Translation context needed with examples

[Mioranduh]

I have noticed while translating that the provided context information that is provided is very limited. Sometimes it is important in order to translate propperly, to see an example of what it looks like on the site.

Some translations are only for specific situations and are hard to reproduce if you want to check them manually.

Would it be possible to be more specific, with pictures or hyperlinks to the pages that show the original text in the context? Or, alternatively, be able to show an example of what your translation is going to look like. (Similar to the forum page, where you have the Example button)

[fruktansvärt]

Definitely!

It can be really difficult sometimes to tease out the intended meaning of a string. Especially when it's a short string. Sometimes you don't even know if it's a noun or a verb.

For longer strings, a whole sentence or more, it's often less of a problem.

[Een]

This is a case of something that may seem simple but is in practice really difficult.

Strings to translate come from the website code, which is plain text telling the server and your browser what to do. There is no image or hyperlink there, and very little usable reference to the structure itself (think about getting a picture or GPS coordinates from an unnamed place described in a novel). So only the current minimal context information can be provided based on the file structure.

The recommendation is to use the most straightforward translation, then while using the website, if you notice something wrong, to look up the string and fix it.

[fruktansvärt]

I get that, really.

Right now the "Context" written under the string is often less than helpful. Is there a way developers could use that to give more information? Or an extra Information field perhaps.

For example, the string "Draw" (context: Stw_animations). Is it a verb (Draw a card) or a noun (It's a draw)? These have very different translations. An information field would help with these short ambiguous strings.

[Een]

Right now the "Context" written under the string is often less than helpful. Is there a way developers could use that to give more information? Or an extra Information field perhaps.

Everything is potentially possible. Just more or less super difficult
We have to balance the effort needed with the time available, and the effective inconvenience to the process; if some strings are left untranslated until a translator has encountered them in context while playing, it should usually not be a huge issue.

For example, the string "Draw" (context: Stw_animations). Is it a verb (Draw a card) or a noun (It's a draw)? These have very different translations. An information field would help with these short ambiguous strings.

Can't find that string as you didn't provide the game name or a link.
Statistically speaking, it's probably a verb, but the best way to know would be to play the game

[fruktansvärt]

It was just an example, one I could find after a quick search.

(It's from 7 Wonders Architects, and the correct answer: it's a noun.)

[Spiegel428]

We have to balance the effort needed with the time available, and the effective inconvenience to the process; if some strings are left untranslated until a translator has encountered them in context while playing, it should usually not be a huge issue.

I get where you're coming from, but at the same time it's disheartening to think that the task one's doing doesn't really need to get done, you know?

I've been wondering lately if something could be done to improve communication between translators. In theory, people can use the forum to discuss issues, but not everyone reads the forum. Similarly, in theory people could message the developer about context, but that involves extra steps to find them, and others won't have the information. Etc.

Would it be possitble to have something like an automatic translator group for a game/language, once someone translates a string? Or some sort of message wall on the translation's game page?

[Een]

We have to balance the effort needed with the time available, and the effective inconvenience to the process; if some strings are left untranslated until a translator has encountered them in context while playing, it should usually not be a huge issue.

I get where you're coming from, but at the same time it's disheartening to think that the task one's doing doesn't really need to get done, you know?

Well, I acknowledge your feelings, but I just explained in this thread that I don't have a practical solution. What should I do, a public seppuku to atone for my failure?

Note that:
- as made clear by the example from fruktansvärt, it's often very short elements that are unclear, so giving them the most straightforward translation is not a huge task;
- as I have expressed in the snippet you quote, leaving them untranslated (so doing no task) until you have played the game and seen them in context (left in English, they should be easier to spot) might be the most practical solution.

So maybe the best would be to update our recommendation to:

Our recommendation is to leave such context-dependent strings untranslated. Then while using the website they'll be easy to spot in context since they are left in English, and at that moment you can look them up to translate them appropriately.

I've been wondering lately if something could be done to improve communication between translators. In theory, people can use the forum to discuss issues, but not everyone reads the forum. Similarly, in theory people could message the developer about context, but that involves extra steps to find them, and others won't have the information. Etc.

