Path: utzoo!utgpu!utstat!jarvis.csri.toronto.edu!mailrus!cwjcc!tut.cis.ohio-state.edu!ucbvax!agate!violet.berkeley.edu!jerry
From: jerry@violet.berkeley.edu ( Jerry Berkman )
Newsgroups: comp.lang.fortran
Subject: Re: Fortran 8x ?
Message-ID: <20567@agate.BERKELEY.EDU>
Date: 18 Feb 89 20:10:34 GMT
References: <20377@agate.BERKELEY.EDU>
Sender: usenet@agate.BERKELEY.EDU
Organization: University of California, Berkeley
Lines: 67

In article <20377@agate.BERKELEY.EDU> naparst@garnet.berkeley.edu (Harold Naparst) writes:
>
>During the public review period for Fortran 8x last spring, I sent in 
>a critique of a lot of the new features in Fortran 8x.  I never got a
>response.  Can anyone tell me what is going on with this new language?
>
>Harold Naparst


The language has changed a lot since the public review.  I attended the
X3J3 meeting last week in Palo Alto last week as a visitor.  Here is my
impression of the status of Fortran 8X.  I am not a member of X3J3,
this list is unofficial.  It is also incomplete; there have been too
many changes to have a complete summary on the net.  For that, you must
read the document in the next review cycle.

In past meetings, the actions have included:
	- removed the concept of deprecation
	- removed the new form of DATA statement
	- removed generalized precision
	- added KIND= attribute to specify processor dependent alternate
		forms of intrinsic types (except LOGICAL)
	- added binary, octal, hex. constants in data statements and
		corresponding data edit descriptors in formats
	- added INCLUDE

In the Palo Alto meeting last week, the following actions were taken:
	- added a pointer attribute
	- added "nonadvancing I/O" which can be used to write without CR/LF
		or to read long records in pieces and removed "PROMPT="
	- allowed KIND= with LOGICAL type
	- added FLOOR, CEILING, MODULO intrinsics.  MODULO differs from MOD
		in treatment of negatives
	- allow characters, structures, and other intrinsic types
		to all be part of the same common block as long as
		long as certain rules are followed

The committee spent this meeting trying to resolve remaining technical
issues and to get a draft standard which would also be acceptable to
ISO, the Internaional Standards Organization, so that both ISO and ANSI
will have the same standard for Fortran.

X3J3 passed a resolution that no new technical proposals may be raised
after this meeting and that no technical issues will be revisited.  The
next meeting, May 8-12, Port Jefferson, Long Island, New York, will be
concerned with polishing the draft, i.e. making minor editorial changes
and fixing bugs and inconsistencies in the draft, and responding to the
public comments received last year.

It is likely that the draft will be forwarded on to X3, the parent body
of X3J3, after the May X3J3 meeting.  If not, then ISO may adopt it's
own standard.  Since there have been substantive changes to the draft,
X3 will hold a second public review period.  X3J3 voted in favor of
another 4 month public review period; however X3 rules call for only a
2 month public review period.

The previous public comment resulted about 450 letters which included
about 5700 individual comments.  The committee intends to answer each
letter, comment by comment, and is in the process of drafting answers.
But first the draft has to be stabilized.  I.e., it would have been
impossible to answer a comment for or against pointers until last
week.  Similarly, about 10 comments thought PROMPT= was not general
enough; now the committee has addressed that problem.

	- Jerry Berkman
	  U.C. Berkeley, Central Computing Services
	  jerry@violet.berkeley.edu