RE: The perfect solution for the UTF-8/16 discussion

2001-06-26 Thread
NNY OF THE FIXED_LENGTH BYTE $B$i$s$^(B $B!z$8$e$&$$$C$A$c$s!z(B $B!!!_$"$+$M(B $B!; $B08@h(B: Markus Scherer <[EMAIL PROTECTED]>;unicore <[EMAIL PROTECTED]>;unicode <[EMAIL PROTECTED]>; Cc: $BF|;~(B: 01/06/26 20:57 $B7oL>(B: RE: The perfect solut

Re: The perfect solution for the UTF-8/16 discussion

2001-06-26 Thread Juliusz Chroboczek
CB> The solution to ASCII vs. EBCIDC would go away if we got all of CB> the hardware to support Unicode natively. Visions of the MMU performing normalisation on the fly during a DMA transfer from the paper tape reader. Juliusz

RE: The perfect solution for the UTF-8/16 discussion

2001-06-26 Thread Carl W. Brown
Markus, I think that big-endian UTF-32 is the only way to go. The solution to ASCII vs. EBCIDC would go away if we got all of the hardware to support Unicode natively. We could forget about bytes and make the 32bit word the least addressable amount of memory. utf-64 would only be used for vani