On 04/25/2001 02:18:40 AM Michael Kaplan wrote:
>I think the biggest problem with such a system is that it encourages
people
>to use a PUA pseudo-encoding rather than do the work to encode a new
script
>if it needs to be encoded. It also encourages private use to become more
>like semi-private use, and this is the kind of thing that should be rather
>vehemently discouraged.
You're entirely right in saying that we don't want people to put off
standardising what should be standardised. There is a valid issue that
should be recognised, though: There are valid uses for the PUA.
[At this point in your message, I thought we were disagreeing because what
you say here seemed to me very black and white. After reading on, though, I
saw that you and I were in agreement. So, the following was at first
intended to counter what I thought you were saying, but now I realise is
consistent with your intent. Having written it, though, I'm not going to
bother revising it - providing this explanation should be adequate and is a
whole lot easier. Hopefully these comments still contribute to the
discussion.]
People need to encode document with not-yet-standardised characters while
those are still in the pipeline. People need to encode documents with
characters that will not be ready for standardisation for some time yet
(e.g. people working with certain types of hieroglyphics may still be
working on their repertoire and may need several more years to do this).
People need to encode documents with characters that may never become valid
candidates for standardisation (e.g. invented scripts, presentation forms
that are needed for use in applications without adequate rendering support,
etc.). These uses of the PUA are absolutely valid.
William is absolutely correct in saying that a receiver needs some way to
know what a sender's PUA data means, and needs a way for his software to
know how to process it as the author would intend. I definitely don't like
his particular suggestion regarding the use of "guidance codes" because it
doesn't solve the problem (unless it is made somehow official, which is an
impossibility and to be rejected outright). But it is entirely appropriate
for communities of users with common interests to have conventions for
dealing with these things. You might consider such "semi=private use"
objectionable, but I must disagree with you on this.
This was done, for example, among the Math community while their characters
were in the pipeline. There was absolutely nothing wrong with that at all.
In fact, it probably helped them to make progress toward getting the
characters they needed standardised since it gave them a way to communicate
with one another and to prototype implementations to see if things worked.
It is appropriate and a good idea for people working on Sumerian Cuneiform.
It's appropriate for linguists wanting to share archives of language data
where not-yet-standardised characters are used.
The only real situation where widespread agreement on the meaning of PUA
codepoints becomes a problem is when those semantics become assumed in
software (unless in special cases where that software is specifically
designed for only those semantics to be used by a market that all agrees on
those semantics), or in any context that can be construed as an industry or
international standard.
>There are enough heavyweights here in the
>"vegetarian" corner that it is worth wondering whether one is really on
the
>side of angels when one chooses to go so far down such a path as to have
an
>official "registry".
The key here is the word "official". Anything that suggests an *official*
registry is definitely bad. I'm with you 100% on that. But for a community
with common interests to say "we agree within this context to assume
semantics that are documented at ..." is perfectly fine. [This is the point
where I realised we were actually in agreement.]
- Peter
---------------------------------------------------------------------------
Peter Constable
Non-Roman Script Initiative, SIL International
7500 W. Camp Wisdom Rd., Dallas, TX 75236, USA
Tel: +1 972 708 7485
E-mail: <peter_constable@sil.org>
This archive was generated by hypermail 2.1.2 : Fri Jul 06 2001 - 00:17:16 EDT