[whatwg] Language codes: RFC 3066 versus RFC 4646

2009-04-02 Thread Alex Bishop
In several places in HTML5 (specifically, sections 3.3.3.3, 4.2.5.3 and 
5.11.1), the specification states that certain attribute values must be 
valid RFC 3066 language codes.


However, RFC 3066 was replaced by RFC 4646 (and RFC 4647) in September 2006.

Is there a reason that HTML5 refers to the older standard or is this 
just an oversight?


Alex

--
Alex Bishop
alexbis...@gmail.com


Re: [whatwg] Language codes: RFC 3066 versus RFC 4646

2009-04-02 Thread Masataka Yakura
I think the current plan is to point to BCP 47, since RFC 4646 is
"soon" to be replaced by its successor [1].

[1] http://lists.w3.org/Archives/Public/public-html/2008Jan/0298.html

On Fri, Apr 3, 2009 at 9:30 AM, Alex Bishop  wrote:
> In several places in HTML5 (specifically, sections 3.3.3.3, 4.2.5.3 and
> 5.11.1), the specification states that certain attribute values must be
> valid RFC 3066 language codes.
>
> However, RFC 3066 was replaced by RFC 4646 (and RFC 4647) in September 2006.
>
> Is there a reason that HTML5 refers to the older standard or is this just an
> oversight?

-- 
Masataka Yakura



Re: [whatwg] Language codes: RFC 3066 versus RFC 4646

2009-04-02 Thread Michael(tm) Smith
Masataka Yakura , 2009-04-03 10:03 +0900:

> I think the current plan is to point to BCP 47, since RFC 4646 is
> "soon" to be replaced by its successor [1].
> 
> [1] http://lists.w3.org/Archives/Public/public-html/2008Jan/0298.html

And the general plan is to at some point(s) (e.g., before taking
the W3C version to Last call) do a thorough review and update of
all the references/citations to whatever the current versions are
that they should point to at that time. It's maybe not all that
useful to try to keep them up to date piecemeal along the way.

  --Mike

-- 
Michael(tm) Smith
http://people.w3.org/mike/


Re: [whatwg] Language codes: RFC 3066 versus RFC 4646

2009-04-02 Thread Julian Reschke

Michael(tm) Smith wrote:

Masataka Yakura , 2009-04-03 10:03 +0900:


I think the current plan is to point to BCP 47, since RFC 4646 is
"soon" to be replaced by its successor [1].

[1] http://lists.w3.org/Archives/Public/public-html/2008Jan/0298.html


And the general plan is to at some point(s) (e.g., before taking
the W3C version to Last call) do a thorough review and update of
all the references/citations to whatever the current versions are
that they should point to at that time. It's maybe not all that
useful to try to keep them up to date piecemeal along the way.


But maybe it is. Checking references earlier catches problems earlier.

BR, Julian


Re: [whatwg] Language codes: RFC 3066 versus RFC 4646

2009-04-03 Thread Ian Hickson
On Fri, 3 Apr 2009, Julian Reschke wrote:
> 
> But maybe it is. Checking references earlier catches problems earlier.

It's WHATWG policy to not fill in the references section until a spec is 
basically complete. For HTML5 this is scheduled for August.

-- 
Ian Hickson   U+1047E)\._.,--,'``.fL
http://ln.hixie.ch/   U+263A/,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'


Re: [whatwg] Language codes: RFC 3066 versus RFC 4646

2009-08-10 Thread Ian Hickson
On Fri, 3 Apr 2009, Alex Bishop wrote:
>
> In several places in HTML5 (specifically, sections 3.3.3.3, 4.2.5.3 and 
> 5.11.1), the specification states that certain attribute values must be 
> valid RFC 3066 language codes.
> 
> However, RFC 3066 was replaced by RFC 4646 (and RFC 4647) in September 
> 2006.
> 
> Is there a reason that HTML5 refers to the older standard or is this 
> just an oversight?

On Fri, 3 Apr 2009, Masataka Yakura wrote:
>
> I think the current plan is to point to BCP 47, since RFC 4646 is "soon" 
> to be replaced by its successor [1].
> 
> [1] http://lists.w3.org/Archives/Public/public-html/2008Jan/0298.html

On Fri, 3 Apr 2009, Michael(tm) Smith wrote:
> 
> And the general plan is to at some point(s) (e.g., before taking the W3C 
> version to Last call) do a thorough review and update of all the 
> references/citations to whatever the current versions are that they 
> should point to at that time. It's maybe not all that useful to try to 
> keep them up to date piecemeal along the way.

I'm now going through the references. I've updated the RFC3066 references 
to point to BCP47.

-- 
Ian Hickson   U+1047E)\._.,--,'``.fL
http://ln.hixie.ch/   U+263A/,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'