Re: 8 bit characters, POSIX, IEEE and flames (Was Re: Two "Simple" I/O Questions)

Re: 8 bit characters, POSIX, IEEE and flames (Was Re: Two "Simple" I/O Questions)

[David Emery]

I can assure everyone that the current IEEE standard is *NOT* a Xerox
copy of the balloted draft with a couple of text edit changes.  In
fact, it was phototypeset at no little expense.  By IEEE rules, the
actual standard MUST BE the same as what was balloted; would you have
it any other way?  We are still working with the IEEE to get them to
release/relax the copyright on the package specs so they can be made
available via FTP or whatever.  Currently, they are considering
providing the package specs on a floppy disk with each copy of the
POSIX 1003.5 standard.    (Current holders of the standard would be
entitled to a copy of this disk presumably for free.)  

As I have pointed out before, the IEEE uses income from sales of
standards to support their work on standards.  Although I am not happy
with the prices they charge, I can assure you that the IEEE
professionals who work on standards (or at least the project edtior
who worked on the POSIX standards) earn every cent of their pay.  

IEEE policies on standards copyright are the topic of much debate
within the IEEE; contact your IEEE or IEEE Computer Society officer to
express an opinion.

Incidentally, as has been discussed here earlier, ISO WG9 voted last
June to change the definition of the Adas type CHARACTER to be an
8-bit character, with the high-order values those of Latin-1.  This is
not a mandatory change (i.e.  it's not a "binding interpretation" in
AI/ACVC terms), but people should stringly encourage their compiler
vendors to support this change.
				dave emery
				(P1003.5 Technical Editor)

Re: 8 bit characters, POSIX, IEEE and flames (Was Re: Two "Simple" I/O Questions)

[Alex Blakemore]

In article <EMERY.93Feb6222019@dr_no.mitre.org> emery@dr_no.mitre.org (David Em
ery) writes:
> I can assure everyone that the current IEEE standard is *NOT* a Xerox
> copy of the balloted draft with a couple of text edit changes.  In
> fact, it was phototypeset at no little expense.

Dave and all,

I must apologize for not reading the stamp on the front that says the
current distribution is is manuscript form prior to publication I was
under the impression that what they were selling now was the finished product.

I was annoyed that for the amount of money they charge ($40-$60), the
standard was not bound - just loose unpunched paper.  I already had
something like that from the balloting.  For that price, I expected
some kind of book. If its typeset, its hard to tell from a good copy.
Its wonderful that the IEEE wants to release an early manuscript, but
I wish they had made that clear.  I would have saved my money.

Perhaps someone will save a few dollars by waiting till the bound
version is ready if thats what they want as a result of my last flame.
If you dont have a copy from the balloting, then its probably worth it
to buy this early copy irregardless of the form.

> Currently, they are considering
> providing the package specs on a floppy disk with each copy of the
> POSIX 1003.5 standard.    (Current holders of the standard would be
> entitled to a copy of this disk presumably for free.)  

This would be a nice benefit but may be unnecessary for everyone.
I'ld be happy if they just made it available to people who request it
(in some electronic form).

> I can assure you that the IEEE
> professionals who work on standards (or at least the project edtior
> who worked on the POSIX standards) earn every cent of their pay.  

The document is very professionally done, I never questioned the
quality of the work.

The editing changes that I mentioned involve things like moving the
rationale and notes to an appendix (which are not officially part of
the standard).  I did not suggest that any content changed after balloting.
-- 
---------------------------------------------------
Alex Blakemore alex@cs.umd.edu   NeXT mail accepted