- From: Mark Nottingham <mnot@mnot.net>
- Date: Sun, 3 Sep 2006 07:16:23 -0700
- To: www-tag@w3.org
By all appearances, the W3C has become a hotbed of socket API development. A colleague recently pointed out that the SVG 1.2 WD contains a "Raw Sockets" DOM interface; <http://www.w3.org/TR/2004/WD-SVG12-20040510/ #rawsocket>. AFAICT, this is neither covered by its charter <http:// www.w3.org/2004/10/svg-charter.html> or in the referenced requirements document <http://www.w3.org/TR/SVG2Reqs/>. Additionally, the Web APIs WG is chartered <http://www.w3.org/2006/ webapi/admin/charter> to produce: > API specifications for other network communication methods. > Network communication methods covered by this deliverable are > network sockets and possibly protocols other than HTTP. This allows > a Web application to perform more networking operations (eg. IRC, > other instant messaging protocols, Java Message Service and Session > Initiation Protocol). Also, it may be necessary to produce > documentation for, or a specification of, connection policy and > security. > Tentative milestones: First draft of specification during December. > Candidate Recommendation 4th quarter of 2006. IIRC people there have mused about including a raw TCP interface, to permit the most general use. Doing so isn't necessarily a bad thing -- after all, the Web is explicitly more than just HTTP -- but it does raise a few questions; * How do raw TCP interfaces relate to the Web architecture? Last time I checked, there was no "tcp" URI scheme. Should a protocol have a well-defined URI scheme before W3C WGs attempt to integrate it? * More generally, what kinds of protocols are appropriate? E.g., why stop at TCP? Why not UDP? Or raw Ethernet frames? JMS is mentioned, even though it's not a wire protocol -- it's an API specification that is implemented by many proprietary protocols. Is that something the W3C should be encouraging? * What, at a minimum, should such API specs include? The SVG draft specified a raw socket API without including a security model (only hand-waving along the lines of "you know what to do; we'll get to it."), and it still stands that way two years later. * How will such APIs be reviewed to assure security and protocol alignment? Any guidance the TAG could give would be helpful, I'm sure. Thanks, -- Mark Nottingham http://www.mnot.net/
Received on Sunday, 3 September 2006 14:16:28 UTC