Ray Saintonge wrote:
cookfire wrote:
I see what you mean, and it is true that we should be careful before transferring translations from one Wiktionary to an other. But don't you think less errors will happen if a larger group of people will be able to see and correct one version of translations? Each from their own viewpoint/background and with an interface in the language of their choice? Now all the Wiktionaries are 'parochial' and if one takes something from one to another, then somebody else changes/corrects this, it is probably never going to be changed in the Wiktionary where the content came from in the first place. The same is true when something gets changed in the original Wiktionary after the transfer has happened. This change/correction will never get notified to the second Wiktionary. In an integrated system, they could be marked 'fuzzy' to indicate something has changed, meaning that things have to be checked again. I agree that this has to be given a lot of thought to set up though.
One thing that I have noted about the initial examples for this proposal is that they started in a subject area where the translations are reasonably well behaved, the names of nationalities and languages, even if there are occasional problems such as choosing between Netherlands and Holland. When you try to extrapolate an idea that works well in particular circumstances to a more general situation you can expect some very serious difficulties.
Ec
Wiktionary-l mailing list Wiktionary-l@Wikipedia.org http://mail.wikipedia.org/mailman/listinfo/wiktionary-l
When an English word is entered into a wiktionary all truisms in the en:wiktionary for that word should be true in the nl:wiktionary, the ja:wiktionary as well. We are after all talking about the same word. When the translation for an English word is XYZ in language F, this should still be true when you describe word English word XYZ in any other language.
What is problematic to be expressed about an English word in English is still dificult to be expressed in any other language.The reason is simple: you are talking about the same thing: the relation of a word in a language to an other language. In whatever language you describe it, you describe that specific relation.
The aim of the excercise is to have a fixed format in which these relations are described. This will result in data about a word that will result in data about the same word presented in another language. When this turns out to be wrong, it is also wrong in the first language.
Thanks, Gerard