I am uneasy about a single URL being effectively bound to two resources (GET and GETSRC return different resources). It opens a potential can of worms in that every HTTP method (HEAD, PROPFIND, LOCK, ...) then potentially needs to be duplicated. And how do I know that GETSRC will work? (Is it defined for all resources no matter what type they are?) Personally I would feel more comfortable with either the current 'source' scheme, another HTTP header field applicable to all requests, or a parameter on the URL or something (/foo.html and /foo.html;getsrc). I can see the cleaness of making one URL computed from the other (identical in the proposed scheme, computed in the ;getsrc parameter scheme above) but I dislike one URL identifying multiple resources and the riple through all the other methods that would be required. AlanReceived on Monday, 29 October 2001 21:58:35 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:43:58 GMT