- From: Bjoern Hoehrmann <derhoermi@gmx.net>
- Date: Wed, 11 Jan 2006 23:04:45 +0100
- To: "Jim Ley" <jim@jibbering.com>
- Cc: "Robin Berjon" <robin.berjon@expway.fr>, <www-svg@w3.org>
* Jim Ley wrote: >you don't, that's clearly an error, but if you just go for rules like >"remove leading and trailing whitespace", then what do you do with things >like: > >color="#ff 00 aa" > >that's easily correctable too, and if you then write the rules that say how >to correct that what do you do with the next example, and the next, and the >next. If these rules are left up to implementations (and I would of course >encourage implementations to agree on error recovery like this) then more >things can be recovered than those that are thought of now. If the specification does not define how to process the above, and you allow implementations to process it in any way they choose, the effect is that you allow implementations to extend the semantics of the format which later limits how *you* can extend the semantics of the format. A simple example is <rect height="100%" width="100%" /> This is perfectly legal SVG 1.1 content, but Tiny does not support this and e.g. TinyLine renders this as if <rect height="100" width="100" /> had been used, which is very different from how the specification does require to process it. If you add new units <rect height="100vh" width="100vw" /> TinyLine and ASV6 both process it as if <rect height="100" width="100" /> had been used, which doesn't quite match the definition of the units. If authors don't know how older implementations deal with unsupported features, creating content that uses new features but degrades grace- fully in older implementations becomes very difficult. Likewise, it might be difficult to introduce new features because implementations want to be backwards-compatible with their old error recovery behavior which contradicts the new features. -- Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de Weinh. Str. 22 · Telefon: +49(0)621/4309674 · http://www.bjoernsworld.de 68309 Mannheim · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/
Received on Wednesday, 11 January 2006 22:04:26 UTC