Godot: Is the language option de_CH really needed?

Created on 7 Oct 2019  路  8Comments  路  Source: godotengine/godot

Hi, I'm Nate.
I'm testing this Engine since two days and I pretty like it.
But I don't understand why there is the Swiss-German language option, because there is no grammar of Swiss-German that means in Switzerland we just write in German...
As result if you select de_CH as language you won't get some translations unlike German (de).

discussion enhancement editor

Most helpful comment

The option exists because one translator added it (see contact info in https://github.com/godotengine/godot/blob/master/editor/translations/de_CH.po).

Translations are at https://hosted.weblate.org/projects/godot-engine/godot/

IMO. it's always a bad idea to start a custom translation for a language variant on such a massive translation catalog. The focus should be on finishing the "standard" language translation, and then this translation can possibly be copied and edited for regional variants which may use different words.

Depending on the actual language and/or the translators though, some disagree with this and start a parallel translation. It's not always clear whether they know it because they have a deep understanding of the differences between language variants and think it's better to treat them as fully separate languages, or if it's because they just felt that their country/region needs representation.

We currently have de_CH which is indeed likely worth removing, as it's only 3% complete, and I don't think the differences with the 100% de warrant this situation.
We also have pt_PT which was started as a copy of pt_BR once both were ~20% complete or so. Now both are 100% complete, but IMO translators would gain time by focusing on a common pt translation before adapting it to their regional variants.
Same with es_AR, whose purpose is not very clear to me.

This is even worth on the massive godot-docs repo: https://hosted.weblate.org/projects/godot-engine/godot-docs/
pt_PT was forked from pt_BR, and did not advance much since. es_MX with 4% vs the 52% complete es seems like a very bad idea.

But for these things we have to defer to the actual speakers of the language to estimate whether branching off (or starting from scratch) is worth the effort. I would personally discourage any fr variant from starting translations.


This is actually an interesting problem. It's be probably nice if de_CH could use de for all the missing translation!

That could likely be implemented, yes. We load all translations in the editor so it would be possible in theory for de_CH to fallback to de strings when there is no translation.

That would be nice to have, though I think in this very case the best solution is likely to remove the always empty de_CH translation, and let Swiss German translators help their German speak improve the quality of the common Hochdeutsch translation.
Then if regional variants warrant branching off, it should likely be done by starting from the 100% complete de translation catalog.

All 8 comments

This is actually an interesting problem. It's be probably nice if de_CH could use de for all the missing translation! I wonder if there is a concept of "language variation". To my knowledge (I live in the north east Switzerland) some words differ between High German and Swiss German, perhaps the entire swiss translation should be emptied and only the different words should be filled in, and for the rest it should default to de.

Id appreciate thoughts on this !

As far as I know, there are just differences between German and Swiss German words in the case of a , (for example : gro脽 - gross).
In that case we could just replace the with ss
Is that right or are there more differences?

The option exists because one translator added it (see contact info in https://github.com/godotengine/godot/blob/master/editor/translations/de_CH.po).

Translations are at https://hosted.weblate.org/projects/godot-engine/godot/

IMO. it's always a bad idea to start a custom translation for a language variant on such a massive translation catalog. The focus should be on finishing the "standard" language translation, and then this translation can possibly be copied and edited for regional variants which may use different words.

Depending on the actual language and/or the translators though, some disagree with this and start a parallel translation. It's not always clear whether they know it because they have a deep understanding of the differences between language variants and think it's better to treat them as fully separate languages, or if it's because they just felt that their country/region needs representation.

We currently have de_CH which is indeed likely worth removing, as it's only 3% complete, and I don't think the differences with the 100% de warrant this situation.
We also have pt_PT which was started as a copy of pt_BR once both were ~20% complete or so. Now both are 100% complete, but IMO translators would gain time by focusing on a common pt translation before adapting it to their regional variants.
Same with es_AR, whose purpose is not very clear to me.

This is even worth on the massive godot-docs repo: https://hosted.weblate.org/projects/godot-engine/godot-docs/
pt_PT was forked from pt_BR, and did not advance much since. es_MX with 4% vs the 52% complete es seems like a very bad idea.

But for these things we have to defer to the actual speakers of the language to estimate whether branching off (or starting from scratch) is worth the effort. I would personally discourage any fr variant from starting translations.


This is actually an interesting problem. It's be probably nice if de_CH could use de for all the missing translation!

That could likely be implemented, yes. We load all translations in the editor so it would be possible in theory for de_CH to fallback to de strings when there is no translation.

That would be nice to have, though I think in this very case the best solution is likely to remove the always empty de_CH translation, and let Swiss German translators help their German speak improve the quality of the common Hochdeutsch translation.
Then if regional variants warrant branching off, it should likely be done by starting from the 100% complete de translation catalog.

That would be nice to have, though I think in this very case the best solution is likely to remove the always empty de_CH translation, and let Swiss German translators help their German speak improve the quality of the common _Hochdeutsch_ translation.
Then if regional variants warrant branching off, it should likely be done by starting from the 100% complete de translation catalog.

@puppetmaster- If I'm not mistaken you're the one who added the de_CH translation. Do you agree with the above proposal?

@akien-mga No problem, I wanted to do this a long time ago myself, unfortunately it didn't work out because of problems with permission.

Any progress on this? I can still see the de_CH translation here: https://hosted.weblate.org/projects/godot-engine/godot/de_CH/

I'll remove it next time I sync translations.

Here's the current version of de_CH.po as backup, if it can be used to retrieve some CH-specific translations to improve a new version forked from de.po (which is 100% complete) - again, after assessing if there's really a use case for a de_CH translation in the first case instead of the pangermanic de.

de_CH.po.txt

I would no longer offer a swiss translation. In the Vivaldi Browser it has one too but it reads badly and is very dialect dependent. Don't think that anyone uses it in an editor.

Was this page helpful?
0 / 5 - 0 ratings

Related issues

mefihl picture mefihl  路  3Comments

EdwardAngeles picture EdwardAngeles  路  3Comments

gonzo191 picture gonzo191  路  3Comments

ndee85 picture ndee85  路  3Comments

n-pigeon picture n-pigeon  路  3Comments