tolgaaaltas It makes zero sense to change the way we slug things in core at the moment. We will make it very easy for extension developers to adjust and make possible UTF-8 usernames. It only makes sense to allow UTF-8 usernames if the slugging strategy supports it. The built-in strategy only works well with ASCII because of the way browsers are.
Some languages, notably the CJK languages using numbers as a slugging strategy because browsers don't display those characters well, while other languages like french, german, etc. can fairly easily be translated into ASCII for slug purposes. I should note that our built-in slugger is (if I remember correctly) capable of converting french, german, etc. into ASCII which means an extension dev in theory could (in the future) simply change the validation logic for usernames. While other languages (CJK) will need to have more drastic changes to both slugging and validation logic. However if we changed the validation logic in core, without updating the slugging it could lead to huge issues.
Our goal is to support all languages across all countries in the entire world. There will be compromises we have to make, hopefully the community can trust us to make these compromises in good faith.
Something I want to note is that internationalization is incredibly difficult to do well, and just some quick google searching on UTF-8 usernames will result in a huge number of github issues and forum issues noting complications of doing so when the backend/frontend doesn't fully and properly support it. We will make core capable of doing UTF-8, we will trust our community to come up with the best implementation of slugging and username validation that works for them.
