W3C home > Mailing lists > Public > public-html@w3.org > September 2007

Re: The only name for the xml serialisation of html5

From: Leif Halvard Silli <lhs@malform.no>
Date: Sat, 29 Sep 2007 21:18:54 +0200
Message-ID: <735416fbfedb8a4e76b8ffe26e8cd43d@10013.local>
To: Dean Edridge <dean@55.co.nz>
Cc: Adam Nash <adamn@wirespring.com>, public-html@w3.org

2007-09-29 03:59:42 +0200 Dean Edridge <dean@55.co.nz>:

>[... ] the most obvious and logical naming conventions for our spec:

[ ... later you added 'simple' and 'semantic' as arguments ...]
> HTML 5 (text/html)
> XHTML 5 (application/xhtml+xml) or (application/xml)

_Why_ this is is logical, semantic and simple seems to boil down to 3 reasonable ideas: 
	1. MIME: A link between MIME type + serialisation name is logical
	2. History: Let's continue to operate with HTML versus XHTML
	3. HTML: As you and Charles later agreed, 'html' should be part of the name for both variants of HTML 5

The possibly problematic point here is 2): Discerning between HTML and XHTML has not been a success till now. A good name should help us get rid of that (and any other) confusion. Otherwise I agree with 1) and 3).

But it might also be worth pointing out that neither 'HTML 5' nor 'XHTML 5' occur in the current draft (well, 'HTML 5' occurs once - in the title). While there are 12 matches for 'HTML5' and 9 for 'XHTML5'. (Plus one occurrence of 'html5' in a domain name.) This point is important: What kind of name are we actually looking for? (I also think that the HTML 5 specification draft should use 'HTML 5' and not 'HTML5'.)

I think that we should just use 'HTML 5' for both variants - as well as when we speak about the semantics of HTML 5. Then, whe we want to refer spesificly to the text/html version versus the xml version, we could use e.g. this:

	HTML 5 text        (text/html)
	HTML 5 application (application/xhtml+xml)/(application/xml)

 1. Goes to the hearth of the issue: MIME. Telling and educating
 2. Help us distinguish between the semantic definition: 'HTML 5'
    and the syntactic definitions: 'HTML 5 text' and 'HTML 5 xml'
 3. 'application' would in itself work as a warning against
    using the XHTML serialisation on the public web.
 4. Avoids all conflics with the XHTML working group

Neither xhtml nor xml part of the name of the XHTML serialisation. For that reason, for the XHTML variant, one could look into one of these names, which would have roughly the same advantages as  'HTML 5 application' - though the link to the MIME type would perhaps be less consistent:

	HTML 5 xml
	HTML 5 xhtml
	HTML 5 xhtml+xml

Note: By using lowercase, we signify 'syntax' and 'MIME type'. 

Of these 3 variants, I find that using 'HTML 5 xml' would be most logical, since 'xml' is the only string that belongs to both the possible MIME types. Also, the subtitle of the XHTML 1.0 spesification say that XHTML 1 is A Reformulation of HTML 4 in XML 1.0. By using 'HTML 5 xml' we are both signifying MIME-type and 'reformulation in XML' simultaneously. It is also the shortest.

If we want even closer parallell to the MIME types, we would perhaps have to use 'HTML 5 xhtml+xml'.
leif halvard silli
Received on Saturday, 29 September 2007 19:19:14 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:49 UTC