Re: forth/fifth generation languages?

Re: forth/fifth generation languages?

[Dave Griffith]

In article <1993Aug9.084733.16628@rulway.LeidenUniv.nl>  
ruiter@ruls41.LeidenUniv.nl (Jan Peter de Ruiter) writes:
> In article <FOUTS.93Aug8152353@cello.hp.com>, fouts@cello.hp.com (Marty
> Fouts) writes:
> 
> |> fifth generation:	A myth the Japanese spent a *lot* of money on
> |>   About a decade ago, MITI decided it would spend 10 years and a lot
> |> of money applying AI to programming, thus solving the software crisis.
> |> The project spent its money and its ten years and just recently closed
> |> down with a wimper.  It looks a lot like programming languages have
> |> gotten about as good as they are going to, probably peaking at C, and
> |> that we are currently in a period of regression, including such
> |> languages as Ada and C++.
> |> 
> 
> Unfortunately, you're not telling us what the goal of the 5th generation
> languages was. I only know that they failed, but what were they trying
> to do?
> 

Hell, most of the time they barely knew what they were trying to do, why should
  
you?  The idea, cynically put, was that with sufficiently advanced hardware and
  
some fancy new language constructs, functional and logic programming would be  
practical for large-scale use.  It was thought that these would be necessary  
for the advanced applications of the 21st century (speech and text recognition 
 
were especially targetted application areas).  Unsurprisingly, this turned out 
 
to be money down a rat-hole as the rest of the world went to OO.  The fifth  
generation project is just about the perfect thing to point to when someone  
tries to argue for government direction of computer science research. 
--
Dave Griffith, Information Resources, University of Chicago,
Biological Sciences Division               dave@delphi.bsd.uchicago.edu
"The faults in bad software can be so subtle as to be practically theological"
--Bruce Sterling

Re: forth/fifth generation languages?

[agate!usenet.ins.cwru.edu!news.ecn.bgu.edu!wupost!math.ohio-state.edu!pac]

In article <FOUTS.93Aug8152353@cello.hp.com> Martin Fouts <fouts@hpl.hp.com> wr
ites:

>It looks a lot like programming languages have
>gotten about as good as they are going to, probably peaking at C

Of course, to some of us, the peak is called "Algol-60", and C is
a deep trench filled with primordial ooze.

Re: forth/fifth generation languages?

[cis.ohio-state.edu!math.ohio-state.edu!darwin.sura.net!dtix.dt.navy.mil!c]

I agree with previous postings about the differences between 1st,
2nd, 3rd, and 4th generation languages.  The statements
accurately represent the conventional interpretation of the
progression. 

I'd like to share an observation that there were correlated
correlated changes in what a programmer was required to know and
how programmers worked with other people.

The 1st generation programmer knew the hardware wiring diagram as
well as every detail of the code.  There were no libraries to
speak of and the applications - while revolutionary - were small
are exceedingly rare.  

The 2nd generation programmer worked with an abstract model of
the hardware.  2nd generation programmers also had a library
system so that they could employ many functions without coding
each one.  This freed them to make larger applications that were
numerous enough to appear in many application domains.

The 3rd generation marked a major change in programming because
fairly ordinary people could master both an application domain
and enough programming skills to write code for it.  That wasn't
the intention, but some line was crossed and the nature of
programming work changed.  Now the sky seemed the only limit.  If
you found a team of domain experts and a team of programmers who
understood domain jargon, you could develop anything, it seemed,
cost not withstanding!

The 4th generation language marks a paradigm shift.  A fourth
generation language is usable by an application domain expert who
is not a programmer at all.  In this paradigm, the programmer is
no longer needed to translate requirements and equations into
software.  Costs go down, and delivery times are shortened
relative to the 3rd generation.

Whatever happens to computer languages, the way people work
together will continue to evolve and that evolution may be more
important than language features.  

Two short, related points:

1) There doesn't have to be a syntax difference between 3rd and
4th generation.  During the STARS Foundations era, I showed how
to do megaprogramming - the synthesis of very large programs from
correspondingly large components - using Ada as a 4GL.  Under the
same program, Lockheed demonstrated that Ada is competent to
express statements usually stated in SQL.  If nobody has a 4GL
for your application - consider shifting your paradigm and
sticking with the established languages: Ada and C++.

2) There are still hard problems associated with 4GLs:
  * 4GLs don't cover all the application domains
	* 4GLs can't cross domains to meet all the requirements of
	  a large system. 
  * 4GLs cost a bundle to develop, and they are usually 
	  nonstandard and proprietary.   

