On Mar 18, Scott Lawrence <slawrence(_at_)globespanvirata(_dot_)com> wrote:
<snip all your arguments that I now accept as being reasonable now I've had
a reasonable intake of Dr. Pepper and cigarettes :-) >
I think you make some good points regarding the ability of independent
developers to find funding. So good that I'm going to make a point to
watch for cases in which it might be in my companies interest to
sponsor such people, and suggest that others might do likewise (this
may already be happening - I don't know).
Perhaps this is adressing the same problem but in a more constructive
manner. The vast majority of web servers out there are going to be Linux or
FreeBSD running Apache. It might be nice if the guys looking after the
stacks in Linux, FreeBSD and the HTTP guys at Apache actually got to come
along without forking out a fortune. I'm not sure what they do now, but I
know that I've seen dicussions on freebsd lists and others where people are
discussing how to implement certain features into some somewhere, where the
conclusion is whoever wrote the RFC should be shot. Obviously, the point of
implementation is the wrong time to start complaining.
I also think that this open-arms attitude to the OSS community might mean a
few more protocols get implemented outside of an individual company's
product range. That has to be a good thing. Well, it is if the protocol or
standard has any value outside of the vanity of a company's R&D department.
The main problem right now though may be one of perception - I've been to
BSD conferences that have cost not dissimilar amounts to IETF confs, but
with more of a social slant. I'm sure that some of the implementors in OSS
world would pull together the money *somehow* to get to IETF if there was a
perceived value. At the moment, the perception seems to be that Cisco and
co. will rubber-stamp things within IETF and the OSS will "just cope with
it" once the RFC appears.
I suppose what I'm saying is, perhaps it's time for the IETF to question how
it should operate in a world where OSS is dominant in the server market
place, and where it is increasingly common for stacks to be implemented by
people who are not paid to do so. It's rather strange really - both Mac OS X
and recent-ish versions of Windows apper to have FreeBSD code in their
stacks. I'll put money on it that few people at FreeBSD have managed to make
it to an IETF meeting.