Re: 64 bit addressing and OOP

Re: 64 bit addressing and OOP

[Richard Riehle]

Mark Thornton wrote:

> Err, but Unicode already requires (I think) 21 bits which is the point of
> this thread.

Thanks, Mark.

I am including this reply in comp.lang.ada

So, will the next Ada standard include support for a 21 bit Unicode,
or am I missing something and it already does?

Richard Riehle

Re: 64 bit addressing and OOP

[Randy Brukardt]

Richard Riehle wrote in message <3E594650.E09A986D@adaworks.com>...
>Mark Thornton wrote:
>
>> Err, but Unicode already requires (I think) 21 bits which is the
point of
>> this thread.
>
>Thanks, Mark.
>
>I am including this reply in comp.lang.ada
>
>So, will the next Ada standard include support for a 21 bit Unicode,
>or am I missing something and it already does?


Probably. See AI-285. (http://www.ada-auth.org/ais.html).

        Randy Brukardt.
        ARG Editor

Re: 64 bit addressing and OOP

[Kent Paul Dolan]

"Randy Brukardt" <randy@rrsoftware.com> wrote:

> Probably. See AI-285. (http://www.ada-auth.org/ais.html).

??? That link doesn't go to anything obviously pertinent, just the top
of some huge and not pellucidly navigable database even whose
component parts are confusingly labeled.

Do you have a direct link or is what you reference really on that page
somewhere?

xanthian.

Re: 64 bit addressing and OOP

[Hyman Rosen]

Kent Paul Dolan wrote:
> "Randy Brukardt" <randy@rrsoftware.com> wrote:
>>Probably. See AI-285. (http://www.ada-auth.org/ais.html).
> 
> ??? That link doesn't go to anything obviously pertinent, just the top
> of some huge and not pellucidly navigable database even whose
> component parts are confusingly labeled.
> 
> Do you have a direct link or is what you reference really on that page
> somewhere?

This, maybe? It's AI-285, anyway.

<http://www.ada-auth.org/cgi-bin/cvsweb.cgi/AIs/AI-00285.TXT?rev=1.6>

Ada and 32 bit Unicode character sets (still?) 64 bit addressing and OOP

[Kent Paul Dolan]

Hyman Rosen <hyrosen@mail.com> wrote:
> Kent Paul Dolan wrote:
>> "Randy Brukardt" <randy@rrsoftware.com> wrote:

>>> Probably. See AI-285. (http://www.ada-auth.org/ais.html).

>> ??? That link doesn't go to anything obviously pertinent,
>> just the top of some huge and not pellucidly navigable
>> database even whose component parts are confusingly
>> labeled.

>> Do you have a direct link or is what you reference really
>> on that page somewhere?

> This, maybe? It's AI-285, anyway.

> <http://www.ada-auth.org/cgi-bin/cvsweb.cgi/AIs/AI-00285.TXT?rev=1.6>

Ah, thanks.  You may well think me mad (many do) but I read
the whole thing, comments and all.

One can of worms remains hidden in what I read:  if you
allow the writing of source code in "native" character sets,
you pretty much must as part of that grant the ability to
use "native" quoting characters, and among the world's
languages there are many of them.  It would seem necessary
to separate out of the "punctuation" classes those pieces of
punctuation that are specifically quoting characters.  This
also adds the complexity that many national quoting
characters have left and right handedness, while the
existing Ada quoting characters are ambidexterous.  Probably
quoting syntax that includes demand for correct quote
matches by handedness would best satisfy the "principle of
least astonishment" for source code writers in those nations
/ languages / character set subsets.

And as held by some but not all of the commenters, if you
standardize these quoting characters for some, you should
standardize them for all, or you end up with "Tower of
Babel"-brand Ada.  I can hardly wait to start using those
dressy lower small double angle brackets to quote strings in
source code; they look so _sophisticated_.

I'd appreciate it if someone would pass this comment along
to whomever is collecting comments for AI-285.

Having made many lives miserable for a short while again,
I'll try to refrain from further comments on the Ada
specific issues I'm reading, after all, in
comp.lang.java.advocacy.

xanthian, whose standards committee (X3H3, X3H33) years
included and enhanced being a nit-picking pedant.

Re: 64 bit addressing and OOP

[Robert I. Eachus]

"Randy Brukardt" <randy@rrsoftware.com> wrote:
>Probably. See AI-285. (http://www.ada-auth.org/ais.html).

Kent Paul Dolan wrote:
> ??? That link doesn't go to anything obviously pertinent, just the top
> of some huge and not pellucidly navigable database even whose
> component parts are confusingly labeled.
> 
> Do you have a direct link or is what you reference really on that page
> somewhere?

http://www.ada-auth.org/cgi-bin/cvsweb.cgi/AIs/AI-00285.TXT?rev=1.6

Randy does an excellent job maintaining that database, and may consider 
your complaint insulting.  On the other hand, he may figure that anyone 
capable of reading the AIs finds navigating the database trivial.

This draft of AI-285 is a good example. You need a working knowledge of 
several ISO standards and the ISO organization, and an expert's 
knowledge of the Ada standard to make much sense out of the discussion. 
  To quote just the issue, not the proposed solution:

SC22 directed its working groups to provide support for the ISO/IEC 
10646 character set:

"JTC 1/SC 22 believes that programming languages should offer the 
appropriate support for ISO/IEC 10646, and the Unicode character set 
where appropriate."

Moreover, the working draft of ISO/IEC 10646:2003 makes use of planes 
other than the BMP.