W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > July to September 2007

Re: [Ietf-carddav] Comments on draft-daboo-carddav-02

From: Stefan Eissing <stefan.eissing@greenbytes.de>
Date: Tue, 17 Jul 2007 09:26:01 +0200
Message-Id: <45452A69-1A66-41AC-AC97-6D09190E578C@greenbytes.de>
Cc: Cyrus Daboo <cyrus@daboo.name>, Wilfredo Sánchez Vega <wsanchez@wsanchez.net>, WebDAV <w3c-dist-auth@w3.org>, ietf-carddav@osafoundation.org
To: Julian Reschke <julian.reschke@gmx.de>


Am 16.07.2007 um 23:13 schrieb Julian Reschke:
> Cyrus Daboo wrote:
>> Whilst that tells you if you have a conflict, you still end up  
>> having to take a guess at another resource name to use to actually  
>> save the new data. Hopefully, the second guess will work.
>> The one benefit I have heard to the POST/ADDMEMBER approach is if  
>> the server has limitations on resource names. Typically those are  
>> servers that proxy their data to/from another, more rigid, system.
>
> Yep. And that's why I think ADDMEMBER would be useful. POST just is  
> too generic.

I think a client would need context for both methods. If you think of  
adding ADDMEMBER to a generic client like "curl", it's not much more  
it could do besides what it already does for POST.

If you talk about a "higher level" client (e.g. more aware of the  
context it works on), the method name again does not make a big  
difference. If AtomPub would have used ADDMEMBER instead of POST,  
what improvement exactly would one expect from that?

Cheers,

Stefan
Received on Tuesday, 17 July 2007 07:26:45 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 10 December 2014 20:02:55 UTC