Re: Next step (Was: I-D ACTION:draft-klensin-unicode-escapes-00.txt

[John C Klensin <[e-mail]>]

--On Thursday, 25 January, 2007 08:01 +0000 "Clive D.W. Feather"
<[e-mail]> wrote:

> John C Klensin said:
>>> But there is (may be) a rough consensus that every scheme
>>> with explicit delimiters (&#xNNNN; or \u{NNNNN}) is better
>>> than any scheme without them? If so, it would be a progress.
>> Much as I would personally prefer that answer, I haven't seen
>> such a consensus emerge yet.
> 
> Well, let's start by seeing if there is.
> 
> I've seen three people here in favour of it. I've seen you
> saying that one variety (the XML one) is ugly, but I don't
> know if you think that that outweighs the benefits of explicit
> delimiters.
> 
> So: is there anybody here - including John - who thinks that
> the chosen format SHOULD NOT use explicit delimiters?

I personally favor explicit delimiters, I have all along.  My
definition of "ugly" involves two things.  The first may be
unavoidable, the second is not:

(i) Use of a sequence of special characters, none of which may
be completely familiar to those who don't regularly use a fairly
full set of Roman characters and both of which, IIR, are
assigned to "national use" positions in ISO 646 BV, meaning that
they may show up in even more different ways in some
presentations.

(ii) If we are going to use explicit delimiters, my aesthetic
and historical sense, and the ability to get help from common
programming-oriented editors and syntax-checkers, favors paired
delimiters over subjectively unmatched ones, e.g.,
   X(nnnn...)
or
   U'nnnn...'
in preference to
   &#nnnn...;

These are, however, just matters of personal taste.

     john