On 2011-07-28 at 14:07:12, Bill Roome wrote: > Actually, here's my proposal in more detail. Each entry in the > resource directory would have one -- AND ONLY ONE! -- media type, and > 0 or 1 accept types. That's sufficient to define the service that > entry provides and the method it expects.
That sounds reasonable, and it does resolve the problem that you don't know what combinations of request and response content type are supported. However, it doesn't reduce the complexity of the complete model. However, if you have a statement that if a resource is indicated with multiple, then it must support all combinations of requests. Given that, the affect on an implementation that doesn't want to deal with this can be minimized if you are careful. > I'd also suggest that the resource directory MUST be authoritative (at > least within any Expires time). Sadly, that's not possible. Even if the directory and the indicated resource are on the same server, there is no guarantee. > Given that, is OPTIONS still necessary? Sadly, yes. However, you are always free to not request it. Similarly, I expect that clients will have to survive if a resource doesn't support it. --Martin _______________________________________________ alto mailing list [email protected] https://www.ietf.org/mailman/listinfo/alto