-----------------------------------------
Paul L. Baker     pbaker@cta.com 
CTA Inc.
6116 Executive Blvd. Suite 800
Rockville, MD 20852
(301) 816-1242

Re: forth/fifth generation languages?

[Robert I. Eachus]

In article <1993Aug9.100625.462@sei.cmu.edu> firth@sei.cmu.edu (Robert Firth) w
rites:

   > Of course, to some of us, the peak is called "Algol-60", and C is
   > a deep trench filled with primordial ooze.

     Come on, Robert, everyone knows that Algol-W was a significant
improvement on Algol-60.  (If you consider Algol-W to be a dialect of
Algol-60, ignore the chiding tone. :-)

     But I do agree that no language before or since has been anywhere
near as beautiful.  It is possible (with effort, but when has
programming not required effort) to write pretty code in Pascal and
Ada, and it is possible to write PL/1, FORTRAN, Algol-68 and even
COBOL that is presentable, but C at its best is ugly.


--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

Re: forth/fifth generation languages?

[Robert Dewar]

Just for the record, there was a fair amount of "primordial ooze" in 
Algol-60 (see for example Knuth's article on "The Remaining Trouble Spots
in Algol-60") Speaking as one who was involved in the attempted 
standardization of Algol-60-modified ten years later, I can assure you
that there were even at that point a number of serious holes.

I also find the syntax junky in places (e.g. the dangling else rule), and
the semantics of procedure calls remains murky (although it was cleared
up in a60-modified, in particular what does it mean if you don't give 
types to the formal parameters -- no one implemented this of course on
typical machines where it is quite important to distinguish integers and
reals).

Also there were huge functional holes in A60, including lack of I/O,
lack of any string handling capability, lack of any parallelism, and lack
of any kind of data abstraction.

Sure I realize that there were elegant features, but, even though this is
a bit off subject for the Ada newsgroup, I think it is important to
counteract a false impression that somehow A60 was this jewel of perfection.
It failed for many reasons, but one of the reasons was errors of judgment
on the part of the designers -- it was by no means perfect

Re: forth/fifth generation languages?

[cis.ohio-state.edu!math.ohio-state.edu!darwin.sura.net!seas.gwu.edu!mfeld]

In article <EACHUS.93Aug9150728@spectre.mitre.org> eachus@spectre.mitre.org (Ro
bert I. Eachus) writes:
>In article <1993Aug9.100625.462@sei.cmu.edu> firth@sei.cmu.edu (Robert Firth) 
writes:
>
>   > Of course, to some of us, the peak is called "Algol-60", and C is
>   > a deep trench filled with primordial ooze.
>
>     Come on, Robert, everyone knows that Algol-W was a significant
>improvement on Algol-60.  (If you consider Algol-W to be a dialect of
>Algol-60, ignore the chiding tone. :-)
>
Was it Dijkstra who said "Algol-60 was a significant improvement on all
its successors"?

Mike Feldman

Re: forth/fifth generation languages?

[munnari.oz.au!sol.deakin.OZ.AU!usenet]

In Article <24b6v7$lo6@cville-srv.wam.umd.edu>
krc@wam.umd.edu (Kevin R. Coombes) writes:
>>I think, a language is not made more complex by adding a feature. 
>>
>
>And so, by induction, a language is not made more complex by adding
>arbitrarily many features ? Even infinitely many ?

Correct (note that the word "necessarily" is implied by the context of the
original quote).

In a similar way, a telepone book is not made more complex by the addition
of arbitrarily many entries, merely larger.

Re: forth/fifth generation languages?

[Wes Groleau x1240 C73-8]

In article <CBKG89.4vx@zoodle.robin.de> uli@zoodle.robin.de (Ulrich Grepel) wri
tes:
>In article <FOUTS.93Aug8152353@cello.hp.com> fouts@cello.hp.com (Marty Fouts) 
 
>writes:
>> Almost no one programs in machine language anymore, because
>> translators are nearly trivial to write.
>
>That's not true. On mainframes most patches are done this way. painframe
>indeed.

>Not only mainframes.  A translator is not trivial when your machine has 
only 16K of magnetic core, no external interfaces, small scale integration
(RTL), only 18 instructions, a design quirk that makes re-entrant subroutines
impossible, ... there's more, but you get the idea.  A stunning example of
"If it works, don't mess with it!"  Of course, I taught someone else how to
program it and moved on to "real" computers   :-)