RE: Submitting DeltaV as Informational RFC? (was: Split ting off core: where we stand)

The fact that an informational RFC is easier to change is exactly why
"proposed standard" is vital for progress with the Versioning Protocol.
The additional stability implied by "proposed standard" is essential
for an organization like mine and Eric's to commit the resources to an
implementation.

It is important to point out here that most of the recent discussions
were about changing the format of the document (e.g. splitting into
two) or about adding new options (e.g. "mutable properties" and
"New-Version" header for PUT), and not about changing the existing
semantics or marshalling.  The main exception is the DTD discussion,
which I hope is resolved via the "per message type" approach, and
which did not receive consensus support as meriting a change to
the protocol.

So I believe that the stability required by "proposed standard" status
has been achieved by the current protocol document.  It is expected
that there will be changes from "proposed standard" to "draft standard",
but these should be the result of real implementation experience
which I believe we will only get if we stabilize at "proposed standard"
status. 

Cheers,
Geoff   

-----Original Message-----
From: Juergen Reuter [mailto:reuter@ira.uka.de]
Sent: Friday, February 09, 2001 5:30 AM
To: ietf-dav-versioning@w3.org
Cc: reuter@ira.uka.de
Subject: Submitting DeltaV as Informational RFC? (was: Splitting off
core: where we stand)


Following the recent discussion about the maturity of DeltaV, I wonder if
it might be an option to submit DeltaV as an informational RFC.  On the
one hand, this might satisfy people that want to see some 'official' RFC
before starting to work on software.  On the other hand, an informational
RFC is probably much easier to change than a proposed standard RFC, which
is a standard tracks document.

Just have a look at the HTTP protocol:  As of May 1996, HTTP was still
issued as informational RFC (RFC 1945), although there had been HTTP
server implementations running since 1990, i.e. for about 6 years.  HTTP
switched to a proposed standard not earlier than in January 1997 (RFC
2068).  The bottom line here is, that HTTP became a proposed standard not
before there were many implementations available revealing all the
bugs and weaknesses of earlier HTTP protocol versions.  In my opinion,
WebDAV suffers from this experience; and DeltaV should do it better than
WebDAV!

Greetings,
            Juergen

Received on Friday, 9 February 2001 07:36:19 UTC