Re: Arch Doc: 11 November 2003 Editor's Draft


First off, I wanted to say that I quite like the latest draft; it's come
a long way, not just with content, but in the structure and quality of
the editing.  It's a good read.  Kudos.

I just have a couple of comments I wanted to make before your f2f ...

The document adequately addresses all of what *I* (YMMV 8-) would
consider the major visible architectural points of the Web, except for
one; the constrained interface (i.e. GET, PUT, POST, etc..)

My position is that it's *extraordinarily* valuable, and that Web
developers, including Web services developers, SHOULD use it, even for
machine-to-machine integration.  Whether the TAG agrees with that
or not though, I think some further guidance here would be valuable,
because it is a question that developers are grappling with today.  I
say "further", because there is already some guidance given on this
topic with respect to the value of GET, which is great.  But that's
just one operation (although arguably the most important one).

Another comment is that I feel that before going to last call, that the
issue I previously raised[1] regarding the (mis?)use of RFC 2119, needs
to be resolved.



Mark Baker.   Ottawa, Ontario, CANADA.

Received on Friday, 14 November 2003 13:56:54 UTC