24148 2013-12-20 14:36:41 +0000 Reference the URL spec 2014-01-24 23:35:20 +0000 1 1 1 Unclassified SVG Linking SVG 2.0 PC All NEW https://www.w3.org/Bugs/Public/show_bug.cgi?id=23968 P2 normal Test Suite 1 zcorpan schepers dschulze ian www-svg oldest_to_newest 97827 0 zcorpan 2013-12-20 14:36:41 +0000 See bug 23968 which is the same issue but for CSS. As for the url query encoding situation in impl, it's as follows: d.e. = document's encoding Element Gecko Blink Presto a d.e. d.e. d.e. feImage d.e. ? utf-8 image d.e. utf-8 utf-8 use d.e. d.e. utf-8 (IE not tested.) HTML uses the document's encoding for almost everything. It's unclear whether we should be consistent or use utf-8 as much as possible. Hixie argues for the former. 97828 1 dschulze 2013-12-20 16:23:45 +0000 Do you mean the URL spec of the WHATWG? Read the latest status document (as far as I know) about stability of normative references (which we need here). I do not think that the WHATWG spec currently fulfills the criteria. 99029 2 ian 2014-01-24 21:29:35 +0000 The WHATWG URL standard is more stable than any W3C or IETF reference on the subject... (and more importantly, more accurate, too) 99044 3 dschulze 2014-01-24 23:35:20 +0000 (In reply to Ian 'Hixie' Hickson from comment #2) > The WHATWG URL standard is more stable than any W3C or IETF reference on the > subject... (and more importantly, more accurate, too) That is great! And still the specification changes nearly everyday and does not fulfill the requirements of the W3C. Which would not be a problem at all, unless you actually want a W3C spec to get a REC at some point. Would there be daily (but of course then outdated) snapshots that can be referenced, the requirement would be fulfilled and we can finally reference WHATWG specs.