- From: Dave Hodder <dmh@dmh.org.uk>
- Date: Sun, 15 Dec 2002 14:20:48 +0000
- To: www-html-editor@w3.org
Here are some of my thoughts on the sample code used throughout the
present XHTML 2.0 working draft:
1) Legacy media (MIME) types and file extensions. Code examples have
quite a lot of references to files with '.html' extensions, plus
there's an occasional mention of the "text/html" media type.
E.g. 11.1.3 has an example reading:
<head>
<title>Reference manual -- Page 5</title>
<link rel="Start" title="The first page of the manual"
type="text/html"
href="http://example.com/manual/start.html"/>
</head>
Personally I think no references to the legacy HTML media type
should be made in this way; '.xhtml' and "application/xhtml+xml"
should be used instead.
(In practice, all web servers associate '.html' with "text/html",
and recent versions of Apache 2, thttpd, etc. associate '.xht' and
'.xhtml' with "application/xhtml+xml". This will probably cause
confusion for a lot of web authors in the future, and the more that
can be done to address that now, the better. Of course, I realise
no firm decision has been made in public as to whether XHTML 2 will
use the "application/xhtml+xml" media type, whether it will use a
special profile to identify itself, etc.)
2) Line breaks within attribute values. I was always lead to believe
that this was bad practice, but the 'summary' attribute in 18.7
does this. Any thoughts?
3) 18.3.4. Sample table. The text example appears to have a
superfluous '=' on the bottom row.
4) The example of the 'meta' element in 12.1.1 has comments that start
with "<--" instead of "<!--".
5) People and organisations in examples. Some examples appear to
refer to real people (Jakob Nielsen, J.R.R. Tolkien, Steven
Pemberton) whilst others seem to refer to fictional people (John
Doe). Whilst not a major point, could it be made more consistent?
Thanks,
Dave
Received on Sunday, 15 December 2002 09:24:51 UTC