I think people should be able to configure systems to handle a broad range of text from popular encoding standards like UTF-8.

Limiting text-space because of communcation is a strange objection that I don't think will hold up over time.

Unicode is a garbage standard that breaks apart so easily. That's why people hate ideas like yours. You're right in an ideal world but not in this baseline reality with Unicode.

Except that it is much better than anything that had existed before it.

The earlier handling of non-English alphabets or writing systems was horrible in MS-DOS and Windows.

While there have been made some serious mistakes in the development of Unicode, its main principles were fine and it does not have any competition.

Feel free to propose and implement a better standard.

Just skim other branches of this tree. Unicode is non-canonical in many ways.

You can't guarantee that the same binary representation reproduce on every machines.

That kind of encoding system has no place "under the hood". That should be obvious.