Re: ADA made me hate programming

[David Thompson <dave.thompson2@verizon.net>]

On Thu, 01 Jul 2010 17:16:52 -0400, Wilson
<leon.winslow@notes.udayton.edu> wrote:

> On Thu, 01 Jul 2010 13:11:41 -0400, <anon@anon.org> wrote:
> 
> >Plus:
> > Reference: "The Development of the C Language" by Dennis M. Ritchie
> >
> >         As for Algol, Ritchie states: "BCPL, B, and C all fit firmly
> > in the traditional procedural family typified by Fortran and Algol 60."
> > He never states that Algol or Fortran was used to build C.
> >
> >
> 
> C is a structured assembly language for the DEC PDP 11.  If you doubt  
> this, go back and examine the PDP 11 assembly language.  All of that ++  
> and -- are a standard part of the assembly language addressing modes.  In  

No they aren't. PDP-11 has only postinc and predec, and only for
addresses. C -- and B on the -7 before the -11 even existed -- has
all. The greater *use* of pointer postinc by early programmers was
likely influenced by the -11 but that's not the language.

> particular, all of those indirect references (pointers) were necessary  
> because the PDP 11 came with a 16 bit instruction set and only 8 bits were  
> allowd for the data address.   Fortran, Cobol and every other higher level  

This is nonsense. -11 instructions are 1, 2 or 3 words of 16bits.
*All* data addresses and offsets are 16bits. *Branches* have
*instruction* *offset* of 8 bits, or 6 bits for (optional?) SOB. 

> language of the time omitted pointers because almost all of their  
> addressing was direct.  (You might also want to compare C to Bliss another  

Nope. COBOL FORTRAN PL/I a68 and Pascal all had at least optional
by-reference arguments, which had to be implemented as pointer or
indirect of some kind, in the worst case (like PDP-8) by modifying
code (which was though okay and even clever in those days).
The latter three had explicit pointers, and COBOL had 'file' buffers
(and sort/merge and later comms) which were often relocated using
hidden pointers. (FORTRAN had ASSIGN for GOTO and FORMAT only.)

> sturctured assembly language for the PDP 11 and then ask which language  
> copied from where.  The two languages offer an interesting contrast.)   

BLISS was developed first for the PDP-10 and then extended to the -11
and later VAX and others. It was certainly similar to BCPL/B/C,
because all of them were designed to be close to 'the machine', but a
slightly generalized model rather than a specific machine. In fact
BLISS as a language is somewhat more capable than C because it didn't
have to be (self)compiled on the -11. 

> Richie started on the PDP 7, but quickly moved to the PDP 11 and may have  
> forgotten the details that led to the final result.
> 
> Also, before you quote Richie or any other author, you need to read a  
> peice by Isaak Asimov on how authors are the last people  to understand  
> what they did and how they did it.  <snip>

Asimov wrote quite a bit about the mysteries and vagaries of the
creative process, but that's not the same thing as technical
documentation or history. Considering that your 'facts' are provably
inconsistent with reality and Ritchie's aren't, I know who I believe.