clear
{{language.name}} Nessuna lingua trovata.
swap_horiz
{{language.name}} Nessuna lingua trovata.
search
Attenzione: questo sito web è stato creato al fine di tenere d'occhio ogni sviluppo. Ogni cosa che pubblicherai verrà sicuramente persa.

Messaggi in bacheca di Aiji (in totale 147)

Aiji
2019-03-19 14:08
[BUG]

- Add two sentences, A and B.
- Add A to a list L.
- While A is being adding (the waiting symbol is spinning), click on the "Add to a list" button => It looks like B was added to the list L (because L disappears from the dropdown), while the waiting symbol on A spins forever.
- Actually, A is correctly added to L (checked by opening L), and B cannot be added anymore unless going to the sentence page of B.

Expected behavior:
- Wherever I click during the addition of A to L, the dropdown menu of A, and its waiting symbol, should be the ones affected when the operation is over, not the one with the current focus.
Aiji
2019-03-13 14:05
There are 1000 orphan French sentences left because I adopted several thousand of them already. I don't have time recently to check them so the counter does not go down.

The system of writing translations and then unadopt them have several flaws, most of them explained by Alan. But for me, the main flaw is that unpractical sentences are added to the corpus and it becomes very hard to deal with them. Typically sentences that nobody would ever pronounce (or write) but that are correct grammatically (and correct translations). Also, sentences that have plenty of translations although incorrect are a pain :) (but that is due to the French corpus history mainly ^^)
Aiji
2019-03-09 13:41
I've thought a similar way before, but I faced an issue: when checking a French sentence, I can correct its structure, grammar, etc. but I can never be sure that the translation is correct, and that makes me uncomfortable. What are your thoughts about this point?
Aiji
2019-03-03 14:43
As shekitten suggested, and as frustrating as it can be, the problem lies in the tool, not in the source.

Having near-duplicates sentences could be very useful to a tool trying to extract information from sentences, as a training set, for example.

I think it is harder for the maker of the second tool to create near-duplicates out of nothing than for the one of the first to avoid near-duplicates that already exist.

As always, most of use-case issues lie in poorly designed tools and not in the source (Surely, the source could be improved, but you got the point I think).
Aiji
2019-02-24 01:34
Regarding the languages setting, I'm using Firefox on Windows and I have no issue typing there, and the setting is still working (I have let it active for a few days)
Aiji
2019-02-14 13:45
Yes.
If I link one French sentence, then all translations will appear.
The problem will not happen if you restrict the langages to be displayed in your profile. That's usually how I do it, but this time I forgot so I faced this situation (maybe I, or somebody else, already mentioned it in the past).
Aiji
2019-02-13 14:41 - 2019-02-13 14:44
Advanced search:
English, display translations in French
Limit to sentences having French indirect translations.

Only English and French are displayed.

Link one French => All translations are now displayed.

Is it intended behavior?
(I do not think so, and that is quite a bother when doing some linking work)
Aiji
2019-02-10 14:39
Let's hope that in time "Using a bot to split and add whole classical novels without any added value into the corpus." will not rise to "contribution of quality" :)
But of course, everybody is free as long as they follow the rules.
Aiji
2019-02-05 12:45
> And I propose that unless and until we have a quick-response team, we should go on
> record as having a policy of encouraging mass import (preceded by proofreading) and
> discouraging or even disallowing bots other than official Tatoeba ones like Horus.

Alleluia.


My problem is that I tend to go with the last option that you proposed (cleaning the evil to keep the good). While we could simply massively cleanse everything...
Aiji
2019-02-02 03:12
Can the admins explain why bots are authorized now?

I believe that past experiences have shown that it is generally not a good idea, the process being more often harmful to corpora than helpful. That for a simple reason, the creator of such bots being less motivated by a good contribution to the project than by their own interest. I respectful way to use bot-generated sentences would be to generate them on your local platform, proofread them and upload only the correct ones.

