On Tue, Nov 3, 2009 at 6:31 PM, Kenneth Gonsalves <[email protected]> wrote: > hi, > I was using the translation memory of poedit to pull in all the Tamil > translations available in my system for automatic translation. The results are > horrifying. It is not just a problem of using high flown words - most of the > translations are completely wrong. People have been translating without the > faintest understanding of the english meaning. Simple things like translating > adjectives as nouns or verbs as nouns and in many cases total nonsense. The > goal seems to be 'how many strings can I do' to get karma or something like > that. I now understand the magnitude of the task of doing kde - over 1 lakh > strings, not only finishing it, but also rectifying most of them. It is > certainly no joke, and we must activate groups who are willing to do this. > Translation is not easy - it is hard and painstaking work, but very rewarding > when the right words are found. And we need a lot of people to do this.
I can understand how you feel looking at the quality of translations. In 2004, KDE translation was taken up by students on mass scale - 20 to 30 students working in a Dishnet centre. While a lot of efforts was made to ensure quality and uniformity of translation, they were not successful. During that time Tamil translation status crossed 95% mark in KDE. That was a good starting point, which could not be taken forward at that time. Options to complete tamil translation of major FOSS applications (KDE, Open Office, Gnome, FireFox, Ubuntu/Fedora/OpenSuSe): 1. A dedicated team of translators working full time This will ensure quality and uniformity of translations. Need financial / institutional support. 2. Spread translation efforts among student community It will be the true FOSS way Cost is less Quality is take what you get. 3. Combination of both A core team of language experts defining terminology, managing commits and ensuring quality Volunteer army to do the volume work. In my opinion, a Tamil Kanini project which aims to take a Linux PC to schools / homes / shops would be one way to achieve the above. The other could be government funded effort (NRC-FOSS?) anbudan, மா சிவகுமார் எல்லோரும் எல்லாமும் பெற வேண்டும் http://masivakumar.blogspot.com _______________________________________________ To unsubscribe, email [email protected] with "unsubscribe <password> <address>" in the subject or body of the message. http://www.ae.iitm.ac.in/mailman/listinfo/ilugc
