- From: Thomas Roessler <tlr@w3.org>
- Date: Sat, 9 May 2009 11:48:29 +0200
- To: WSC WG public <public-wsc-wg@w3.org>
Looking at Yngve's message here: http://www.w3.org/mid/op.us37ivemvqd7e2@killashandra.oslo.opera.com ... I wonder what the next step is. I think we have the following open points: 1. We need to add a MAY clause after the first paragraph of 5.4.1: > When TLS error conditions occur, user agents MAY choose to abort the > connection without any further user interaction. The guidelines in > this section apply when user agents choose to cause a user > interaction in the case of TLS error conditions. 2. We might want to make it more clear above that a user agent can very well both follow the detailed guidance in 5.4.1 (e.g., for the top-level frame), and choose to do the hard abort in some cases -- e.g., XMLHttpRequest, and that the latter is indeed the recommended course of action. 3. I wonder whether we want to come down more strongly on what choice user agents should make when for inline elements and XMLHttpRequest. Yngve's note above suggests that there is no uniform approach right now. I don't know whether we need to attempt unifying it, or whether the approach given above is enough. Thoughts? -- Thomas Roessler, W3C <tlr@w3.org>
Received on Saturday, 9 May 2009 10:34:16 UTC