Re: advantages of forth over other languages

John Passaniti wrote:
> John Doty wrote:
>> At the cutting edge, the best solution is unobtainable. The choice is 
>> solution or no solution. Later on, when you understand what you're 
>> doing, you can work on "best". But again, programming knowledge is of 
>> limited use in judging what "best" means in a real application. The 
>> domain rules.
> 
> Well, we're back to this nonsense.  As we've all seen countless times 
> before, you appear to live in a world where someone with the job title 
> "programmer" is little more than a code monkey, hired only because they 
> can take the ideas of the "domain expert" and express them in code.  The

> extent of how much this strict stratification is real or is the product 
> of artificial barriers you construct in your mind is anyone's guess.

In my experience it's barriers in *programmers'* minds. I see powerful 
evidence that that barrier exists in *your* mind, but you are blind to 
it. It does not exist in mine. If a program is the right solution, I'll 
write a program. But I might also compute a logarithm with a diode (what 
fraction of programmers would ever consider that alternative?).

You give lip service to the idea that programmers are also domain 
experts, but then you also refer to programmers as masters of 
specialized arcana like "design patterns". Very few real domain experts 
know what a design pattern is.

I mostly work with experimental physicists. Compared to them, 99% of 
people who call themselves "programmers" are just code monkeys. But 
physicists can write pretty good code.

You're like the fish that has reached the surface of the water and 
thinks he's reached outer space. You have a long way to go.

> 
> All I can tell you is that companies that run on that model (strict and 
> limiting roles described by job titles) is increasingly rare in my 
> experience.

Sure. But do the programmers actually master the domain, or do they just 
pretend? Do you even know how to tell?

Programming as a profession seems to be particularly attractive to 
narrow minds. I guess the artificial world of the computer is less 
threatening than the real world.

> 
>> Thanks. But this stuff comes easily to me (one minor project I'm 
>> working on these days is learning Verilog). It *doesn't* come so 
>> easily to others whose expertise I must rely on. Genius takes many 
>> forms. So it's im****tant to keep the methodology as straightforward as 
>> possible, even if something more sophisticated might be "best" by some 
>> irrelevant metric.
> 
> We also seem to have a different definition of "genius."  For you, it's 
> apparently someone who is completely invested in some domain and never 
> strays outside that narrow role.

No. Usually it's somebody with multiple strengths. But nobody is good at 
*everything*, although gaining skills even in areas where you can't be 
excellent is valuable.

>  Your "genius" can't take on the role 
> of programmer, because gosh, that's a completely different title.  And 
> we all know how im****tant titles are to you.

You choose to completely misunderstand my point.

> 
> That to me isn't a genius.  That's someone who knows a lot about 
> something.  Real genius is interdisciplinary.  Real genius is not only 
> coming up with brilliant ideas and insights, but in being able to 
> realize them competently.

Indeed. And that's why programming systems must cater to the broad needs 
of genius rather than to the narrow prejudices of code monkeys.

> 
>> Or because they are free from the mental box of programming, and have 
>> other priorities outside of it. That can make someone extremely 
>> effective at getting work done with a computer, even if their programs 
>> are chaotic messes. Of course, cleaning up after such people is a 
>> valuable skill, too.
> 
> A "genius" who is "free from the mental box of programming" is someone 
> who is next to useless.

You just contradicted yourself. "Real genius is interdisciplinary." How, 
then, can someone restricted to programming be a genius?

>  What we're ultimately talking about here are 
> systems that must be expressed in terms of hardware and software.  A 
> "genius" who can't conceptualize their design in terms of real-world 
> hardware and software is engaging in something abstract.

And programming is the most abstract form of engineering.

> 
> Nearly as useless is the "genius" who "gets work done with a computer" 
> but that work is impractical.  I worked with a "genius" who came up with

> a beautifully elegant way to express a digital signal processing task. 
> He even came up with software that did the signal processing to show it 
> operating.  The only problem was that his software didn't run in real 
> time (a requirement), and he and I later proved that his software 
> couldn't run in real time on the hardware we had available at the time. 
>  Physical limits like cost, processor and memory bandwidth made it 
> impossible.

That's pretty much what happened with the Hubble planning software but 
the other way around: a team of professional programmers at a NASA 
contractor developed, at enormous expense, a system that could not keep 
up with real time. The software that worked was written by an 
astrophysicist with spacecraft operations experience. "Real genius is 
interdisciplinary."

I can think of quite a few other examples of this sort. Programmers tend 
to overvalue elegance and logic. They also tend to optimize the wrong 
things. Domain experts with problems to solve tend to be more practical.

> 
> If that is a genius to you, then you have a low threshold for genius. 
> Again, the true genius is one who can cut across disciplines and realize

> systems from the bottom to the top.  True genius is a mixture of both 
> scientist and engineer, and sometimes marketing and management.

I agree. But I have rarely encountered anyone like that who would call 
themselves a "programmer".

-- 
John Doty, Noqsi Aerospace, Ltd.
http://www.noqsi.com/
--
History teaches that logical consistency is neither sufficient nor 
necessary to establish practical, real world truth. Those who attempt to 
use logic for that purpose are abusing it.