Re: Forth and Co - The Return of the Jedi

Jerry Avins wrote:
John Doty wrote:
Jerry Avins wrote:
John Doty wrote:

...

So PolyForth wasn't Forth?

It was a pair of languages based on Forth ideas. Forth isn't a language, it's a methodology. Moore called it "A New Way to Program a Computer", not a new language.

Forth, like other languages, has words and sentences. Do you seriously contend that Forth is _not_ a programming language?

There are many Forth languages. Forth is a methodology.


Certainly ColorForth shows that Chuck Moore does not appreciate stagnation!

One person's stagnation is another's consistency; both are useful. Being predictable is a large part of being reliable.

That's the reasoning behind the infamous Ariane V rocket failure. Don't revise the software, even to remove useless calculations, because the results might be unpredictable. But when the useless calculation threw an uncaught exception...

Excess baggage is never a good idea.

Exactly. So why does Standard Forth require support for an interpreted pidgin language? It's been 30 years since Sachs demonstrated that it is unneeded.

I got burned, once, by removing code that I thought was a vestige, only to learn that some I/O chips -- not the one that happened to be plugged into the test board -- needed a substantial delay before they would operate properly.

When you're revising code, you need to understand the requirements. But you also need to understand the requirements when you decide *not* to revise the code for a new application area, as the Ariane V example shows.


Simplicity is a large part of being reliable, so what is one to make of a Forth standard that *mandates* unnecessary complexity? Every Standard Forth system must implement *two* languages with different grammar, but confusingly overlapping vocabulary. So unnecessary. So inconsistent.

It's perfectly consistent with all the early Forths I used.

Yes, but it has been obsolete for 30 years.

That you've "progressed beyond" them doesn't unForthify (deForthicate?) what came before you.

Forth is a methodology. The pursuit of simplicity is a critical part of that methodology. If you don't follow that methodology, you are no longer on the Forth path.


Fortran code from the 1960's, before Forth existed, still runs. Maybe you should consider Fortran for your next high reliability project. ;-)

Nah. I never liked punch cards.

Vacuum tubes have much more consistent parameters than transistors. Were vacuum tube computers more reliable than transistor computers?

One of their consistencies was short life.

So, your statement that "Being predictable is a large part of being reliable" is hardly universal.

Extreme unpredictabilty on small scales can lead to extreme predictability on large scales.
--
John Doty, Noqsi Aerospace, Ltd.
--
Specialization is for robots.
.

Relevant Pages

  • Re: How would you do this in forth?
    ... that give the language its syntax. ... forth based methodology to the forth language, ... portable by any Forth standard I know of, but still it will port to ...
    (comp.lang.forth)
  • Re: new here, my lang project...
    ... related to OOA/D rather than OOP. ... stability and maintainability are primary goals for OO ... So your language needn't worry too much about OOness and this forum ... employing any methodology consistently and not using a methodology. ...
    (comp.object)
  • Re: Forth and Co - The Return of the Jedi
    ... Forth isn't a language, it's a methodology. ... Moore called it "A New Way to Program a Computer", not a new language. ... you should consider Fortran for your next high reliability project. ... Were vacuum tube computers more reliable than transistor computers? ...
    (comp.lang.forth)
  • Re: why learn C?
    ... Understanding algorithms does not help you understand languages, ... language that is targeted towards that methodology it is then easier to ... only have to learn the language, not the language and the methodology. ... right now i am putting emphasis on learning algorithms. ...
    (comp.lang.c)
  • Re: Users of the BON notation among Ada users ?
    ... Because, as a metter of fact, UML is the only standardized 4GL. ... over time (or whose language was taught at Universities etc.). ... For the same reason I dislike Ada ... For what regards enforcing a precise methodology through UML, ...
    (comp.lang.ada)