1

Is that feasible?? Would the actual editors of those languages be able to make sense out of them?

2

Translation sites suck. I don’t think “create a bunch of incoherent low-quality content in the hope that other people fix it later” is the right thing to do.

3

While it is possible that you can tell what the other person wanted to say, it won’t help much. You may want to make an attempt to pick an article from Spanish wikiHow and use google to translate it. See you can make sense of it:slight_smile:

4

Good point. Just a thought I had.

5

In languages, take Spanish for example, have different dialects and words in certain countries, and variations of the language. So overall I don’t think it would be feasible, it would confuse people and potentially ruin articles. I’ve thought of this though, but translation sites aren’t completely reliable.

6

I tried that with French wikiHow and got told to stop. The mistakes the translation engines make are so big that it’s hard to fix.

7

If I got this right, I don’t think ttrimm meant for the articles to be translated by the tool then saved straight to the alternative language wH site. I mean I think it is a good idea that only needs some … I don’t know, polishing maybe. What I understood from Teresa’s question (and I could be wrong) is that we have something where you can choose an English wH article, some api tries to translate it to the desired language (say using google’s api) and then the article is saved as a draft somewhere in the editor’s personal space. The editor then reads the translation, fixes any inconsistencies and polishes the article. When the article is ready, the editor pushes some “Save” button where the final translated article is saved to the alternative-language-wH. This means that the editors using this tool will have to know both English and the alt-lang. This might be a tool that is available to everyone, but when an article is translated using this way, a “designated” editor/admin or group of editors/admins get a notification. The designated “chiefs of translators” for that language (or whatever you want to call them) can then see whether the article was merely translated using the tool and not edited, and in such case, they have a right to “disapprove” it, immediately deleting it and notifying the editor who did the translation. On the case where the translator did a good job, the article is “patrolled”, “translated”, “approved” or whatever term we might come up with, and kudos is given to the translator. Of course, if such an idea gets approval and sounds appealing, I think I’ll go hiding and advice ttrimm to do the same, as the engineers in wH will surely loveus. With a passion! (That last note was just a joke of course) I think this idea, as I understood it from ttrimm, is one that I would like to see. At least it deserves more investigating and looking into. Thanks for reading. /me goes hiding … lol (jk)

8

The translation sites are far from accurate (I use 'em for my Spanish homework, and that doesn’t turn out well). Translation should be from a person and legit.

9

I thought about something like this as well, it could possible serve the other language wikiHow’s provided of course that the editor, does work after the article has been translated.

10

Google Japanese translation failed. They added unnecessary words in a very basic sentence than what I was taught.

11

You could use them but you’d have to be able to speak the language. Translation sites get the gist of the sentence but it either overcomplicates it or doesn’t add enough.:slight_smile:

12

@Elyne : “The editor must work after the article is translated”. Now that I agree with. That’s why I suggested the article is first saved as a draft and transferred only if the translator saves it to the other language. @JamDonkey : Again, that’s something I agree with. The translator editor must be able to speak and write in both English and the other language. @Maluniu : I don’t know about Japanese, but in Arabic, the translation is about 70% correct which is a good percentage of work reduced. It will make transferring articles (and good quality ones) into other languages a matter of editing the article instead of completely rewriting it in the other language, which, imo, is a step forward:slight_smile:

13

How about, if a non-dual language person worked with someone who could translate. Maybe make translating teams?

14

@Ttrimm That could work (sounds like fun too!).:slight_smile:I’d be happy to help with that!:slight_smile:

15

Yes, I do the same for my French and some really weird stuff have appeared out of there, but…

… I wouldn’t know about Arabic, but English-French is also about 75% accurate. And if you know something about the language you can fix the weird stuff. For French, I send it through Google Translate, and even though I know very little French, I can see most of the mistakes and fix them (I use very simple French =p). And it saves me a lot of typing and time, instead of doing the whole thing on my own with a dictionary. I think this idea is worth looking into, but if translation is done, it should definitely be done by someone who knows both languages. Or maybe even translating teams, like @Ttrimm said.

16

This is an interesting idea. It does seem like it would make translating easier. Are there enough people doing this systematically that building a tool would make sense though? The reason I ask is because it doesn’t seem to take too long to do this manually: copy the text, paste into translator, copy translated text, paste into other language wiki, save draft. I tried it and it took about a minute and a half: http://screencast.com/t/sDETxhMf (and that’s with me figuring out how to copy from Google translate). It would definitely save time if there are dual-language people doing a lot of translating at once, but I guess that’s the question I’m getting at - are there?

17

I’m doing now what you already describing a lot, but it still would be faster if there was such a tool, it’s not that the copy/pasting that takes a lot but fixing the text is. It would be a lot faster - maybe not for one but especially if you do multiple articles. So I would def be interested.