Naše data jsou k dispozici pod různými typy licencí Creative Commons.Více informací
Pokud se vám tyto stránky líbí, můžete nás podpořit.
Zeď (7 222 témat)
Tipy
Před položením dotazu se podívejte na často kladené otázky – FAQ.
We aim to maintain a healthy atmosphere for civilized discussions. Please read our rules against bad behavior.
LeviHighway
Před 11 hodinami
AlanF_US
Před 13 hodinami
cafoc64474
Před 18 hodinami
EugeneGS
Před 18 hodinami
cafoc64474
Před 19 hodinami
gillux
Před 3 dny
gillux
Před 3 dny
LeviHighway
Před 4 dny
EugeneGS
Před 5 dny
Ooneykcall
Před 5 dny
Estas mia unua tago en la Tatoeba..
Bonvenon Nagasiva.
Hi people!
I fixed some bugs on the typing game. Basically, it is faster now and the sentences were updated. And the server will be running 24/7, I hope.
http://braulio.home.dyndns.org/tatoeba/race/
Cool! I can get ~55 words per minute in English and Lojban. :-)
I noticed a slight problem though. In Lojban, sentences have to be separated by the word ".i". In your game, they get run together and it's actually really hard to read. :P
Add Ossetian to the list, please! :)
http://tatoeba.org/epo/sentence...ne/indifferent
I guess tomorrow I'll be able to, since Tatoeba generates the CSV download file every saturday: http://tatoeba.org/download_tat...mple_sentences
Fixed!
I guess there are problems with other languages too. There are the right to left ones...
There are problems with the Mandarin Chinese sentences too, in your typing game there's no differenciation between simplified and traditional characters...
Very nice! It's a good idea; it shows what can be done with the corpus.
I would suggest however, that the language selection page could also show the language names beside the flags, so people can choose a language without waiting for all the flags to download. You could also use the IMG tag's ALT property, to show a mouseover tooltip for the images.
Does anyone know what font Tatoeba uses? It's really pretty. I think it's Trebuchet MS but I'm not sure. I'm talking both around the site and for the logo (can't tell if they're the same).
I can confirm that around the site it is Trebuchet MS. I used firebug to check it pretty quickly.
What would be a very cool, an d i guess useful toll would be to make the Tatoeba "guess what tags you might want to add". I know this looks similar to my previous proposals (http://tatoeba.org/deu/wall/sho...7#message_4827 about guessing a users languages), but i think this one is more useful:
For example if there are some sentences containing the word "beach" that are tagged with "holiday" then the next sentenceyou stumble upon containing the word "beach" could suggest adding "holiday" as a tag. It should be greyed out somehow and be next to the tags, so that you would just need to click on it to add it like i also suggested in a previous wallpost http://tatoeba.org/deu/wall/sho...6#message_4636 about using the tags of the translations as a suggestion.
Of course this would still require the users to think about whether a certain tag fits to the sentence, as for example "beach" could also be contained in a sentence about the D-Day landings in the Normandie.
What d o you think about my suggestion? Would it be practically feasable? How much effort would this cost?
It would be not so easy actually, as for the moment we only store the sentence as "one block" so without word by word decomposition, So it would require to develop a segmenter, ok for English it's not so hard (you can make one who give "quite good" result by splitting on space and punctuation) and for Chinese I've already, but for other language it will be more complicated. The language with a lot of flexions (French, German), agglutinativ language (Turkish, Japanese) etc. for . Moreover after it will require to know how to store that in a way it can easily used by the system you want. And even once you have that, without even speaking about context, your sentence will contain a lot of words, adverb, verb, etc. so how do you tell the system that "beach" is relevant but "softly" is not? It requires some kind of "machine learning" algo.
So feasable, hmm yes, nothing impossible in the absolute, though I will not put this feature in the "easy to code" ones ^^
How much effort ? Quite a lot as we have nothing at all in the current system which can be used as a basis (except my Chinese segmenter), but developing tools to analyze the sentences is what I plan to do after the release of the new version of tatoeba, so even though I didn't think about your idea, which is a good one, at that time I will try to see if I really can do it or not :)
I didn't read the whole comment, but flexions should be regruped in the future for a better search feature, as many maniii online dictionaries have. Maybe we ask to use Wiktionary data... or ask directly to proprietary "Fair Use" dictionaries (http://digicoll.library.wisc.ed...2&q1=hall%F3).
> how do you tell the system that "beach" is relevant but "softly" is not?
Well, i was thinking of just a threshold, like say 50 occurrences of the combination "beach"(word) and "holiday"(tag). So as "softly" is not likely to have very often the same tag, it shouldn't get beyond the 50 occurrences limit. Another option would be to subtract all combinations of "beach" and a different tag like "nutrition" (like out of e.g. "We are eating at the beach."), so that "beach" would get below the threshold, because it's too diffusely connected. This would also apply to general words like "is", "have", "do" etc.
Well maybe a "machine learning" algo would be better, but a quite simple subtraction would imho be enough.
well, i understand the difficulty about the segmenter, but maybe it would even work, just not as good, without the segmenter (by just splitting at the blanks). If the suggestion needs a certain "handmade" threshold, it would be quite possible that all grammatical forms would be contained (even though combined words would need the segmenter).
After all its cool to have this on your "feature request list" :)
Ĉiuj esperantaj frazoj estis adoptitaj.
Estis - sed daŭre aldoniĝas novaj orfaj frazoj, ĉar homoj maladoptas ilin :-( Aktuale estas denove 39 orfaj frazoj en Esperanto.
Thanks Pharamp, Nero. I guess I could learn to open my eyes:)
Ahah that's not a problem. We are here to help ;)
Wie kann ich nach deutschen Sätzen suchen zu denen z.B. die Esperato Übersetzung fehlt?
Nero, how will I find this page again, if I lose/forget the link?
Black bar on the top > Browse > Browse by sentence > select your options :)
Sorry, it's "Browse by language" ;)
Or at the main page where it gives the sentence count, click on the number next to the German flag.
Ni bezonas sonoj por la frasoj!!
...
Nu, tio estas jam antauxe menciita problemo, kiu ekz. cxi tie priparoligxis: http://tatoeba.org/eng/wall/sho...8#message_3888
Mi iom samopinias kun vi, sed estus tre malbunte cxi tie sen la flagoj, cxu ne?
Kiel ajn, se vi bezonas helpon pri la sistemo de tatoeba, bv. min demandi, aux eble pli rapide cxi tie skribi.
Kiel mi komprenis, proponantoj de nova lingvo proponas ankaŭ flagon aŭ alian emblemon por ĝi — ne temas pri arbitra decido de la administracio. Teorie povas esti lingvo sen flago aŭ kun ia komunuza emblemo.
Por la venonta Tatoeba tago mi sugestas ke Esperantistoj kontrolu ĉiun frazon kun "po". Por prepari oni legu PMEG: http://bertilow.com/pmeg/gramat...rtetoj/po.html
Kion vi pensas pri tiu frazo:
http://tatoeba.org/eng/sentences/show/530136
Ĉu oni forviŝu la vorton "po"?
jes oni forviŝu ĝin. Mi lasis mesaĝon tie.
Ĉi tie vi povas trovi ĉiujn tiajn frazojn:
http://tatoeba.org/eng/sentence...rom=epo&to=und