Mur (6 950 sujets)
Astuces
Avant de poser une question, soyez sûr d'avoir lu la FAQ.
Nous cherchons à maintenir une ambiance amicale pour des discussions civilisées. Veuillez lire nos règles contre les mauvais comportements.
gillux
il y a 17 heures
mraz
il y a 2 jours
mraz
il y a 3 jours
maaster
il y a 3 jours
mraz
il y a 3 jours
morbrorper
il y a 4 jours
maaster
il y a 4 jours
mraz
il y a 4 jours
morbrorper
il y a 4 jours
Nandixer
il y a 4 jours
Hallo Leute :)bin neue hier...was geht?
...oder was geht nicht. :)
Ganz unten links gibt es "Häufig gestellte Fragen". Es kann nicht schaden da mal reinzuschauen.
*** author tags ***
I think we should harmonise all author tags. They should include the full name of the author:
"by Goethe" -> "by Johann Wolfgang von Goethe"
There may be some exceptions where it doesn't make a lot of sense to use the full name, but I think the full name should be the standard.
"by Mark Twain" would be "by Samuel Langhorne Clemens".
Obviously, we would still use his pen name.
I hope we don't need to use the full name of Dom Pedro I (first ruler of the Empire of Brazil). His name was:
Pedro de Alcântara Francisco António João Carlos Xavier de Paula Miguel Rafael Joaquim José Gonzaga Pascoal Cipriano Serafim de Bragança e Bourbon
The tag would be:
[by Pedro de Alcântara Francisco António João Carlos Xavier de Paula Miguel Rafael Joaquim José Gonzaga Pascoal Cipriano Serafim de Bragança e Bourbon]
But I agree with the idea.
I tried adding long names before, but the tag has a character limit.
Of course it would be an advantage to have a separate window for all by-tags, listed in alphabetical order allowing an easy survey of the names already used, in order to avoid parallel forms.
(But even now during the entering of names in the tag line we already can see similar tags listed below.)
I suppose it wont be necessary to use names consisting of more than 3 parts (like "by Wolfgang Amadeus Mozart" for instance). Of course even with this limitation a name may be quite long. In such particulare cases a tag like "by Wolfgang A. Mozart" could be a solution.
Sure, full name is fine.
There are also FROM tags to consider [from La Mort et le Bûcheron].
And another comment - BY and FROM support English language only.
How about using some symbol instead of words?
Comics text bubble? i.e. webdings 125 (cannot insert it here)
> There are also FROM tags to consider [from La Mort et le Bûcheron].
All non-English FROM tags were converted to English in the past. I'm not sure this is a good idea, but probably we should stick to it?..
Hopefully, tags will be translated by the Tatoeba interface in the future.
I'm not sure whether we should use from. It may be enough to indicate the author. Some may say we shouldn't go to much into details, but principally this is a good idea.
Some names risk not being clearly identified as such if they are not prefixed by "by"
I totally agree.
[eng] A few useful queries for Esperanto. These are all pairs (two sentences linked between them and not linked to any other sentence):
epo-srp (407 epo sentences):
http://www.divshare.com/download/16666736-047
epo-rus (609 epo sentences):
http://www.divshare.com/download/16666735-b87
epo-por (6466 epo sentences):
http://www.divshare.com/download/16666734-955
epo-nld (549 epo sentences):
http://www.divshare.com/download/16666733-58d
epo-ita (487 epo sentences):
http://www.divshare.com/download/16666729-5e1
epo-hun (2279 epo sentences):
http://www.divshare.com/download/16666728-df6
epo-fra (1194 epo sentences):
http://www.divshare.com/download/16666727-dde
epo-deu (3154 epo sentences):
http://www.divshare.com/download/16666726-975
why Esperanto?
because Esperanto is a valuable bridge from your language to languages such as Persian, Serbian, Hungarian and Portuguese. Also, the size of the material in Esperanto untranslated to English is 98978 (71.46%), even though the size of the material in Esperanto isolated (nor directly nor indirectly translated) from English is just 33594 (24.25%).
