here's some of the comments from the reddit thread i found that on : "It wasn't a problem when it wasn't affecting illustrators" and "It wasn't a problem until it became widely commercialized."Yes dude, its billions being dumped into translation software, how did I not see this...
here's some of the comments from the reddit thread i found that on : "It wasn't a problem when it wasn't affecting illustrators" and "It wasn't a problem until it became widely commercialized."
I mean, you keep going on about axes to grind, when really you are the one keeping this going. I already know my position is correct,
I've already figured out that the OP was going for "evil corpos/capitalism is bad"Well, not really. You keep mentionning theft when the problem lie with potential copyright infringement (which is incorrect, or at least a heavy dose of hyperbole: theft and IP violations are two different things in most jurisdictions), and despite the fact that there are options for scraping the Internet legally/acquiring data as Adobe did for Firefly (among others). There have been pharmaceutical companies that released hastily tested medecine making people sick, but we don't say that medecine kills people in general, and if I were to do that, I think someone might say I have an ax to grind with the pharmaceutical industry.
"Fuzzy logic" has been a part of professional translation for decades, primarily for maximizing reuse of pre-existing glossaries (ie translations) and making it easier to show near matches. While I think AI will further the ubiquity of Google Translate, if accuracy and context really matters, a translator is going to need to see the text. AI will continue to improve and figure out the egregious mistakes better and better but if quality matters, you'll still have translators involved.Before it was widely commercialized, it was an ace in the sleeve for translator who would use automated translation software to improve their productivity. They don't have a problem with the tech, they have a problem with anyone having access to it.
That’s one of the ways we weeded out bad translators when I worked for a translation firm. We’d poke at Google translate until we found a dozen or more words that it translated badly in the target language then make a point of including those words in a translation test. Anyone who returned a bad translation of those words was turned down because either they didn’t know any better or they were using Google translate. Still works, too. Just takes a little bit more work."Fuzzy logic" has been a part of professional translation for decades, primarily for maximizing reuse of pre-existing glossaries (ie translations) and making it easier to show near matches. While I think AI will further the ubiquity of Google Translate, if accuracy and context really matters, a translator is going to need to see the text. AI will continue to improve and figure out the egregious mistakes better and better but if quality matters, you'll still have translators involved.
That’s one of the ways we weeded out bad translators when I worked for a translation firm. We’d poke at Google translate until we found a dozen or more words that it translated badly in the target language then make a point of including those words in a translation test. Anyone who returned a bad translation of those words was turned down because either they didn’t know any better or they were using Google translate. Still works, too. Just takes a little bit more work.
Yep. That’s exactly what we did. We wrote the sample to be translated with an emphasis on the words that Google translate screwed up on and had good translators we knew check the sample translations submitted by potential new translators.That can be helpful, but the best thing of course is to get a sample translation (or hand them a paragraph or two) that's reviewed by a trusted translator.
Well, we were the firm in this case. It was one of the things we did to filter out bad translators we would use for contracted projects.But that also doesn't always work if you hire a translation firm because they might throw a couple of people with mixed skills at a project.