Hi Will

I didn't want to say 'because it was always like it is now'. 
But Java has some well-defined patterns, also regarding the i18n stuff,
which should /probably/ apply (ie. how you want to handle localized texts
from properties, totally different again?).

Anyway, for me maybe it is most natural to resolve language specific
values, as it follows a general pattern which is globally accepted.
To switch the default-language, the localization just has to be removed
from the base-name. 
The 'Convention over configuration'-pattern maybe also plays into this
problem.

I don't see many cases for changing the fallback language quite often,
yet. And I don't believe it would be a big challenge to achieve as it is
now.


Just my opinion I couldn't keep to myself, thought.
I was working in i18n-issues some big-times in the past years.


Kind regards
Marco

-- 
15 Years of Excellence - http://netcetera.com

Marco Glur | [email protected]
T +41 44 247 79 20
Netcetera AG | 8040 Zurich | Switzerland | http://netcetera.com

>-----Original Message-----
>From: [email protected] [mailto:user-list-
>[email protected]] On Behalf Of Will Scheidegger
>Sent: Montag, 3. Oktober 2011 13:41
>To: Magnolia User-List
>Subject: Re: AW: [magnolia-user] Troubles with i18n node wrapper
>
>
>Hi Marco
>
>Sorry, but I'm not convinced by your arguments
>
>- "It should be done that way because it always has" is not very helpful
when
>there is a problem.
>- Renaming attributes twice... hmm... I don't know about that. Sounds
like a
>big hassle to me. And I'm quite sure that some attributes would be
missed.
>- And I've never encountered the situation where multi-language was
>_removed_ from a site.
>
>The only argument that _does_ make sense is that this allows he
developers
>to add a second language later and things would still work. But that
would also
>be the case when you would check for the [attributename]_[languagename]
>nodedata first and only fall back to the [attributename] nodedata when
the
>other one does not exist.
>
>-will
>
>On 03.10.2011, at 13:17, Marco Glur wrote:
>
>>
>> In my opinion it should be transparent for i18n support to work
>> according the well-known java message-resources pattern (no suffix in
>> name=fallback) and most stable, of course, respecting existing code.
>>
>> Maybe valid migration path could be to
>> a) rename all existing 'base' values that are i18n with the suffix
>> according to the old fallback language
>> b) then rename all suffixed from the new fallback language to have no
>> suffix
>>
>> The default behavior is anyways not to have Multilanguage, so no need
>> to reference a certain language in the node-data names with a suffix.
>> The decision which is the fallback/default locale should instead be
>> taken at the most early point of setting up a new CMS.
>>
>> Another obstacle having a default-suffixing always would be:
>> How should it work still, when once you remove Multilanguage support
>> for a site completely, need to migrate the node-datas manually, then?
>>
>> I wouldn't see any change of concept here in future.
>>
>> Kind regards
>>
>> Marco
>>
>> --
>> 15 Years of Excellence - http://netcetera.com
>>
>> Marco Glur | [email protected]
>> T +41 44 247 79 20
>> Netcetera AG | 8040 Zurich | Switzerland | http://netcetera.com
>>
>>
>>
>> ----------------------------------------------------------------
>> For list details see
>> http://www.magnolia-cms.com/community/mailing-lists.html
>> To unsubscribe, E-mail to: <[email protected]>
>> ----------------------------------------------------------------
>>
>
>
>
>----------------------------------------------------------------
>For list details see
>http://www.magnolia-cms.com/community/mailing-lists.html
>To unsubscribe, E-mail to: <[email protected]>
>----------------------------------------------------------------



----------------------------------------------------------------
For list details see
http://www.magnolia-cms.com/community/mailing-lists.html
To unsubscribe, E-mail to: <[email protected]>
----------------------------------------------------------------

Reply via email to