This last list is for those Esperanto sentences to one-click distance to be directly linked to an English sentence. They just need an advanced contributor to check if they match, and press "link". The list is fairly big, though:
http://www.divshare.com/download/16666928-1ce
(note: it might help to have this post translated to Esperanto)
[epo] Jen kelkaj utilaj serĉo-rezultoj pri Esperanto. Ĉiuj ĉi estas paroj (du frazoj interligitaj kaj ne ligitaj al aliaj frazoj) :
epo-srp (407 epo-serbaj frazoj):
http://www.divshare.com/download/16666736-047
epo-rus (609 rusaj):
http://www.divshare.com/download/16666735-b87
epo-por (6466 portugalaj):
http://www.divshare.com/download/16666734-955
epo-nld (549 nederlandaj):
http://www.divshare.com/download/16666733-58d
epo-ita (487 italaj):
http://www.divshare.com/download/16666729-5e1
epo-hun (2279 hungaraj):
http://www.divshare.com/download/16666728-df6
epo-fra (1194 francaj):
http://www.divshare.com/download/16666727-dde
epo-deu (3154 germanaj):
http://www.divshare.com/download/16666726-975
Kial Esperanto?
...ĉar Esperanto estas valora ponto de via lingvo al lingvoj kiel ekzemple la persa, serba, hungara kaj portugala. Ankaŭ ĉar la amplekso de la materialo en Esperanto, kiu estas (ankoraŭ) ne tradukita al la angla estas 98978 (71.46%), eĉ se la amplekso de materialo en Esperanto izolita de la angla (nek rekte nek nerekte tradukita) estas nur 33594 (24.25%).
La lasta listo estas por Esperanto-frazoj en unu-klika distanco de rekta ligo al angla frazo. Ili simple bezonas progresintan kontribuanton, kiu ekzamenos ĉu ambaŭ tradukoj kongruas, kaj premos "ligu", se ili kongruas. Sed la listo estas sufiĉe granda:
http://www.divshare.com/download/16666928-1ce
multan dankon!
Korekto:
*unu-klika distanco* > legu: "unu-klaka distanco"
Of course such lists are very useful; they allow to translate much faster. Thank you, Marcello! Kompreneble tiaj listoj estas tre utilaj; ili permesas traduki multe pli rapide. Mi dankas al vi, Marcello!
[not needed anymore- removed by CK]
I speak Russian and Ukrainian with my parents, Russian with most people around me; I've learnt Belarusian at school and don't use it much.
Is it possible to put a flag icon nearby the language-select drop-down box? I believe it would decrease the number of the "wrong flag" mistakes. My bad, I forget it every time.
I have a question about the information contained in the jpn_indeces file (AKA, the original Tanaka "B" lines). Is it being maintained any more? If so, how can one contribute to it? I've noticed a few errors (for example, the Japanese sentence 82502 is linked to English sentence 0 in this file, while it is properly linked to the English sentence 321215 in the links file, and indeed in the live database on the website). I'm the author of a tool for learning Japanese and I have been using the Tanaka corpus up until now. I'm switching over to the Tatoeba data, but I rely on the B lines to quickly search for vocabulary in the example sentences without having to search for all possible inflections.
On a slightly related topic, it seems to me that maintaining at least a list of dictionary forms for words in each sentence would be extremely useful for tool writers. Searching the database for every possible combination, or automatically making a list of dictionary forms for each word is difficult at best.
I suggest you look around here:
http://tech.groups.yahoo.com/gr...t/message/4578
*** Esperanto Tatoeba mondomapo 2011 ***
Al ĉiuj esperantlingvanoj:
Bonvolu vidi ĉi tie la mondomapon de la tatoeba kontribuaĵoj en esperanto en 2011
https://docs.google.com/spreads...r&pli=1#gid=14
La nombroj estas la nombroj da frazoj en esperanto aldonitaj de la kontribuantoj en ĉiu lando indikita en siaj profiloj.
