Charlotte đŠ 
charlotte
Whatâs the preferred order of a language dropdown?
Right now i have it sorted by language code which seems reasonable but some language codes for some smaller languages are kinda weird, so the ordering might be unexpected
Unicode codepoint would have issues with languages that are written in multiple scripts (say, serbian) where the two language variants are split apart due to using different scripts.
Romanization has issues with language (families) which share one or two orthographies but have different pronunciations and romanization schemes. As such stuff like arabic or chinese languages would be split apart in the list, which is probably unexpected.
I could of course group languages and sort with the other two methods but then i wouldnât know which is the âcanonicalâ language to sort by
1
2
2
de weledelzeergeleerde Hx Anne C.A. "Joe 'p" Baanen, aggressieve mechanisator en effectieve genderaar
Vierkantor@mastodon.vierkantor.com
@charlotte i want to vote for whichever scheme Wikipedia uses, and that seems to be by ISO code.
0
0
1
oh i just remembered that due to the ravages of mastodon, the 5th option is not visible to mastodon users. If you want to vote for âBy romanization, but with languages groupedâ you could also just like this post
1
0
2
Peter Brett
krans@mastodon.me.uk@charlotte I recommend "by default Unicode collation algorithm" instead of "by Unicode codepoint."
1
0
1
gabbo the wafrn guy 
gabboman@app.wafrn.net
ï»ż@charlotte@akko.chir.rsï»żÂ alphabetically acordingly to what is being shown
#i-hate-so-much-when-'españa'-is-after-'sri-lanka'
0
0
0
âlolâ âfedifriend (cat aspect)
samgai@vocalounge.cafe@charlotte A difficult question.
There's also the option to sort using a language-neutral collation, which puts letters like A Ă Ă Ă together, making it a bit closer to what people expect than just sorting by code points which may put similar letters far apart.
In any case, I think that the representation which you use for sorting should be visible. I've seen a lot of UIs fail to do this.
1
0
1
âlolâ âfedifriend (cat aspect)
samgai@vocalounge.cafe@charlotte Wikipedia has got a nice search function which seems to respond to ISO codes, endonyms, and exonyms in ANY language known to Wikipedia. For example, you can be on English Wikipedia, search for ăă€ăèȘ, and get to German Wikipedia that way. But that may be overly complex for your use case.
0
0
1
Peter Brett
krans@mastodon.me.uk@charlotte That's correct. There's a default, generic #Unicode collation algorithm which is "good enough" in many contexts.
For specific language, region and/or context, the more specific collation is called a "tailored collation".
0
0
1
@charlotte I feel like the answer here is "any option that feels consistent to a lay person", which rules out a lot of options, like code points.
The ones I have always found most usable are the ones that list languages by their English name, and sort it by that as well. I learned the English word for Swedish pretty quickly :)
0
0
0