[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: UTF-8+names

• To: xml-dev@l...
• Subject: Re: UTF-8+names
• From: "Simon St.Laurent" <simonstl@s...>
• Date: Sun, 19 Oct 2003 19:17:38 -0400
• In-reply-to: <20031020.000707.19366645.Tony.Graham@S...>

Tony.Graham@S... (Tony Graham) writes:
>> Unicode itself ran out of room and put in surrogates.  Now it seems
>
>Yes.  I don't doubt that 65,000 characters seemed like enough back in
>1988, or that a (fixed) character size larger than 16 bits would have
>been an even tougher sell back when Unicode was getting established.
>
>> that we've run out of patience and added yet another layer of
>> processing in the middle.
>
>Yet the proposal under discussion doesn't attempt naming either 65,000
>characters or 1,000,000+, so I don't see why surrogates have anything
>to do with it.

It's another level of indirection between characters and bytes.  Lots of
people (who've ever encountered them, anyway) gripe that surrogates
complicate processing - and they're just a dead-simple algorithm.  This
proposal makes the impact of surrogrates on the distance between bytes
and characters look trivial by comparison.

• References:
  • Re: UTF-8+names
    • From: Tony Graham <Tony.Graham@S...>

• Prev by Date: Re: UTF-8+names
• Next by Date: RE: UTF-8+names
• Previous by thread: Re: UTF-8+names
• Next by thread: Re: UTF-8+names
• Index(es):
  • Date
  • Thread