W3C home > Mailing lists > Public > public-html@w3.org > August 2008

Re: meta content-language

From: Leif Halvard Silli <lhs@malform.no>
Date: Thu, 21 Aug 2008 08:45:35 +0200
Message-ID: <48AD0F0F.3030602@malform.no>
To: Ian Hickson <ian@hixie.ch>
CC: public-html@w3.org

Ian Hickson 2008-08-21 03.36:

> The alternative as I see it is to disallow it. The problem there is that 
> people will get error messages in conformance checkers, and those error 
> messages aren't very useful.

Instead,  when HTML or BODY is lacking @lang, let the validator 
inform the author that the Content-Language pragma will be used as 
fallback for <html lang=""> - the same way as in HTML 4. (It is 
likely that those that specify the content-language without also 
specifying <html lang="">, would be /interested/ in specifying 

What is the problem with keeping it as it is in HTML 4?

OK, perhaps decouple it from HTTP as such. And replace the HTML 4 
algorithm, which does not distinguish between HTTP and the meta 
pragma. And please specify that the first language is the one 
which will be used if several are present.

But do not mess with the basic fact that it specifies the language 
  preference(s) of the intended audience. It is - litterally - a 
_meta tag_!

>> Since <meta> can and should be permitted to contain a list of languages, 
>> an algorithm that sets the content language to only the first language 
>> is inappropriate.
> Right now, the HTML5 spec says that only one language is allowed. What 
> should the processing be if multiple languages are provided? It doesn't 
> appear that people use it to provide multiple languages, they only provide 

To say that only one language - the first - will be used for text 
processing if several are speficied, is fine. (HTML 4 does not 
give rules for this.) But to disallow adding several languages, is 

Those that provide multiple languages use it according to the HTML 
4 specification, I guess. That is: they don't use it to "provide" 
multiple languges.

> On Sat, 16 Aug 2008, Leif Halvard Silli wrote:
>> There is currently nothing, I think, which prevent authors from making 
>> the META content-language element appear several times in the same 
>> <head> element. Did the algorithm take this into account?
> Only the first is used. The others are non-conforming.

For "Content-Type" (the Charset/Encoding) I see that it is the 
first meta, if there is more than one, which counts.

But for Content-Language, at least in Webkit, Firefox and Internet 
Explorer 5 fo Mac, it is the last instance which counts. (Though 
Internet Explorer 8 beta 1 looks at the first - it is the only 
one.) That at later rule overrules a previous one, is very common 
to authors.

(That you are willing to change how this works, indicate that it 
currently isn't actually a very important feature. At the very 
best, it has been used as a method for passing CSS to particular 
browsers ...)


> Note that none of this applies to the HTTP header, I'm only talking about 
> the <meta> pragma here.

Is it your intention that the Content-Language header, when coming 
  from HTTP, shall not have any effect on the document with regard 
to language? While the META Content-Languge pragma shall?

(In HTML 4, there is no distinction, which means that the page 
also inherits the language from the HTTP Content-Langauge header.)

> On Fri, 15 Aug 2008, Phillips, Addison wrote:


>> In addition to being incompatible with existing Web content
> Which Web content? Could you show us pages that actually use multiple 
> language tags in the Content-Language <meta> pragma?
>> I really don't see why we need to change the Content-Language meta tag 
>> from indicating the target audience to indicating the processing 
>> language. Since browsers don't make use of this information today for 
>> processing the text, we'd be better to make existing practice formalized 
>> than to change semantics.
> They do make use of it.

Are there any pages which actually relies on the language 
information coming from the META element? For instance, how many 
pages aare there that use the :lang() selector, and which gets the 
mathching language from the content-language pragma?

(I suppose that if some author is actually using the :lang() 
selector for anything, then he/she is also already using the lang 

leif halvard silli
Received on Thursday, 21 August 2008 06:46:23 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:37 UTC