Defining Minimum Criteria for I18N Support

[Rahul Bhalerao]

Hi all,

I think the term 'Language Support' has been used in more vague sense
so far (please correct me if I am wrong).  Also there has been a
difference between a language being supported technically and a
language that is supported with all the localization. Thus there has
been a need to define the terms with more clarity. I think it would be
good if we have two different sections for a language support namely,
i18n and l10n.

Recently I have tried to formulate a minimum criteria for a language
to have "I18N Support". The same is documented here:
https://fedoraproject.org/wiki/I18N/LanguageSupportCriteria

On the same lines we may even have similar criteria for l10n support.
I would like all to do a little review of the above and provide some
feedback so that we could finalize this.

Thanks,

[Ankit Patel]

Rahul Bhalerao wrote:
> Hi all,
>
> I think the term 'Language Support' has been used in more vague sense
> so far (please correct me if I am wrong).  Also there has been a
> difference between a language being supported technically and a
> language that is supported with all the localization. Thus there has
> been a need to define the terms with more clarity. I think it would be
> good if we have two different sections for a language support namely,
> i18n and l10n.
>
> Recently I have tried to formulate a minimum criteria for a language
> to have "I18N Support". The same is documented here:
> https://fedoraproject.org/wiki/I18N/LanguageSupportCriteria
>
> On the same lines we may even have similar criteria for l10n support.
> I would like all to do a little review of the above and provide some
> feedback so that we could finalize this.
>
> Thanks,

Hi Rahul,

Thanks for taking an initiative to define the criteria for language 
support. However, I am not sure whether the term (Language Support) used 
over here is correct or not. Because any package or feature of the 
application included in Fedora have any specific criteria for *support*. 
The criteria defined on the above URL could be called as a *Language 
Inclusion Criteria* in terms of I18n and L10n both instead of *Language 
Support Criteria*. Anyways, FLP team would be happy to define the 
Language Inclusion Criteria in terms of L10n I think.

I could be wrong here, please correct me if I am. Because as far as I 
know, for Fedora, "Support" means voluntary help from community 
contributors through mailing lists, forums or IRC. That's it, I think! 
No specific criteria for it.

So, let's hear the inputs from other members on this thread!

[Rahul Bhalerao]

On Fri, Jul 11, 2008 at 1:10 PM, Ankitkumar Rameshchandra Patel
 wrote:
> Thanks for taking an initiative to define the criteria for language support.
> However, I am not sure whether the term (Language Support) used over here is
> correct or not. Because any package or feature of the application included
> in Fedora have any specific criteria for *support*. The criteria defined on
> the above URL could be called as a *Language Inclusion Criteria* in terms of
> I18n and L10n both instead of *Language Support Criteria*. Anyways, FLP team
> would be happy to define the Language Inclusion Criteria in terms of L10n I
> think.
> I could be wrong here, please correct me if I am. Because as far as I know,
> for Fedora, "Support" means voluntary help from community contributors
> through mailing lists, forums or IRC. That's it, I think! No specific
> criteria for it.

IMHO, 'inclusion' does not sound an appropriate word here. For
example, it would mean that the listed criteria be met even before the
inclusion process starts. Whereas the purpose here is to have a
reference to check after the inclusion process is completed.

The targeted problem here is that, how do we crosscheck if a
particular language is actually a part of fedora. The problem with
l10n/i18n work is that it is scattered through too many packages and
sub-components of the system. This makes it difficult to identify if
the language is usable or not on a given system. This also adds
ambiguity for developers to make the language go through a _feature's_
process. Since a 'new language' in fedora is a significant achievement
for user experience and amount of work for l10n and i18n teams, it
makes sense to identify it through the features documentation and even
possibly the 'release notes'. Thus, if we propose a new language for a
particular release as a feature we have to ensure that a third party
(anyone beyond i18n/l10n team) has a set of checkpoints to verify the
completeness of the feature. This has been the main reason to document
the minimum criteria for support.

Now, I agree that the term 'support' may have varied interpretations
depending upon the context. For fedora, I think it should mostly be
from community perspective. The same term has been used by gnome as
well. The objective is not-to-discourage particular activity on a
particular language and yet to identify few others that have good
status of overall work and usability. This helps developers to set a
milestone and users to set expectations. The same could be true for
fedora. (We have already used it in past but more vaguely.)

This is not proposed as a criteria for 'commercial' support, but I
think the same can be derived on similar lines if needed.

Thanks,

[Sankarshan (সঙ্কর্ষণ)]

On Fri, Jul 11, 2008 at 11:42 AM, Rahul Bhalerao  wrote:
> Recently I have tried to formulate a minimum criteria for a language
> to have "I18N Support". The same is documented here:
> https://fedoraproject.org/wiki/I18N/LanguageSupportCriteria

If the above URL is a proposal - do please mention as such on the page.

~sankarshan