> Jim W: > > Maybe we have agreement. The key word for me is > "required". I don't think properties should be > REQUIRED for search arbiters. I never said they > should be DISALLOWED. The spec., being quiet on > this issue, accommodates my position (it doesn't > say they are disallowed, and it doesn't say that > you can't have them). Is your position that the > spec. should require them, or is it OK to be > quiet on that issue for the first release? I think search arbiters should be required to support the RFC 2518 base set of properties. If they do, downlevel clients will be able to see search arbiters in collections, and will be able to tell that it is not a resource type they understand. I see this limited property support being important for interoperability with downlevel DAV clients. If the spec. is silent on the properties issue, I feel it will cause interoperability issues, since I can easily see a client being developed that depends on the property information being present, and behaving poorly if it is not. - JimReceived on Thursday, 24 June 1999 20:08:43 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 22 March 2009 03:38:04 GMT