Would it be possible to have something like an automatic translator group for a game/language, once someone translates a string? Or some sort of message wall on the translation's game page?

Anything will require some effort from translators at this level.

There is already a way to communicate by leaving a comment about your translation to explain your translation choices when it's not completely obvious. It is very close to a translation wall (focused on action rather than just discussion).

The forums by language have the advantage to allow for interested translators to subscribe / unsubscribe from forum notifications.

While possible, creating groups automatically as you suggest would create a lot of groups (42 languages, 535 games => 22470 groups already) and not necessarily improve the situation (very few requests about more communication between translators along the years + effort still needed to go to the group and post + effort still needed to follow up on the group activity + no way to unsubscribe from group notifications at the moment, so the only way would be to quit the group, but you'd get re-added at each new translation...)

So it's not a development I'd dive into with confidence about a clear positive impact.

[cigma]

If you haven't played the game on BGA yet (which is very common if the game is new here), there are some ways which can help to understand the context:
- Search the expression in the english rulebook. Compare the passage with the rulebook in your language, if available (often on https://boardgamegeek.com/).
- Watch some game videos to get an impression of the gameplay.
- Watch some games in play or replays here on BGA
- Play the game yourself on BGA
- If translating a tutorial: Go through the tutorial step by step in another tab while translating. Verify the translation the next day, after it was published.

At least that's how I do it. Even though this won't help in every situation, I hope you'll find something useful for you in this list

[Spiegel428]

Trying not to mess up the quotes, wish me luck.

Well, I acknowledge your feelings, but I just explained in this thread that I don't have a practical solution. What should I do, a public seppuku to atone for my failure?

Surely that's too lenient for your crimes! No, I was just hoping to brainstorm something. And sometimes knowing what options are right out and why helps to come up with an alternative.

[snip] - as made clear by the example from fruktansvärt, it's often very short elements that are unclear, so giving them the most straightforward translation is not a huge task; [snip]

Sure, however the reason I brought up better communication is that I think there are a variety of cases were the translation could benefit from a discussion, because there isn't necessarily a one-to-one correspondence to every word in two given languages, even if the context is clear. "The" most straightforward is a bit of wishful of thinking.

But I think this goes back to the (understandable) tension between "we'd like to do the best translation" versus "we're fine with an ok translation, really, given the lack of resources".

So maybe the best would be to update our recommendation to:

Our recommendation is to leave such context-dependent strings untranslated. Then while using the website they'll be easy to spot in context since they are left in English, and at that moment you can look them up to translate them appropriately.

For the FAQ, you mean? Looks good to me.

There is already a way to communicate by leaving a comment about your translation to explain your translation choices when it's not completely obvious. It is very close to a translation wall (focused on action rather than just discussion).

I think that fulfills a different need, it's more of a change log, useful when someone already knows something is off. (It's been such a long time since I saw a comment that I had to test it to see how it looks on the page). Also, not everyone is going to realize their choices are not completely obvious.

[snip] a lot of groups (42 languages, 535 games => 22470 groups already)[snip]

Sorry, I meant game OR language.

[snip] very few requests about more communication between translators along the years

Speaking for myself here, but there's been times when I just decided not to translate a game rather than go through the hassle of trying to reach other translators. So maybe there are few requests because it's so clunky that people give up... Part of the hassle for me is that if I see a forum with very little activity I automatically assume nobody will read/reply. Or maybe it's just my language that's not very organized.

+ effort still needed to go to the group and post + effort still needed to follow up on the group activity + no way to unsubscribe from group notifications at the moment, so the only way would be to quit the group, but you'd get re-added at each new translation...)

So it's not a development I'd dive into with confidence about a clear positive impact.

I can see that. I suggested groups because they're something that already exist and automatically adding also exists for playing a game, so less implementation, but at the same time, they're not the most usable thing either. Also, I fully admit that I may be the outlier who doesn't mind scrolling by a lot of stuff on the welcome page , I already have sooo much stuff there. Perhaps if one day the groups are revamped we can revisit this idea.