- From: Aaron Swartz <aswartz@upclink.com>
- Date: Mon, 08 Jan 2001 19:20:12 -0600
- To: <xml-dist-app@w3.org>
I just finished reading the XP Reqs and I had a few questions about how XP would fit into the web infrastructure. It seems to be acknowledged a popular use of XP will be as a RPC mechanism. However, other XML-based RPC protocols (XML-RPC, SOAP) have been criticized because of their lack of URIs and misuse of HTTP. The major problem seems to be that information only available through some sort of RPC request cannot be given a URI (and thus cannot do any of the things that result from that). Additionally, such systems use a single HTTP method (generally POST) and thus muddy the semantic meaning behind an HTTP method. As an example, there is a SOAP stock price RPC where a SOAP envelope containing a ticker symbols in responded by an XML message containing the stock price. Many argue that such a simple system should be implemented using HTTP GET instead of SOAP. With that introduction, I have the following questions for the WG: 1) Do you feel that such RPC/HTTP services are "bad" and would you recommend against their use with XP? 1a) If so, do you plan to structure XP in such a way that the negatives of such services are lessened? 2) Do you expect to preserve the semantics of HTTP requests (HEAD, GET, POST, PUT) with the HTTP binding? 3) Do you have plans to attach a URI of some sort to an XP message? (It seems that this would certainly ease use cases 805, 807 and 809 (Routing, Tracking and Caching).) Thanks for information you can provide on this subject, -- Aaron Swartz <me@aaronsw.com>| my.info <http://www.aaronsw.com> | <http://my.theinfo.org> AIM: JediOfPi | ICQ: 33158237| the future of news, today
Received on Monday, 8 January 2001 20:21:43 UTC