Re: [Ietf-caldav] Last Call comment on Etag requirements in draft-dusseault-caldav-12

[Julian Reschke <[e-mail]>]

Lisa Dusseault schrieb:
> 
> On Jun 20, 2006, at 10:27 AM, Wilfredo Sánchez Vega wrote:
> 
>>   Not really, no.
>>
>>   HTTP defines ETag.  An HTTP server should be able to use the same 
>> ETag logic on all HTTP resources, and not treat ETags for calendar 
>> resources differently than others.  Not all users of ETags are going 
>> to be aware that calendar resources are special.
>>
>>   My concern is that if there is *any* inconsistency between the 
>> general solution when it comes and CalDAV's, that an implementor may 
>> have to choose between being compliant with CalDAV or the more general 
>> ETag spec, or may have to continue to implement special semantics on 
>> calendar resources for purposes which are better served by the other 
>> spec.
>>
>>   I realize that "the other spec" doesn't exist today, and that this 
>> is a total drag.  Can't we take your one paragraph and put it into its 
>> own document?  I don't know IETF process very well, so I don't know 
>> what the next steps should be, but as an implementor, I'm 
>> uncomfortable with the prospect of dealing with two independently 
>> written specifications for the same behavior.
> 
> We basically tried that.  What it turned into was this 
> <http://www.ietf.org/internet-drafts/draft-whitehead-http-etag-00.txt> 
> with no consensus on the basic model or apparent drive to come to 
> consensus.  Got any feedback on that draft?

It's a shame we didn't make progress on that, but that doesn't 
invalidate the approach taken (making this an issue orthogonal to 
WebDAV, CalDAV or Atom).

Best regards, Julian