Closed kathy-phet closed 5 months ago
Before: After:
Question, is this preferences dialogue set up in such a way that translators can translate the string for the name of their language? (I just note that many of the language names are not translated (e.g., Ahmharic, Armenian, Bengali, etc.). I'm guessing that that's just because translators haven't yet translated them, no?
Not yet, but ideally that would be set up in the future. Whenever we have a request to change it, we will be able to change it without a MR.
@solaolateju wrote
@jonathanolson for the sort, It appears we are currently using the language code (for example, es/español is close to English ). I suggest we use the English string. Allow for more control and easier navigation for users.
For example,
Iranian Persian (fa/fas) and Afghanistan Persian (fa_DA/prs) would appear in different places using either the code or locale name (endonym). They are currently named Persian and Persian (Dari). They would help teachers/students/translator find all the options available to them. They probably would miss it if they are separated.
Sola - I see what you are saying here, but the sort will look very awkward overall, because the user doesn't have insight into the two letter codes.
@jonathanolson - can you paste an example of what it will look like with this kind of sorting based on the two letter codes or english names.
@kathy-phet I was also referring to and prefer sorting with the English name over the two or three letter codes. I believe most native speakers will know what their language is called in English. This way the list is sorted in English and displays the local names. It would end up looking like what we have on the translated sims page but a different layout.
I agree with your position on two-letter codes, users are not familiar with the two-letter codes and would find it confusing. If you consider we have custom two-letter codes as well, a lot more names will be out of place. This applies to three-letter codes.
I see what you mean now looking at Albania/shqipe
. The sort would look off. Let me think about other options.
I found 2 alternative in a quick search
@solaolateju -
sorting similar script together, with, for instance, Latin based scripts coming first, followed by other scripts (Cyrillic, Greek, Han, Hangul, and so on." This is similar to what Jonathan tried doing in the last shared sort.
Yes, I believe @jonathanolson used a sorting standard to lead towards the current behavior. JO can you comment here on the resource used to set the sorting order.
@solaolateju - Does this approach seem OK to you after talking through the options?
We sort based on the browser's unicode-based sorting for English, which will also try to correctly sort text content from other locales. More specifically, it uses https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/localeCompare.
Current (sorted by unicode - so alphabetizes for English and then sorts for other types of characters) :
Sort by Locale code:
Sort by English name:
I'm not sure about showing English names of languages. It seems VERY standard to show the translated language name, e.g. wikipedia:
@solaolateju @RVieyra @DianaTavares - Can you comment on your opinions about above sorting options? We need to finalize this thread and move forward.
I vote for the "Current" solution. It is the easiest to understand visually.
Whatever is decided, I hope this ordering will be applied to the website translation drop-down as well as the simulation filter.
I also like more the "current".
@jonathanolson - I checked with @RVieyra and she said Global team discussed. They are all comfortable with proceeding with the "current solution", so please proceed, and close this when code review and qa testing is complete!
I'm OK with the current as well.
Closing since this is deployed!
Right now the sort on the Localization Tab is very odd. @jonathanolson is looking for a better way with the locale work.