I think an international auto-blog would be a very interesting side-project. It would certainly be a useful and noteworthy thing, even with some manual tweaking of output. The idea is inherently powerful, even if the the implementation would probably be less so.

But in the short term we'll get more mileage from localising ET for non-English use, perhaps with a simple link to a 'translate this page' feature on something like Babelfish. It would be a simple and quick mash-up which would do the job almost as well as a three year research project would.

I wouldn't see developing new language community spin-offs as a bad thing, as long as there was some cross fertilisation between them.

If ET became multilingual overnight, we'd soon be swamped by content. We might be better off having roving front pagers who can cross-post specific topics.

Possibly the easiest way to implement this would be to run everything on a single server in a single database, with language tags for user filtering, and either individual cross-post tags for specific language crossovers, and a global show-all tag for ultra-special posts that would interest everyone.

With a bit of code you could possibly persuade the mash-up to show the translations automatically, with an option to have volunteer translators who can spend 15-20 minutes beating a machine translation into something that makes local sense.

PerCLupi:

I think that the language problem is not what we have to resolve first.

However it is the problem which might best attract EU funding.  I don't see a problem with different sub communities being created who converse primarily in their mother tongues provided there are at least 90% accurate translation engines around to make those conversations at least partly accessible to others, and particularly to people who have some, but not sufficient capability in a particular language without some translation support.

This will help create more dialogue between otherwise diverse linguistic sub-groups who at the moment have no means of interacting meaningfully on-line.   Just becuase we don't have a 100% solution doesn't mean that an 80-90% solution shouldn't be attempted, particularly if bridging the remaining 10% communication gap requires human effort and creates better linguisitic skills, social interaction, and sense of shared effort.

What we are proposing is to create an infrastructure capable of supporting enhanced dialogue between peopl with different primary languages.  The infrastructure itself will never be a substitute for people making the mutual effort to understand each other better.

"It's a mystery to me - the game commences, For the usual fee - plus expenses, Confidential information - it's in my diary..."

I agree with you: the main goal should be to maintain and develop ET as the community and "thinking network" it is. However, we could, in parallel with ET and without modifying ET, launch an experiment about a real-time multi-lingual forum in which the ET users could participate.

"Dieu se rit des hommes qui se plaignent des conséquences alors qu'ils en chérissent les causes" Jacques-Bénigne Bossuet

Yes, I agree. I'm wary of splitting the community if we have several language sites. I'd be keener on having a way to translate more stuff, post more bilingual columns, and have  non-English diaries in the main site.

In the long run, we're all dead. John Maynard Keynes

Clarification:

I have raised just one point as a problem and to harness energies in the process of improving. I have not rejected the possibility of using an existing mechanism that can facilitate that non-English-speaking Europeans consider that it is easier to enter ET.

My view was that an interesting ET attract more visitors than a multilingual ET.

If possible facilitate muntilingual
understanding, very well.

PerClupi, I agree.  Even the small quadriligual exercise I started took much longer than I wanted to spend on it and a full article done in this way would be ridiculously energy-wasting. There might be some way to make it work better but it would take a lot of thought and collaborative time. The original English content to be translated wpuld need to be pared down as much as possible, more drastically than one would normally do for the sake of clear writing. A one-page multilingual translation would need to flow easily from one language to the next, perhaps in the same paragraph rather than distinctly separated.

The instant-translation services I've seen, eg Babelfish and Google, are hilariously bad. Content on my own website was put through one of those word-grinders in Spanish and came out in a tongue no one on this planet would understand.

Blaugustine