- From: Jim Ley <jim@jibbering.com>
- Date: Wed, 1 Dec 2004 18:27:48 -0000
- To: <www-svg@w3.org>
"Boris Zbarsky" <bzbarsky@MIT.EDU> > Again, this is the sort of issue we want to avoid in SVG. Absolutely! hence why I think script inserted in this shouldn't be specified, as you say the current wording leaves opent the ability to add script nodes whilst the document is loading via script, and that also needs to be specified or outlawed, I strongly believe outlawed, since specifying it is way too hard. >> (where addChild is an appropriate script to insert that HTML into the >> current document) > > I'd have to trace through this to be sure, but I suspect fred() will call > confirm() in this case. Interesting, that would mean that appendChild and document.write behave differently - I should really look into it myself, but this sort of stuff is best avoided due to inconsistencies - you also I think need to define behaviour to get around things like http://jibbering.com/2004/10/loopcrash.html (will crash recent mozillas) entering infinite loops. >> There's a similar problem with every working group of course, there's no >> CSS2 errata and no SVG 1.1 errata > > Excuse me? Whilst there are indeed no SVG 1.1 errata, the CSS2 errata are > found at > <http://www.w3.org/Style/css2-updates/REC-CSS2-19980512-errata.html>, are > quite extensive, and are clearly linked from the CSS2 specification (see > the "Errata" section at <http://www.w3.org/TR/CSS2/>). No, that's a "proposed errata", it's not an errata, at least it admits to not being maintained which is an advantage over other working groups. > It doesn't in recent versions, as far as I can tell. No, does in 5.5, not in 6 after checking. Cheers, Jim.
Received on Wednesday, 1 December 2004 18:58:50 UTC