- From: Dan Brotsky <dbrotsky@Adobe.COM>
- Date: Thu, 3 May 2001 13:41:10 -0400 (EDT)
- To: Peter Raymond <Peter.Raymond@merant.com>
- Cc: ietf-dav-versioning@w3.org, w3c-dist-auth@w3.org
>I am in favor of Lisa's suggestion of having three timestamps.
I'm not sure Lisa was actually making that suggestion :^), but in any
event I don't like the idea of requiring this kind of timestamping at
all. It adds detail to the abstract "DAV resource" that
unnecessarily biases it towards file-system implementations (where
content and properties are likely to be stored separately and
modified independently).
I believe that the modification date property is in DAV because it's
there in HTTP 1.1. And its semantics are left vague exactly to allow
many different implementations.
The move to etags in 1.1 was specifically intended to more generally
answer the question "has the content of this resource changed." We
shouldn't now go back and try to resurrect modification dates as the
way to find out about content changes.
Similarly, I don't think we should tie modification dates one way or
the other to property changes. if clients need a way of asking "has
this resource changed in any way - content or properties" then we
should add a "property etag" in addition to the "content etag" that
now exists.
dan
--
Daniel Brotsky, Adobe Systems
tel 408-536-4150, pager 877-704-4062
2-way pager email: <mailto:page-dbrotsky@adobe.com>
Received on Wednesday, 9 May 2001 10:54:56 UTC