Skip site navigation (1) Skip section navigation (2)

Re: Easy language switching for the documentation website

From: Magnus Hagander <magnus(at)hagander(dot)net>
To: Cédric Villemain <cedric(dot)villemain(dot)debian(at)gmail(dot)com>
Cc: Dave Page <dpage(at)pgadmin(dot)org>, damien clochard <damien(at)dalibo(dot)info>, PostgreSQL WWW List <pgsql-www(at)postgresql(dot)org>
Subject: Re: Easy language switching for the documentation website
Date: 2011-07-09 16:43:26
Message-ID: CABUevEzKry4OZV6bRU-2i4UNrO2w7XyvnZhw3Tzuo6y7Q9MZ0w@mail.gmail.com (view raw or flat)
Thread:
Lists: pgsql-www
On Sat, Jul 9, 2011 at 00:48, Cédric Villemain
<cedric(dot)villemain(dot)debian(at)gmail(dot)com> wrote:
> 2011/7/8 Dave Page <dpage(at)pgadmin(dot)org>:
>> On Fri, Jul 8, 2011 at 3:01 PM, Magnus Hagander <magnus(at)hagander(dot)net> wrote:
>>>
>>> This thing relies upon the URLs being the same, right? So if we want
>>> to do that, why do we need to fiddle with javascript - we could just
>>> put links directly into the templates and have "real links" instead of
>>> javascript, no?
>>
>> Yeah. I also wonder if it's a good idea to add so many "official"
>> looking links to sites that we don't manage. We have spent a lot of
>> time ensuring that our primary site is mirrored across multiple
>> servers and is able to handle error conditions gracefully. I'd rather
>> see the foreign language docs on the same servers, and also with the
>> same stylesheets, banners and menus etc, to avoid confusing the user
>> by sending them to a completely different site.
>
> I fully agree here.
>
> IMHO debian use the right way to do that:
> http://www.debian.org/releases/stable/amd64/pr01.html
>
> Add a .de or a .fr or .. at the end offer you the relevant page. By
> default it is english, and by default it gets your browser language.
> Nothing prevent us to have a list of supported language to help the
> user navigate the documentation.
>
> It should also be easy enough to write on the page when then
> translation seems outdated (from file timestamp between original and
> translated version for example), I didn't check but the issue is not
> limited to PostgreSQL and I am pretty sure some open source projects
> did find a good way to handle that).

Timestamp does not work for determining if it's up to date. Withan
older timestamp it's almost guaranteed to be out of date, but with a
newer timestamp there is nothing saying *what* was changed at the
later time.

I think system like gettext allow a more sophisticated way than just
timestamp comparison, and we need something along that line. But I
don't know the details of how they work.

-- 
 Magnus Hagander
 Me: http://www.hagander.net/
 Work: http://www.redpill-linpro.com/

In response to

pgsql-www by date

Next:From: Thom BrownDate: 2011-07-11 09:47:08
Subject: Re: pgfoundry down
Previous:From: Cédric VillemainDate: 2011-07-08 22:48:36
Subject: Re: Easy language switching for the documentation website

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group