It looks like a no brainer to me. If you have some program based on a standard, any non-incremental changes to the standard mean spell hard work: you have to read the updated and old spec side by side to just to determine whether something is now broken in your program, never mind what needs to be done. Of course that guy is opposed to that.

Assurances that, don't worry, it's all gonna be backward compatible, are neither here nor there, from that perspective.