- From: <bugzilla@soe.ucsc.edu>
- Date: Thu, 3 Feb 2005 10:08:44 -0800
- To: w3c-dist-auth@w3.org
http://ietf.cse.ucsc.edu:8080/bugzilla/show_bug.cgi?id=78 Summary: Value of ETag and getlastmodified properties on multiple bindings Product: WebDAV-BIND Version: -latest Platform: Other OS/Version: other Status: NEW Severity: normal Priority: P2 Component: 03. Properties AssignedTo: julian.reschke@greenbytes.de ReportedBy: lisa@osafoundation.org QAContact: w3c-dist-auth@w3.org Issue: The Bind specification is currently silent on how consistent live properties, including ETag, must be across bindings. This has been discussed somewhat on the list but no agreed-upon language added yet. The behavior of servers here affects greatly how bindings-aware clients might implement synchronization. It's very similar to the interoperability problems that have plagued the WebDAV community over whether the last-modified date and etag should change when the properties of a resource are changed with PROPFIND so we should avoid a similar ambiguity in the case of bindings. Proposed Text: "The ETag and Last-Modified date values refer to the state of the resource, not the state of any or all bindings. Thus, the ETag, the Last-Modified date and the value of the 'DAV:getetag' and 'DAV:getlastmodified' properties MUST NOT vary according to which binding is used to access the resource." Some discussion on the list recently included the notion that since some servers use the URL as part of the ETag, this requirement cannot be made. I don't agree with that personally because implementing the Bind specification requires far more changes than simply how ETags might be generated. Certainly there are successful implementations which use another method to generate ETags which does not suffer from variance according to URL, and those methods can be emulated. However, should the WG conclude that this requirement cannot be made or met, then the Bind specification still needs to include text specifying that the value of the ETag MAY vary across bindings, and whether the value of last-modified MUST be the same. ------- You are receiving this mail because: ------- You are the QA contact for the bug, or are watching the QA contact.
Received on Thursday, 3 February 2005 18:08:46 UTC