This is not an official answer (I have nothing to do with Google), but: As I understand it The Google translator needs three things to "learn" a new language:
1. A large body of text in the target language. Google gets this by indexing the internet. 2. Parallel texts, that is texts of which you have copies in multiple languages, including the target language Pashto in this case. Google gets these from official sources (like the UN, which translates a lot of documents in its work), and as it learns a language better, it may be able to identify parallel texts by itself when it finds them on the internet. 3. Manual work from Google engineers. Exactly how and what they do I don't know, but they have said there is some. The thing your 10000 strong army of Pashtun speakers can contribute to is point 2 ;) If you translate Wikipedia articles to Pashto using the translator toolkit (http://translate.google.com/toolkit/) then they are easily accessible to Google, and will probably have structure which makes them particularly easy to feed into the translator. It may well be that it learns from other types of translated web articles as well, but Wikipedia it seems to have special support for. The toolkit knows about Pashto although it can't translate it automatically yet; you can tell it that's what you're translating to, and Google will be able to use that information. There is little use in contacting Google, they do not have staff on hand to answer the huge number of people who want to tell them something. But we know that they are trying to add as many languages as they can, and I know at least two on the Google translate team look over this forum on occasion. -- You received this message because you are subscribed to the Google Groups "General" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/google-translate-general?hl=en.
