You may have more chance asking this in other more populated forums if you want an answer ^^ We still don't have any native Japanese speakers hanging around here =(

yes for me too :(

I'm re-posting with my "real" account... I used by mistake my "test" account (tranglich) earlier.

So for the explaination, we were updating Tatoeba in order to integrate some bugfixes, but we hadn't really tested everything properly =/

I was thinking, perhaps some of you guys (not only contour and Pharamp, but anyone reading this) would be interested in being our beta testers, or as I prefer to call it "bug hunters".

Send me a PM if you are interested :)

Also (same item) should be "looking for people" not "looking forward people".

ok thanks, that the correction will appear on next launchpad update

Sorry I forgot about that. I modified the extraction script so that it the full path is displayed in "Location" in Launchpad.

The [M] and [F] flags are only required in English sentences used in WWWJDIC. Ideally that information would be stored separately from the main sentence text and added as required later.

The {stuff}{1} like {this}{2} is an old mark-up system that is not currently in use. I would not add it anywhere it isn't already used, and Jim doesn't want it near the sentences he uses with WWWJDIC. ;-)

I actually have no idea if every speech is in the public domain. But if you are only adding one or two sentences from a speech, then you should be fine. If you are adding the whole speech, then do make sure that it is in the public domain.

I would still encourage you to translate what is already here though. It wouldn't be very productive for you to add manually, one by one, sentences from a long text :P
Because yes, we will someday have a feature for mass adding.

As for having a special section for long text, we surely will have that too, someday. But in a much much more distant future than the mass adding. There is still a lot to do with sentences, before we move on to long texts.

One more question (sorry), is there any plans in the future for mass adding of sentences to tatoeba?
[for example importing sentences to tatoeba from a txt file of for e.g. sentences separated by white characters]

In the absolute it's not a problem for me, but maybe then we can put them in a special section, reserved for long text, this way you will not need to worry about the current sentence length limitation and moreover this way people who're looking for long text, will find them easily
what do you think?
and Trang ?

and is there a limit on the number of characters a sentence on tatoeba can contain?

I think Zipangu has developped some mantra for this :P
anyway, congrat to have reached 2,000+ sentences in arabic it's already amazing :)

If it was to be done manually, then one of the French sentences should be :

1) unlinked from its Japanese and English translations
2) link to the other Japanese and English sentences (which would then both have 2 French translations)
3) the Japanese-English pair without French translations has to be deleted.

But this should be done automatically by the duplicate deletion script, when we run it again.

What about allowing users to directly link to audio files hosted on free (or not) hosting sites that allow audio streaming (for easier integration)?

We've discovered Rhinospike quite recently actually. But you know, we were in the phase of releasing all these new things, we didn't have time to consider contacting anyone (because we still need to live our lives and stuff ^^). Anyway, we just checked their license, it appears they are distributing their audio under CC-BY, just like we do! So yes, we will be contacting them and see what happens :)

Now, concerning not recording with your laptop's microphone, the key point was not really that you cannot record with your laptop's mic, but that *qualiy* is important. Usually, your laptop's mic will not give the best quality. However, I do not have any experience in sound, I don't know what is the cheapest way to get the best sound. But this is more something that you'll have to discuss with Shtooka.

Now the problem we're facing is a "Fast, cheap, good. Pick two" type of problem. Tatoeba is cheap for sure. If we pick "fast", then we could have lots of audio, not necessarily optimal for beginners, but always better than nothing. If we pick "good", then the audio we get will be more sustainable, in a way (no need to throw it and replace it), but this will exclude people who will not have the patience and determination to aim for quality (like 99% of people).

Despite of this, I decided to pick the "cheap and good". We are not in a hurry, and actually, we NEED time. The whole audio integration process has barely started. We don't really have a good platform to add audio... Right now, it's quite artisanal. You would have you rename your audio files with the id of the sentence and send them to us via email or something, so that we upload them in our server. Although, the good thing is, Shtooka has a good software that will let you record "massively".

Anyway, besides of that, audio is not really the "core" function of Tatoeba. Not that I don't want it, I've ALWAYS wanted it. But Tatoeba itself still needs a lot of improvements in order to reach a broader audience... not just in terms of features but also in terms of performance (that is, making it faster, and making sure it doesn't crash if we suddenly have thousands of visitor a day).

We decided to introduce audio in order to get awareness and mostly feedback (just like you did), so that we can start thinking about how we can adapt Tatoeba to make the audio integration easier. But things won't really be settled before AT LEAST a couple of months ^^