-----Original Message-----
From: Simon St.Laurent [SMTP:simonstl@simonstl.com]
Sent: Saturday, October 14, 2000 1:00 PM
To: xml-dev@xml.org
Subject: RE: sunshine and standards development
> Competition with the W3C in those sectors is simply not feasible in a world
> where W3C Recommendation=Standard. Right now, that simple equation, which the
> W3C itself officially disowns, is definitely in operation. While RELAX is
> making a bold move through ISO, I'm not sure whether ISO can compete with the
> W3C for developer mindshare.
A W3C Recommendation is not a "standard" any more than the Windows API is a "standard" or Java is a "standard". They have no legal standing (unlike ISO or ANSI specifications), they are wide used if they they are useful, and useful if they they are widely used. Actually, I suspect that virtually no W3C Recommendations are fully implemented by more than a handful of vendors -- Look at the recent discussion of the ownerDocument property on DOM nodes, or the perennial question of XML parser conformance. I don't see this as a big problem, because W3C Recommendations are really more like treaties among competitors to explore the "solution space" in a way that causes consumers the least inconvenience than "standards" that must be adhered to.
Since Recommendations such as XML useful because they are used (and vice versa), I don't see the barriers to entry that Simon does, i.e., that Relax can't compete with W3C Schema for developer mindshare. SAX is the obvious example of an ad hoc group coming together to meet a common need much faster than the W3C or IETF could act; SOAP might be another, and XQL yet another. I've heard it said that XML is almost another example, having come in "under the radar" before the full-blown W3C process became cast in concrete. Ad hoc solutions that are clearly good starting points tend to get adopted as de facto standards.
So, I see a continuum here from:
* Ad-hoc specs that come from a small group, often but not always
with a "major power" sponsor.
* Pragmatic "let's compromise for the good of the consumer" specs from
industry consortia, or de-facto standards from a dominant vendor.
* "Real" standards that rigorously define a vendor-neutral specification
and strict conformance criteria.
I believe that it's obvious that "sunshine" is extremely important at the ad hoc end of the spectrum; when people are brainstorming a problem, limitations on the free exchange of information or the range of possible solutions are clearly detrimental. On the other hand, "sunshine" probably probably hurts when it's time to just broker a deal to get a real standard cast in concrete.
[I keep thinking of historical analogies ... such as the autocratic Bismarck keeping Europe more or less at peace for decades in the 19th century, but having it all fall apart once more open politcs limited diplomats' room to maneuver in the 20th; or perhaps the fact that many "charismatic" US presidents such as Lincoln, Roosevelt or Kennedy would be unelectable with all the "sunshine" in the political process of today; instead we get to choose between people whose main accomplishment is to have avoided taking a firm stand on anything.]
In this scheme of things, the problem is that people see the W3C as both crafting technologies
out of thin air and having their Recommendations treated as if they were standards. These activities MUST be separate, as much as it would be convenient for us all to move directly from revolutionary technology advances to universally accepted standards without a period of chaotic natural selection in the middle. I personally would like to see the W3C more open to outside involvement at the "ad hoc" end of the continuum, more determined to take time to get specs right (with clear prose, extensive implementation experiene, etc.) before they become Recommendations, and perhaps more willing to defer to the ISO on the really messy business of casting Recommendations into concrete Standards. Nevertheless, as long as the W3C serves as the "treaty organization" within which pragmatic compromises that result in consensual Recommendations are made, I don't believe that more "sunshine" would help.
