It was actually Gunnar's original suggestion, but I originally objected due to making the majority of programming languages slightly more complicated for XMPP RTT (most of Windows and Mac C++, C#, Java)
But most people in the know, seems to agree -- now I have 3 votes of confidence -- so I'll make the slight change to XMPP RTT to count based on code points. Cheers, Mark Rejhon 2011/6/23 Gunnar Hellström <[email protected]> > Mark said in the UTF-8 / UTF-16 discussion: > >> However, I am thinking of following Simon's excellent suggestion. >> >> What do you think of his suggestion of using "code point" counting for >> length and position attributes? > > [snip] > Yes, counting in code points is the right decision. You do not need to >> comment what that means for the programmer. > > [snip] > Just specify that in the protocol, p and n are counted in Unicode code > points. >
