Re: [whatwg] Should navigator.language and and/or HTTP Accept-Language include locale?

On Wed, 25 May 2016 23:56:09 +0100, Nils Dagsson Moskopp  
<nils@dieweltistgarnichtso.net> wrote:

> Anne van Kesteren <annevk@annevk.nl> writes:
>
>> On Wed, May 25, 2016 at 7:32 AM, Михаил Гаврилов
>> <mikhail.v.gavrilov@gmail.com> wrote:
>>> I propose to standardize locale settings (datetime,  number
>>> delimeters) can be specified only by user via the operating system
>>> settings. Sites should not change the locale that the user has chosen
>>> for himself.
>>
>> There is a proposal to do that at
>> https://github.com/whatwg/html/issues/171 but it has not meaningfully
>> progressed thus far.
>>
>> Note also that some sites go out of their way to localize form
>> controls to fit the language of the site. To the extent they will use
>> libraries rather than <input> and friends. Arguably they're wrong, but
>> there are cases where that does make sense. And let's not forget that
>> users often forget to set these settings properly, that's probably
>> part of the reason why sites do that kind of thing.
>
> In my experience, the “users might have forgotten to set this, so we
> default to something that is not what the user asked for and create a
> better user experience” is something that is often invoked by developers
> that do this stuff, but I have never seen data for/against the claim. Do
> you have evidence for/against it improving user experience on web pages?

I don't *have* evidence I can share, but I have *seen* clear evidence that  
language settings for english in particular are seriously likely not to be  
representative of the people who have them.

Which is hearsay, of course. I would be very happy if someone can share  
clear evidence, but I'm prepared to accept the claim otherwise.

cheers

-- 
Charles McCathie Nevile - web standards - CTO Office, Yandex
  chaals@yandex-team.ru - - - Find more at http://yandex.com

Received on Thursday, 26 May 2016 05:55:25 UTC