mdmkolbe writes on Slashdot:
Given that on a system with Unicode, you almost never want to do a non-normalizing case-insensitive match and that it is hard for a user to efficiently implement their own normalizing case-insensitive match, it seems an odd corner of the rectangle to omit.
Cowan writes:
Alternatively we could specify that -ci procedures always normalize, or that -ni procedures are always case-insensitive, since the details of the normalization are not exposed anyway.
The WG voted to reject this proposal and remove the *-ni procedures altogether.