Re: [alreq] Review Proposed Draft: Unicode Technical Report #53: Unicode Arabic Mark Ordering Algorithm

Here's my initial feedback:

Any algorithm that's too script-specific—like the algorithm proposed here, applying to only one script—would only create chaos among implementations and users.

If there's a need to fix normalization for some Unicode sequences which is not possible because of the stability promises, a new algorithm may be introduced to replace the old kind-of-broken one, with all the new character properties needed. This has happened to Unicode Identifiers and Unicode Segmentation algorithms already, and therefore can be done for Unicode Normalization as well—if there's a major issue, as it's been claimed by the draft that there is.

-- 
GitHub Notification of comment by behnam
Please view or discuss this issue at https://github.com/w3c/alreq/issues/143#issuecomment-332263726 using your GitHub account

Received on Tuesday, 26 September 2017 16:53:06 UTC