W3C home > Mailing lists > Public > ietf-dav-versioning@w3.org > October to December 2001

RE: Baselines and Bindings

From: Clemm, Geoff <gclemm@rational.com>
Date: Thu, 11 Oct 2001 09:21:08 -0400
Message-ID: <3906C56A7BD1F54593344C05BD1374B10488C45B@SUS-MA1IT01>
To: ietf-dav-versioning@w3.org
I believe that capturing the relative names of resources
from other configurations should be required baseline behavior
(i.e. a MUST).  If you could include that in your baseline
extension writeup, that would be great!  If someone comes
up with an important reason for this information not to be
captured, I would be willing to weaken this to SHOULD, but
I believe we should first try for MUST.

Cheers,
Geoff

-----Original Message-----
From: Peter Raymond [mailto:Peter.Raymond@merant.com]

[Geoff wrote]: 
>Although the protocol does not require the baseline to 
>remember the names of members from other baselines, it certainly 
>can do so... 

Thanks Geoff, that makes sense. 
1. Why not make this the recommended behaviour (that servers SHOULD capture
the names of resources 
that appear in other configurations, rather than not include those resources
in the baseline). 
2. Another possibility would be to make this client-driven, eg let the
client send a XML element 
in the body of the BASELINE-CONTROL or CHECKIN of a VCCn request to indicate
that it wants 
names captured for resources that appear in other configurations). 


It seems like such crucial data (in order to make the baseline a consistent
snapshot of the 
baseline-controlled collection). 
Any preferences as to which of the above approaches? 
I could add it to my proposed baseline feature extensions/clarifications? 
Received on Thursday, 11 October 2001 09:21:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 13:57:42 GMT