It makes me sad that when people invest time and energy to increase the global quality of a corpus, suddenly a random Johnny comes and insert thousands of sentences beating that down, with zero effort, not showing the minimum respect as to proofread them. (Voltaire + Maxence = 11K, although one of them is supposed to be human)

Now I see that Voltaire has been red-marked. I guess that is CK's doing after one of my comment yesterday. That sill does not explain why it was accepted at first :)
Aiji
2019-02-02 02:45
I got the same six sentences with only their Turkish translations when doing the same search. As CK mentioned, could you give us the complete URL of your results page?
Aiji
2019-02-02 02:32
I understand that feeling!
Several users, I included, have their own list of "My sentences", to serve that purpose.
Mine is here for example: https://tatoeba.org/fra/sentences_lists/show/6375
When adding sentences, we need to add them to list, so sometimes I forget to do so, but now original sentences can be distinguished from translations, so let us be patient and wait for a new research option :)
Aiji
2019-02-02 02:27
Then let me apologize for not being precise enough. I was talking only about this specific case.
As another example, I would not unlink "Qu'est-ce que tu fais ?" and "Que fais-tu ?" if they were linked. Personally, I don't find it necessary, but such a link is not harmful and "may" be helpful to somebody.
Aiji
2019-02-01 14:38
I don't think it is possible to do that with the current tool. I think what you want to do is to cross the search for translated sentences to a search of original sentences of the author, right?

Maybe it will possible in the future however.
Aiji
2019-02-01 14:34 - 2019-02-01 14:35
Your reasoning seems biased to me, at least on two points: european-centered, and purpose-centered (Tatoeba is used to learn a language).

>But they contain the same vocabulary, the same grammatical structure, the same word order, and so on.
It is funny how you do not say "the same meaning", although it is my mistake for saying "different sentences" and not "different meaning". However, since we're debating in English, you'll surely excuse me for that mistake.
And now, if you can tell me contexts where "Mange." and "Mange !" are the same, in terms of meaning, and not of "trivial transformation", I'll be happy to hear some of them. But then I guess "Tu vas au restaurant." and "Tu vas au restaurant ?" are also similar sentences that can be linked. And I shall be doomed.

Long story short: don't tell me how to maintain our French corpus, French natives are here to debate on that. :)
For most of the other points, Thanur has explained them probably better that I could have.
Aiji
2019-02-01 14:25 - 2019-02-02 03:06
The request has been black-holed

Error: The requested address '/fra/sentence_comments/save' was not found on this server.

I'm not sure when this happens. I think it was when I posted a comment. The comment was not saved.
EDIT: It happened on the Wall as well, just now :(
Aiji
2019-01-29 14:45 - 2019-01-29 14:46
I'm not sure that I completely understood what sentences you're linking together so I wouldn't say that I disagree but I think I might disagree^^
Languages that possess an exclamation mark, or whatever punctuation sign it may be, possess it for a reason. The fact that people don't know how to use them or overuse them is certainly sad, but « Mange. » and « Mange ! » are two completely different sentences.

Therefore, « linking them as synonyms » is, for me, a mistake. And I actually unlink this kind of pair when I see them. We wouldn't use the one instead of the other. And if we translate these two examples into Japanese, for example, they will give us two quite distinct sentences.

On a more global scale, I am against linking sentences of the same language together, except in the case of expressions/sayings/regionalisms/not dictionary-friendly. Doing so could lead users / learners of the said corpus to unfortunate mistakes, believing two words have the same meaning out of all context.
Aiji
2018-12-12 00:13
That's an interesting discussion, but we can go down forever and ever ^^
We can split "awkward" sentences into correct awkward, academic awkward, regional awkward, etc. With Tatoeba current system, they probably should be tagged, but not everybody thinks about it, or can, and in the absolute tags are here to give meta-information and help the person who would not understand why such a sentence is correct.

A simple example: The French « C'est quelle heure ? » is (clearly) grammatically incorrect. However, this is the natural way to ask the time in some regions... Hence, even "natural-sounding" sentences are up to debate. Etc. :)
Aiji
2018-12-09 06:19
I'm part of those who do not believe in "good" sentences. That is a very bad word, impractical to use.

That being said, following Tatoeba's rules, why would you classify as "good" any of 2, 3, or 4?
In 2, if the punctuation is really erroneous, that is not simply a matter of appreciation but a real mistake in the sentence, it needs to be corrected.
Similarly, 3 and 4 need to be corrected.

I think I see your point, but from the experiment point of view related to what TRANG was talking about, they would weigh as "bad" sentences in the balance to extrapolate to the whole corpus (of course, that is not possible, but I think you see what I mean).

As for proofreading, I agree with you. Proofreading is best done with a fresh set of mind.
Aiji
2018-12-03 14:14
Merci merci