Сan we make Latin script sort after Cyrillic in the language selection for Russian (and Belarusian, when it is added)?

Currently I used things like ‘язык CycL’ (‘CycL language’) instead of ‘CycL’ to make it sort in the end of the list, but it is not quite a good idea because it creates problems for those who use keyboard shortcuts.

I believe it would, at least it's how it is written in modern loanwords, e.g. Μπρίστολ (Bristol), Μπορντό (Bordeaux). Τατοεβα would mean sth like Tatoeva.

But maybe there should be 2 different spellings: Τατοεμπα for modern Greek and Τατοεβα for ancient. :P

BTW, Татоэба is appropriate only for Russian and Belarusian. In Ukrainian, it would rather be Татоеба.

> Τατοεβα
Maybe Τατοεμπα?

I think tranlsating logo is not too much work, but it would look good.

Can you please plot the graphs for Russian?

IMHO "Don't add word-by-word translations" is easier to understand than "...add translations that sound natural...".

Find a Chechen, Yakut or Kalmyk who objects to marking the Russian language with the current Russian flag first, please.

> I have no problem with the flags.
+1

I have to disagree. Naturalness is not a boolean value — sentence can be more natural or less natural. Some sentences are more natural, some are less.

The idea is to find the most natural way of saying something.

IMHO the original structure of the sentence should never influence the choice of the translation.

Lojban is hard. People will not learn it. I think computers will use normal languages.

The problem with comments is that they are not exportable, so you ca’t use them outside Tatoeba.

Something better is needed. :)

I quite agree. Forcing people to write our numbers would make corpus less representative.

However, generating the readings automatically may turn out to be quite problematic. For example, it Russian it would require to do the syntax analysis to determine the case of the numeral.

BTW, Zifre, do you have any downloadable or online reference materials on Chinese in zhùyīn?

I think it is based on the browser or operating system settings.

Thank you. I’ve changed “perse” to “persan” in Launchpad (it is the site where translation is done), but it won’t appear in the interface until the next inteface update. :(

Excuse my answering you in English, I don’t speak French (but I sometimes understand it), I’ve just made the change you’ve suggested.

Vi ankaŭ laboras kun mi. = You work with me too.

Mi havas aŭtomobilon. = I have a car.

Mi havas multajn amikojn. = I have many friends.

Many of them were collected by the professor Tanaka.

Many of them were just added by different people.

Ah, forgot to mention that a main features that distinguist Tatoeba from other corpora are:
· CC-BY license (you can use the sentences in any way you like, without paying anything, and do with them what you want)
· multilingual nature

But some points are lacking:
· there is no morphology checking (since doing it for all the languages supported is a giant amount of work)
· there is no way to differentiate homonyms
· sometimes sentences are not so representative as full texts, so some statistics may be biased (however, using the long texts would raise copyright issues)

Theoretical point of view:
Tatoeba is a language corpus. It is an essential thing for creating language-processing software, and just any software. For example, Braulio Bezerra created a programe to practice typing that uses the sentences from Tatoeba.

Pragmatic point of view:
· You can use Tatoeba as a big phrasebook.
· Since there are many native speakers around, you can try translating something into a foreign language and ask a native speaker to correct you, to improve your skills. But it’s advisable to leave a comment “Needs native check” in such case.

sorry, I meant "one Tengwar sentence"

Well, now, when we have just one Tengwar font, it’s not a big issue. :D