Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!ames!amdcad!sun!pitstop!texsun!convex!mozart!rowan
From: rowan@mozart.uucp (Steve Rowan)
Newsgroups: comp.lang.fortran
Subject: Re: FORTRAN 8x parser/lexer .NE. compiler
Keywords: FORTRAN 8x
Message-ID: <663@convex.UUCP>
Date: 20 Oct 88 14:03:16 GMT
References: <656@convex.UUCP> <15739@agate.BERKELEY.EDU>
Sender: news@convex.UUCP
Reply-To: rowan@convex.COM (Steve Rowan)
Distribution: comp.lang.fortran
Organization: Convex Computer Corporation, Richardson, Tx.
Lines: 38

In article <15739@agate.BERKELEY.EDU> link@stew.ssl.berkeley.edu (Richard Link) writes:
>>
>>Granted, an Ada compiler is simpler than a FORTRAN 8x compiler
>
>Oh my God! Is this really true?
>
>The biggest gripe against Ada is it's size; in the same vein that an
>elephant is a mouse designed to mil specs.

My assertion that an Ada compiler will be simpler that a FORTRAN 8x
compiler is based on the following observations.

1. FORTRAN 8x contains every feature of Ada except tasking.

2. EVERY legal FORTRAN 77 program is a legal FORTRAN 8x program.  That
   means one has to deal with insignificant blanks, interspersing DATA
   statements with executable code, no reserved words, and so forth.  No
   one has proven to me that all the modern language features can be
   parsed in program units with statements like

	 DO10I=1.5
	 IF(I)=5
	 DO10I=1,5
	 IF(I)5,5,6
   
3. Any production quality FORTRAN 8x compiler will have to accept all the
   standard extensions like REAL*4 and INCLUDE statements that X3J3 has
   chosen not to standardize.

4. FORTRAN 8x has features like the powerful array notation that do not
   exist in Ada.

I could go on but I am sure X3J3 will address many of these issues when
it responds to my public review comment and those of the 600+ other
concerned FORTRAN users and vendors.

Steve Rowan
convex.COM