From: Ed Gerck <egerck(_at_)NMA(_dot_)COM>
...
It is all about as interesting as
another recent arrival's descriptions of how "we" talked about the
Internet in cafeterias in the old days before it really existed.
Since I made that comment... yes, that is what we (maybe not you) did
back in 1992 when I started to notice and use the Internet. That's only
10 years ago. I guess hindsight is always 20/20, but the idea of
self-regulation needs IMO a grain of salt. At least, we could have
public non-conformance notes.
...
There's nothing wrong in remembering when one first encountered something.
It's not good to urge changes on an organization based on reminisces of
having been around at the beginning despite not having made even the second
generation. 1992 is so long after the old days that it counts as "today."
The current noise not withstanding, there are fewer non-conformance
problems today than there were 1992, provided you only care abount
non-conformance that causes interoperability problems with open standards.
There are now plenty of "public non-conformance notes," particularly
compared to 1992. Open any trade rag and you'll find plenty. That they
are only a little less unreliable than they are incomplete shouldn't be
surprising given what they cost readers. Never mind that they cost
vendors 100 times the $100 that has been proposed here, and that's
assuming the vendor doesn't need to buy advertising or trade show booth
space to be considered. Those of us who have been attended sessions
where more complete and less reliable non-conformance notes are generated
know that they cost orders of magnitude more than $100 even when admission
is free, since you must spend a week or two of senior engineer time.
We also know that their results are rarely public in any meaningful way,
because otherwise vendors could not afford to attend. For example, one
sure way to end Connectathon (is it still alive?) would be to announce
that all of the results would be published.
(1992 was before conformance tests? Sheesh! I think the 1992 Connectathon
was one of the last that I attended. Then there is RFC 1025.)
Those who want any sort of conformance noting, certificating, or
testing should apply to the many commercial and non-profit organizations
in those businesses. All of them charge a lot more than $100 for
anything but membership does. For that matter, mere membership in
the non-profit consortia and forums is usually more than $1000.
It is at best incredibly ironic to rail against minor Microsoft's bugs
while demanding some kind of trademark certificate or conformance
test. Isn't that exactly what Microsoft offers and demands for the
official Windows compatibility service mark?
It's one thing to be open minded and open to change for the IETF
or anything. It's something else to let your brain fall out.
Vernon Schryver vjs(_at_)rhyolite(_dot_)com
(In the interests of full disclosure, I encountered the net in 1972
at the console of TIP 25/DOCB. I was disconnected in the late
1970's and early 1980's. I also don't claim to have ever been in
any cafeterias talking with those who were responsible, at least
not until long afterwards. When I finally met some of them in the
flesh, I hope I didn't pretend inside experience and insight.)