Request for feedback: WebDAV property datatype draft

Hi,

we (greenbytes & SAP) have been working with typed WebDAV property 
values for quite some time now. The Internet Draft describing the 
extension has been stable for a long time, the recent changes describing 
mostly optional features such as:

- some per-property flags (protected, hidden)
- property displaynames
- WebDAV SEARCH operators for multivalued properties

Recently I learned that another well-known server vendor (Xythos) is 
going to support the basic parts of the protocol as well.

We have dicussed making the draft a working group deliverably several 
times now (I think last time in spring '2003). As far as I can tell, 
many of us did agree that we should do this, but there has been 
disagreement about the scope (many would like to see property schemas, 
discovery of types, and so on...). Although all these things are 
interesting, they are *hard* to define and to get consensus on. On the 
other hand, the base protocol itself is simple and can exist even in the 
absence of schemas, resource types and other extensions.

Given the time constraints (as a community) that we apparently have, it 
IMHO makes sense to publish an RFC about those things that are already 
well understood and implemented by multiple parties; and then leave all 
the other interesting stuff for future developments (maybe we can get 
back to that once we have RFC2518bis, BIND, REDIRECT and possibly 
LOCKING out of the door).

Therefore I'd like to propose to take
	<http://greenbytes.de/tech/webdav/draft-reschke-webdav-property-datatypes-latest.html>

as a base, removing all nice-to-have-but-non-essential features that 
were added since

<http://greenbytes.de/tech/webdav/draft-reschke-webdav-property-datatypes-02.html>

(being the aforementioned property flags, displaynames, and SEARCH 
operators) and publish this as an "experimental" RFC.

It would be nice if the working group would adopt it; but it's not 
really essential.  Otherwise, publishing it as a private submission as 
"informational" RFC would make a lot of sense (after all, it's a useful 
addition to WebDAV properties, and there already exist independant 
implementations).  Note that in this case, the IETF is going to come 
back to the working group and it's chairs asking whether this 
specification collides with possibly ongoing working group activities; 
thus we should discuss this before; see RFC2026, section 4.2.1).


Best regards, Julian

-- 
<green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760

Received on Friday, 9 July 2004 16:11:44 UTC