- From: Andrew Cunningham <lang.support@gmail.com>
- Date: Thu, 1 Aug 2013 09:08:17 +1000
- To: Marcos Caceres <w3c@marcosc.com>
- Cc: acunningham@slv.vic.gov.au, www-international@w3.org
- Message-ID: <CAGJ7U-Vi0rKx3X1OcSyEHipg8wM19+o73i6nbC4AVoi=fveY5w@mail.gmail.com>
On 01/08/2013 4:26 AM, "Marcos Caceres" <w3c@marcosc.com> wrote: > > On Saturday, 27 July 2013 at 00:28, Andrew Cunningham wrote: > > > Talking about writing systems becomes more problematic.. > > Since you are talking about 2 to 4 BCP language tags at least > > Language > > Country > > Script > > Variant > > > > I will check the apps that we have in the FxOS marketplace, but I don't think any have ever gone beyond language-country. But yeah, idea is to keep it simple. Understandable > That's good feedback. I guess I'm wondering what the ideal is to cover the 80% use case? Like I said, the content that is motivating Mozilla to add this API rarely, if ever, get's localized beyond two sub tags (that's not to say that it shouldn't, but that's just what we are seeing so far - though the good news is that developers are localizing content, which is great!). I'm going to check all the content that we have got so far just to make sure. > For the 80% mark the approach you've outlined would be sufficient. Although localisation takes infrastructure, skills, knowledge and resources that many languages do not have. My main concern is that what is adopted is a mechanism with a certain degree of future proofing. If built on what is done now, may not be compatible with needs of some languages in future. In our own work we've steered away localisation projects in the languages we work with, because localisation models in use, seem to be missing components needed. Andrew
Received on Wednesday, 31 July 2013 23:08:55 UTC