Changing the autofill for Spanish language in Lokalise

On the Lokalise chat, @cristina.gareta wrote:

Hi Peter, is there any special reason to translate first the root language Spanish instead of the “dialect”? Would it be possible to copy the Spanish - Spain to the root language? (translated words)

It is certainly possible, but I want to make sure this makes sense for the project. As I look at the Spanish language translations now:

• es is 14.3% done
• es_419 (Latin America) is 17.6% done
• es_ES (Spain) is 62.6% done

So there is clearly more in Spanish - Spain right now, but there is also at least some part of Spanish - Latin America that isn’t in the root language.

The way the Lokalise is set now, changes to the root language are automatically copied to the dialects and marked as “Unverified”:

There is a filter in the Lokalise user interface to see which strings are marked as “Unverified” and either verify them or adapt them to the dialect.

In Lokalise’s “Language Autofill” settings, I have options for setting a string as unverified when it is updated (“Turn on to automatically set updated language translations as Unverified”) and to override (“Turn on to override existing translation in target languages”). Given the unequal percentages for the root language and the dialects, I think this makes sense:

1. Make a one-time batch language autofill from Spanish - Spain to Spanish with “Set unverified” unchecked and “Override” checked. This should also trigger the existing language autofill from Spanish to Spanish - Latin America. For Spanish - Latin America, I expect the effect will be that there are a lot of new translations in that dialect that are marked as Unverified, but anything that is in Spanish - Latin America now will remain unchanged (because of the language autofill rule in the screenshot above).
2. Do the same for Spanish - Latin America to Spanish. The effect will be that anything in Spanish - Latin America will be added to Spanish, and if the string isn’t in Spanish - Spain yet, it will be added and marked as “Unverified” (but the existing Spanish - Spain strings will be unchanged).

Going forward, I think it is worth discussing whether it makes sense to have the changes done in the root language first. My assumption was that doing so would make it easier for Spanish - Spain and Spanish - Latin America to reuse each other’s efforts. But I’m not familiar with the Spanish dialects, and so if there is a better workflow for translations we can put different auto-fill rules into place.

Questions

1. Is the 1-2 plan described above a good way to bring the Spanish translations to the same level?
2. How should the Spanish translations use the autofill functionality to the greatest benefit?

The other Spanish dialect translators here on discuss are @cristianchisaba and @jalexsc. There is also Iñigo Lapitz, but they don’t seem to have signed up for an account here on discuss.folio.org. I will send a message to them directly.

Hi Peter, thank you for your prompt answer. Latin and Spain versions of the Spanish language are similar, so the core language may be translated by any of us. However, if we implement FOLIO in a library, the library will require the local version of the language, because some words are different and the language of the interface should be friendly (despite both of them are understandable)

For us (Iñigo and me), the 1-2 plan is ok to reuse efforts done until now. In the future, we will start translating both, the root language from English and the Spain version.

Looking forward to hearing from the rest of the Spanish translators

Thanks, @scanbit. I’m looking forward to hearing from others, too. I should have put a time frame on my first message. Unless I hear objections, I plan to do the 1-2 steps on Friday.

Okay – this work is done in Lokalise and is making its way into the source code repositories now. Thanks for everyone’s input, and please let me know if I can make the process easier.