Bedaŭrinde, el la 207 esperanto kontribuantoj, nur 74 indikis la landoj kie ili vivas en siaj profiloj...
Se vi volas indiki ĝin, bonvolu informi min kaj mi aŭtomate ĝisdatigos la mapon.
Mi bedaŭras ke la landaj etikedoj estu en la franca, sed grafikaj dokumentoj en Googlo ne ankoraŭ apogas esperanton...
In my profile there is a stats counter of sentences owned and another of contributions. The numbers are different. What do they count? thanks.
...and I forgot comments on sentences that also account each for 1 contribution...
Still don't get it.
My "sentences owned" shows about 1200, but I inserted just few new items.
My "contributions" shows about 3700, three times higher than the 1200.
Please clarify.
When you insert a new sentence, from scratch, not translating => sentences + 1,contribution + 1
When you translate a sentence, what you actually do is insert a new sentence (your translation) + insert a translation link => sentences + 1, contributions + 2
When you insert a comment on a sentence => contribution +1
My observations show different behavior of the counters:
Adding a single translation promotes the "owned" counter by 1 and the "contributions" counter by 3 (not by 2).
Sorry, you are right. It's 3, because it's 1 for the sentence and 1 for each link, and a translation is 2 links (1 in each direction).
OK
thanks
When you insert a new sentence, it accounts for 1 contribution.
But if you translate a sentence, it accounts for 2 contributions (1 for the newly inserted sentence + 1 for the translation link).
*** Concerning Tags****
I've merged @Needs Native Check and @NNC into @need native check
I've moved "XXX" and "not for safe search" into list instead of tags.
I believe is "needs", though.
Will you merge @change and change?
oh yes I didn't see that,
I also didn't know about @change and change, so yes i will merge them too.
I've merged "Needs native check" into "@needs native check" (and added the missing 's')
I've also merged "change" into "@change"
Note to corpus maintener:
Before starting to mass delete any tag, please talk about it there before.
I do it myself, so I hope everyone will follow this, for the greater good of transparency here.
Suggestion: Tag deletions should appear in the logs.
not a willigness problem, just never had time to implemented it, so for the moment it's about how to deal without that.
There are some strange tags like "by Tatoeba:al_ex_an_der"; "by Tatoeba:Muiriel" ; "by Tatoeba: sysko" and so on.
I don't know how and why they have been created, an if ou click there you will see that all are linked to sentence 1392027 and that seems to make no sense at all.
I created many of them. I recycle the wall.
There used to be a rule that sentences from Tatoeba users were prefixed by "Tatoeba".
The problem is there is a bug with the character ":" in tags since I don't know when..., so the click doesn't work anymore. I reported this already on this wall.
Super ! Bravo !
Maintenant et afin d'arrêter de gâcher le temps des contributeurs, il reste à définir qui a le droit ce créer des étiquettes, à quelles fins et quelles autorisations cela nécessite...Vaste programme !
J'ai dû mal comprendre le privilège et on dirait que des "gestionnaires du corpus" s'amusent à effacer les étiquettes définies par les contributeurs avancés...C'est très drôle, et ça dénote une grande maturité...En voyant de tels agissements, on comprend tout de suite pourquoi ils sont devenus gestionnaires. C'est évident !
Oui donc maintenant essayons d'éviter les étiquettes ayant pour but de montrer leur absurdité
Pour ce qui est des modalités etc., je vais tâcher d'y réfléchir, là je suis en train de peaufiner mon détecteur de langue.
Cool, thanks. Marcelo is right though, it should be "@needs native check".
perhaps you should also merge "@needs native check" with "@check" and "check" XP
This German sentence
http://tatoeba.org/eng/sentences/show/430786
has to be corrected. But since September 2010 comments remain unanswered. Will any corpus maintainer take care?
for these cases it is good to tag these sentences for "@change". This way a moderator can go through the sentences tagged and evaluate if it's time to make the changes effective by his own.
I do that from time to time (but mostly for Spanish sentences)
Thank you! I'v followed your advice.
Dankon! Mi sekvis vian konsilon.