On Thu, Dec 17, 2009 at 10:54 AM, Jonas Sicking <jo...@sicking.cc> wrote: >>> From: Anne van Kesteren <annevk <at> opera.com> >>> Subject: Re: [XHR] LC comments from the XForms Working Group >>> Date: 2009-10-08 15:31:27 GMT >>> >>> On Tue, 17 Jun 2008 05:24:48 +0200, Boris Zbarsky <bzbarsky <at> mit.edu> >>> wrote: >>> > Anne van Kesteren wrote: >>> >> It would change the conformance criteria. I'm not sure that's a good >>> >> idea. Especially since the use case put forward is mostly theoretical. >>> >> Overall, I'm still not convinced this is a good idea. >>> > >>> > It doesn't seem necessarily that theoretical to me, for what it's >>> > worth. Anne, do you happen to have a more or less complete list of the >>> > current dependencies of XHR on Window, buy chance? I think that >>> > information would be very helpful in seeing where things stand. >>> >>> To wrap this up, I changed XMLHttpRequest some time ago so it can be used >>> in other contexts as well now. If you reuse it you have to define the >>> XMLHttpRequest origin and XMLHttpRequest base URL. >>> >>> My apologies for being a bit stubborn on this earlier. It was mostly >>> because I was hesitant reworking how everything was put together, but it >>> turned out that had to happen anyway. >>> >>> Hopefully it can now be of use to the Forms WG. >>> >>> Kind regards, >>> >>> -- >>> Anne van Kesteren >>> http://annevankesteren.nl/ >>> >> >> So, to be clear, here's how do complete the change for the specific >> dependency that Anne calls about above. >> (This process is repeated for each dependency of XHR on HTML5.) >> >> Cf. section >> http://www.w3.org/TR/2009/WD-XMLHttpRequest-20091119/#origin-and-base-url >> >> Each XMLHttpRequest object has an associated XMLHttpRequest origin and an >> XMLHttpRequest base URL. >> >> This specification defines their values when the global object is >> represented by the Window object. >> When the XMLHttpRequest object used in other contexts their values will >> have to be defined as >> appropriate for that context. That is considered to be out of scope for >> this specification. >> >> This text still results in a normative reference to HTML5. So change the >> XHR document to this: >> >> Each XMLHttpRequest object has an associated XMLHttpRequest origin and an >> XMLHttpRequest base URL. >> >> This specification does not defines their values; they MUST be defined by >> the host integration. >> For an example integration with [HTML5 informative reference] see [XHR For >> HTML5 informative reference] >> >> Further, the actual definitions would be removed when the actually occur. >> >> Then the new rec-track document XHR for HTML5 would say this: >> >> Each XMLHttpRequest object has an associated XMLHttpRequest origin and an >> XMLHttpRequest base URL. >> >> This specification defines their values when the global object is >> represented by the Window object. >> >> And then go on to cite contain the actual text of the definitions pulled out >> from XHR. > > Ah, thanks for the concrete example. This makes it clear what you are > suggesting. > > What you are saying makes sense. However it seems to add unnecessary > overhead to split the spec in two to accomplish this, for the spec > editor, for someone implementing the spec, and for someone using the > spec. It would seem to be much lower overhead to put these things in > an appendix or something similar.
Though I just realized that I'm not sure all dependencies can be solved this way. How would you for example break the dependency on the event loop, currently only specified in the HTML5 spec (but implemented in basically every piece of software with a modern UI)? / Jonas
