W3C home > Mailing lists > Public > www-international@w3.org > April to June 2008

Re: Language tag education and negotiation

From: Leif Halvard Silli <lhs@malform.no>
Date: Mon, 28 Apr 2008 03:49:06 +0200
Message-ID: <48152D12.2080506@malform.no>
To: Asmus Freytag <asmusf@ix.netcom.com>
CC: Andrew Cunningham <andrewc@vicnet.net.au>, www-international@w3.org

Asmus Freytag 2008-04-27 05.40:
> On 4/25/2008 11:45 PM, Andrew Cunningham wrote:
>> basically no one size fits all, there needs ot be flexibility.
> For people who are bi- or tri- lingual (in the case of mutually 
> un-intelligible languages) there's another situation that can crop up, 
> which is entirely not handled by the current scheme.

You could say that all Norwegians are bilingual about Nynorsk and Bokmål.

> What one wants to be able to do sometimes is to get not this or that 
> language as a static preference, but the language in which the 
> material was written in originally. That is, if a Spanish site offers 
> an English translation, a bilingual person might well prefer the original

I can tell you that it has happened that I had to check the Bokmål text 
of for instance Norwegian income taxation rules, because the translation 
was unclear. However, you don't bring in the question: How do we know 
that it *is* a translation? You also assume - as with Law - that there 
is always one spesific language version which constitutes original.  You 
must also consider that you undermines the purpose of parallell language 
versions if you tag any of them with "this is a translation". For 
instance, I find the many Norwegian translations of English wikpedia 
pages often worthless. Just knowing that it is a translations, makes me 
not want to read it. (Before, all had to learn to play the piano badly, 
now they must act translators ...)

> It's of course not the case that content is routinely tagged by its 
> status as original/derivative or original/translation. A rough 
> approximation could be achieved by using the domain name. I.e. serve 
> English pages by preference, but for .it, serve Italian pages by 
> preference. Something like that. Or, alternatively, some UI gizmo that 
> sits in a corner somewhere and points out if language negotiated 
> material is available in another primary language, so that a single 
> button would switch (and some history mechanism to record choices).

Assuming that the .no domain contains Norwegian texts, sounds like a 
good default, as long as the Accept-Language message from the UA doesn't 
give other messages - which the server supports and has files for.

> I would casually estimate that the size of the user population, 
> world-wide, for which this kind of solution could be of interest is 
> larger than the population of Norway and Sweden combined. ;-)

Oh, that was a qute smiley.

> I think Leif's issue is quite real, and my point here is that its part 
> of a broader design problem of how to serve users the most useful 
> contents when they aren't in the typical situation of one primary 
> language and one or more quite secondary languages, but where their 
> command of  different languages doesn't neatly fall into this model, 
> including where the languages are only weakly differentiated (Leif's 
> case).
>
> So, if you are starting a campaign to educated every one, if might 
> make sense to address more than one facet of the shortcomings of the 
> current model.

I kind of lost the exact thing you were proposing, other than defining 
default languages for each top level domain.
-- 
leif halvard silli
Received on Monday, 28 April 2008 01:49:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 19:17:17 GMT