IMO. The guys that came up with the variable-length encoding should be tried for treason to humanity :)
I agree with the proposed action, but I don't think the problem is the var length so much as the incredible amount of "appears to work" going on as a result of pretending to be backwards compatible. Things would work much better if they failed more noticeably when done wrong.
For example, UTF-16 is a variable width encoding, yet when someone fails to handle it, they don't move on until it's fixed. ("How do I get rid of the space...")