It should be exponential. Bursts are perturbations, which get evened out in the long run, like you said. I fit it, and the exponential model fits quite well.

According to this, Tatoeba may very well reach 1,000,000 sentences on the 1023rd day of its existence. C'est-à-dire, in 273 days, or roughly 9 months.

But that's an even-weighted model. If we give the recent few months more weight (though I don't know any justification for doing so), it might fall down to 5 months or even less...

OK, you inspired me. I'm going to do it in Excel right now, and tell you when the million will come.

The growth should be exponential though. I bet you we double in 4-5 months. One million in 6 months.

You did the calculations?

> Not in English. 'noone' isn't correct in English either.

I distinctly remember my high school English teacher going over this, and saying that "noone" was technically the correct way to do it. And everyone was like "What?!" But yea, no one does it that way, really. She didn't make us do it, either.

The procedure for grammatical tags is pretty simple, IMO:

1) Replace all "(Language)" with "". (this one was my fault, as I didn't know you could sort tags by language at one point)

2) Decompose all the "[# Person] [formal/informal/plural/singular]" into two tags (one for the person, and one for the degree of formality, etc). I think that I'm prolly responsible for about 90% of the grammatical tags...

> I haven't tried making "accidental" tags, but "proverbb" doesn't show up when entering "prov...". Is it possible that this is an issue with your browser, rather than the script.

That's because this was a hypothetical example. Try typing in 2nd, and see if you get "2nd personn" as the first choice.

But yes, I do also have an issue with my browser where it often won't let me scroll down with the arrow keys. That would be an individual problem.

I think the bug with the erroneous tags should probably be looked into, as well. For example, if I misspell "proverb" as "proverbb", and then delete the tag and enter the correct one, the autocomplete remembers the incorrect (even though no sentences have that tag). So next time if I want to tag, I'll get "proverbb" as the first choice. I guess I'll try using escape more often.

Sorry, sysko, I know you put a lot of work into it.

> Firefox has a good auto-complete feature anyway, so it doesn't do much for me.

You can get them to work together? Because my Firefox autocomplete is gone now... I kind of want it back.

In fact, I often find myself racing against the system, and typing the tag I want as quickly as I can so that the auto-complete doesn't have time to activate...

Maybe I'm the only one who finds the auto-completion feature on the tag more of a burden than a benefit... but could it be possible to implement a "turn-off autocomplete" option?

It often either suggests to me tags that were entered erroneously and then deleted, or a whole list of tags from which the one I want is somewhere in the middle. The problem is that hitting the return key doesn't implement the tag I've typed, but rather selects the first tag on the list.

Really, really annoying...

сборник = collection (as in, a short story collection)

Yea, you're right. Those would be under @change.

That being said, I think @Needs Native Check and Needs Native Check should just be merged. But yes, I don't agree that they should be associated with @check.

@check seems to be very general and can even include things like wrong language flags and punctuation mistakes. Native check is more specific and thus more efficient. A native speaker simply looks at all the tagged sentences in his/her language and can immediately say whether it's fine or not (without having to worry about all the clutter from @check).

Neden?

Just don't tell Big Brother, sysko...

Il existe une façon... (mais je me tais)

In any case, Uighur and Turkish are only about 50% similar... That could lead to some interesting communication. Maybe later.

Yes, I saw the comment and replied (right?)

You know I *hate* Latinized Uighur...

Indeed, there are very many apple-related sentences...