2006/4/25, Robert Sayre <[EMAIL PROTECTED]>: > > On 4/24/06, Bill de hÓra <[EMAIL PROTECTED]> wrote: > > > > I'm -1 to this part. Why do we need to divide safe and unsafe operations > > using URIs, when HTTP has methods for doing that across one URI already? > > I'm also skeptical, but won't stand in the way. Maybe I don't > understand the requirement. Is the thought that authoring clients and > non-authoring clients will be "sharing" the same feeds, so we must > have elements the non-authoring clients will ignore?
If I want to link to an existing entry from one I'm authoring, I shouldn't have to go through the web-site to find it. I should be able to find that entry in my authoring client and ask it to make a link to it, the authoring client should then pick the "non-authoring URI" to make the link. Or are you assuming the server will "translate" the URIs? -- Thomas Broyer
