well, it would be convenient to have such technology, although at the present i think bilingual or multilingual people are efficient enough in translating communications--with the exception perhaps of "literary texts." but that's a whole new argument. in the business, academic, enterntainment, and government sectors, however, i think we're getting along just fine with our human translators.

on the other hand, this ultimate translator may help "empower" sectors of society marginalized by dominant discourse practices. for instance i have observed that the laws of our land (consitution and by-laws, legal documents, courtroom procedures and legal discourse in general) are encoded in what seems like a "hermetically sealed colonial language". and this happens not just in my country. this is the reason why a lot of unfair legal activities go unchecked, and why it is mostly people from the middle to lower classes who depend so much on lawyers (and other legal machinery) or else fall victim to conspiracies of fraud, deception, corruption within the judicial system (and primarily why money becomes a determining factor in litigation).

if the ultimate translator makes the language of legal discourse accessible to these sectors of society, citizens in general would be empowered enough to independently defend themselves with full knowledge of their rights and privileges in accordance to the provisions of the law, and they may also actively participate in the creation of more equitable laws. if this happens then the ultimate translator would expend its ultimate benefit to humankind.

magandang ideya yan! sa pamamagitan ng 'Ultimate Translator', maiintindihan ng mga hindi pinoy o kahit sinong nakakaintindi ng tagalog ang aking komento sa ideyang ito.

(if the 'Ultimate Translator' was already invented, people that do not understand what i just said above can decode and translate it to whatever language they prefer.)

Yes, there are currently automatic translators on the internet, e.g. Google. You just need to front it with an automatic language identifier.
Google - please take note. I can't translate the initial comments because I don't know what language they are in!

I think this will come, but currently automatic translators are c**p.
Like the old joke, somebody ran "Out of sight, out of mind" through an automatic translator and back again, and got "blind idiot".

A wish is not an idea

I Agree, but i think its near to impossible to programm such a super translater :) and if it would happen... please let this genius includes á é(accents!?) and that languages kompatible 2 each other, because in print media design its a war to make some complete different font languages compatible 4 both of them.

I don't see how it would be near impossible to program one. Difficult, perhaps. Definately time consuming. But not impossible.

Someone could create a 'language' of concepts that exists within the computers. To add support for a language, one needs to build two translators. One that translates to the desired language from the internal 'language of concepts' (LoC) and another that translates from the desired language into the LoC.

Translations into this LoC could be logged, indexed, and later searched. The translator can also make use of spare time by translating news feeds and encyclopedia articles into the LoC. These, and other translations, can be indexed for future use. Who knows.. Perhaps this could allow a machine to answer questions instead of just bringing up lists of articles that match keywords.

If you have an idea for making this happen, please post it. But I prefer not to read peoples' complaints; devoid of any attempt to propose a unique idea.

A complaint is a statement that offers only pre-existing knowledge. An idea is an attempt at providing a "unique" perspective potentially contributing to improvement. "Unique" as in having never been thought of before.

Think about your concern, and propose a method of "HOW" to go about addressing that concern. Don't give up, just expand your thoughts further!

There was an article in the New Yorker a month or two ago about how intractable the translating problem is proving to be. Only cut-down, domain-specific packages are helpful. The technology has been dreamed of for fifty years, and lots of effort has been invested, but progress has been limited.

The purpose of this forum is to suggest NEW ideas. This idea is as old as the hills.

the google problem is easy to fix, it feels stupid just saying it its so simple. - write a code to compare some of the words to different languages and choose the language that gets the best match

myparadigm: I've been saying the same thing for years...just compare a few words and you'd have a pretty good guess which language...of course the proper translation is still an issue.

English to a foreign language - fine,Google is doing the job. A foreign language to a different foreignlanguage- a difficult proposition.