W3C home > Mailing lists > Public > www-svg@w3.org > November 2004

Re: SVG 1.2 Comment: image/svg+xml;charset=""

From: Chris Lilley <chris@w3.org>
Date: Tue, 23 Nov 2004 15:11:11 +0100
Message-ID: <914842425.20041123151111@w3.org>
To: Ian Hickson <ian@hixie.ch>
Cc: Robin Berjon <robin.berjon@expway.fr>, www-svg@w3.org

On Tuesday, November 23, 2004, 2:19:57 PM, Ian wrote:

IH> On Tue, 23 Nov 2004, Robin Berjon wrote:
>> > 
>> > Because otherwise, as I wrote in the comment, content will be handled
>> > inconsistenctly based on whether the UA supports the MIME type 
>> > directly or just via RFC3023 support. (As seen, for example, with the
>> > W3C validator.)
>> So, enforcing a broken behaviour upon SVG to be consistent with systems
>> that don't support SVG is a good idea? :)

IH> Enforcing interoperability is a good idea. The issue of whether allowing
IH> encoding information to be included in the metadata channel instead of in
IH> the markup itself is a good idea or not is irrelevant; that ship sailed
IH> years ago, with the publication of RFC3023.

Which was then marked as architecturally bad, in the Architecture of the
World Wide Web document; and RFC 3023 is not being revised.

IH>  What matters at this juncture
IH> is making sure that implementations of SVG and XML processors in general
IH> parse SVG documents in the same way.

Which they  do, as long as they do not make up parameters that don't

However, since you like making tests Ian, perhaps you could set up a
test where the XML is in encoding a, the xml encoding declaration says
its encoding b, the charset says its encoding a, and see how many XML
processors will update the xml encoding declaration so that the instance
is well formed locally when saved to disk. I would be very interested to
see results of such a test.

Note that such a test instance would be non well formed when processed
server side. Since XML server side processing happens rather a lot, this
would be a bad idea. This is why the TAG says that duplicate metadata is
a bad thing, and says that the encoding should not be specified if the
document already specifies it.

 Chris Lilley                    mailto:chris@w3.org
 Chair, W3C SVG Working Group
 Member, W3C Technical Architecture Group
Received on Tuesday, 23 November 2004 14:11:12 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:01 UTC