08:42
<Lachy>
hey, I'm working on my presentation (which I'll be presenting on 3 Aug) and I'm trying to list and compare the benefits of using HTML vs. XHTML and explain when each is appropriate.
08:42
<Lachy>
Any ideas about the benefits and use cases for XHTML?
08:44
<hsivonen>
Lachy: first, I'm assuming you mean XHTML 1.0:
08:44
<hsivonen>
Lachy: embedding SVG and MathML
08:44
<Lachy>
no, XHTML5
08:44
<hsivonen>
ooh.
08:45
<Lachy>
the presentation is called Developing with HTML5
08:45
<hsivonen>
well, anyway, embedding SVG and MathML in it
08:45
<Lachy>
yeah, got that one already (that's the only one I had)
08:45
<hsivonen>
embedding it into XSLT transformations (for producing HTML5 or XHTML5)
08:45
<Lachy>
also XSLT, but I'm not sure if that's really a benefit ;-)
08:45
<Lachy>
I suppose, some people might want to do that
08:46
<hsivonen>
using the XML representation as the internal representation in non-browser apps
08:46
<Lachy>
like in a CMS?
08:46
<hsivonen>
that's not syntax but using streaming or tree representations
08:46
<hsivonen>
yes
08:47
<hsivonen>
or in a conformance checker :-)
08:47
<Lachy>
yeah, but most developers are building a conformance checker
08:47
<Lachy>
*aren't
08:48
<hsivonen>
well, any app that wants to do non-browser things with HTML5 and wants to do so in a robust way could use an XML pipeline with an HTML5 parser on input and an HTML5 serializer on output
08:48
<hsivonen>
which means there's no XML *syntax* involved but the APIs / data models are
08:48
<Lachy>
I'll see if I can turn that into some kind of flow chart, showing the authoring in (X)HTML, storing as XHTML, and serialising as HTML5 to the client
08:59
<jgraham>
Yeah, for something like Genshi the XML serialisation will work better
09:00
<jgraham>
http://genshi.edgewall.org/ (Python templating language that uses SAX-like streams internally and uses a subset of XInclude to process fragments)
09:04
<hsivonen>
Lachy: after all, all the XML toolchain stuff is pretty cool. it's just that the server-to-browser step doesn't work in IE and the ingestion step to the pipeline is brittle if using a real Draconian XML parser
10:14
<rabies>
Morgen.
16:53
<Navarr>
Pardon for the probably very odd question; but what is the point in HTML5 being made if its less strict? Isn't the point of standards to make them more strict to follow a specific set of rules so that browsers read and display them correctly on a universal level?
16:54
<gavin_>
depends on what you mean by "more strict"
16:54
<gavin_>
it's important that the UA requirements are completely and consistently defined
16:55
<gavin_>
the authoring requirements don't need to be "strict" to achieve interoperability
16:56
<Navarr>
but with the advancing work of XHTML to combine into XML to have further freedom in what a user does, why continue HTML? (these are very basic questions, being asked by a 16 year old interested in the web)
16:56
<gavin_>
I'm not sure I understand your assertion
16:57
<Navarr>
well, XHTML 2.0 is being worked on by the W3C, allowing the combination of HTML (written using XML schema?) to combine with other types of XML (SVG, MathML,ect.) why continue HTML?
16:58
<gavin_>
the goal of the XHTML2.0 working group isn't to "combine HTML with other types of XML", as far as I know
16:59
<gavin_>
their goal is to rewrite HTML, and they're doing it in a way that is incompatible with the web
16:59
<Navarr>
ah, i see.
16:59
<Navarr>
Thank you very much, I'm just kind of curious.
17:01
<Philip`>
About 99.95% of the web uses HTML instead of XHTML, so work on HTML is more relevant than XHTML for the majority of users and authors
17:03
<Navarr>
Isn't it also true that most of the web uses wrongly formatted HTML?
17:04
<gavin_>
yes
17:04
<Philip`>
http://triin.net/2006/06/12/HTML suggests about 97.5% of pages are invalid
17:04
<Navarr>
wow.
17:05
<Navarr>
So, what will effectively be the difference between HTML5 and XHTML2?
17:05
<Philip`>
There are basic syntax errors on about half of pages, before even looking at whether they're using real HTML elements and using them correctly
17:05
<gavin_>
http://blog.whatwg.org/faq/ will probably answer a lot of your questions
17:06
<Navarr>
thank you ^^;;;
17:06
<gavin_>
though it's perhaps a little bit biased :)
17:07
<Navarr>
yea
17:07
<Navarr>
i see it has been worked on to try to make it less biased
17:16
<Navarr>
thank you for that
22:49
<annevk>
hmm, 1848 new e-mails
22:49
<zcorpan>
annevk: wb
22:51
<annevk>
anything new?
22:53
<zcorpan>
annevk: i updated http://html5.googlecode.com/svn/trunk/parser-tests/