much like the fact that the IETF is
by design open. I fully realize that a W3C working group
can operate in an open manner, but absent a requirement to
do so, it leaves the chair open to
accusations of bias and reduces the ability of detractors to
obtain adequate
redress.
There is precedent
for a dual IETF/W3C path. If someone can identify both the
additional benefits and the additional costs of pursuing the W3C
presuming a baseline of an IETF working group; and if leaving the
door open for this possibility in July or later doesn't give the
IESG heartburn now; I personally am willing to entertain this
discussion. In fact, I would be willing to make a substantial
commitment of my own time to make this happen.
It is not a question of having to "settle" for one or the
other. Despite reservations, I'm sure that all the issues can
be worked and either organization would make a fine home for
Atom. However, if a choice must be made, and a choice is an
option, the IETF was from the beginning, and remains, my first
choice.
WFM: the default mode for the IETF is to strongly support casual lurkers who drop in when they see something important to them being missed, the W3C claims it's possible for them to support such a thing. Though with the way dropins in the Atom community are greeted with open cries of "where's your #$&*(% spec text?" maybe it doesn't matter ;)
If the Atom standardization process can't be made as open as it is now in W3C, then W3C is no option. IMHO. If W3C can't bend their rules a little to makt this work, it won't work. As easy as that. I have no strong feelings on this, but I've always had my heart at W3C, and feel that they are a bit more «serious» without saying IETF is an «unserious» standards body.
I also feel that Atom is more at home in the W3C, being so close to XHTML, WWW and all. I think W3C has a lot of goodies to offer, and that the Atom specification might «weigh» more if it's a W3C-REC than an RFC. I've always wondered why OASIS haven't been seriously considered as a standards body for Atom, though. Maybe it's just because it hasn't been mentioned loudly enough.
I think that Atom will be a great format and API no matter what standard body it gets embraced by; the only important thing for me, is that the WG is open to non-members of the standards body.
I believe that the ideal situation would be for Atom to have a "home" in IETF yet still benefit from significant participation from W3C. If the W3C members really think that Atom is important, then the fact that it is in the IETF shouldn't keep them away.
In his posting about The Real RSS Debates Robert makes a few useful comments to move discussion of syndication formats and protocols away from the destructive "RSS vs ATOM" topic and on to things people actually need to talk about, despite his...
I haven't seen the precedents, but should mention I've left a few pointers as to why the IETF may not be the best approach at this time.
However, the entrance of interest from the W3C, imv, changes a few things.
Like, one possibility would be for the W3C to take over Atom/RSS 1.0.. while parallel specing continues elsewhere (for example, mebbe the people LEADING the way in the area of actually getting some IETF working group going). These trailers or lurkers or trollers pointing out things that have been missed?? Creates a lot of eddies and blind allies, seems...
I'm looking forward to Brent Simmons' recent spec efforts, as well, because some things like that could make a lotta this here discussion a moot point, again imo/o...
This morning, I got a trackback from a Korean weblog. Unfortunately, if you look at how a typical trackback is sent, you see that character encoding information is not provided. Something to think about the next time you are tempted to think that yo...
[more]