On Thu, Apr 10, 2008 at 9:53 AM, Doug LaRue <[EMAIL PROTECTED]> wrote:
> ** Reply to message from "Bob La Quey" <[EMAIL PROTECTED]> on Thu, 10 Apr
>  2008 09:34:50 -0700
>
>
>
>  > On Thu, Apr 10, 2008 at 5:42 AM, Todd Walton <[EMAIL PROTECTED]> wrote:
>  > > On Tue, Mar 25, 2008 at 2:34 PM, Bob La Quey <[EMAIL PROTECTED]> wrote:
>  > >  > Agreed. Let's try and figure out what programming is,
>  > >  > and what problem it is trying to solve. I would be
>  > >  > satisfied with a decent statement of that.
>  > >
>  > >  Translating?  Making human wants known to the machine.  Successfully
>  > >  writing out what you want the machine to do, in the machine's
>  > >  language.  So, programming is inherently functional.  We don't program
>  > >  computers to tell them we love them or just to say, "I had a good
>  > >  day".  We program to make something happen.  Programming is action.
>  > >
>  > >  Though I suppose sometimes, when we program, we're just stating facts.
>  > >   But always facts relevant to what we're about to tell the computer to
>  > >  do.
>  > >
>  > >  So... programming is telling a machine what to do?  Seems pretty
>  > >  straightforward.
>  > >
>  > >
>  > >
>  > >  -todd
>  >
>  > If it is so straight forward then why is it so hard to do?
>  >
>  > I think it is because we do not know how to do it well.
>  > Which goes back to the fact that despite your statement
>  > do not know what it is we are trying to do or how to do it.
>  >
>  > BobLQ
>  >
>
>  Defining the problem is a very large part of the problem since doing even
>  the simplest of things requires many many minute details at the programming
>  level
>  be exact. But, like building a 20 story building, planning planning and more
>  planning
>  is required to actually do it right IF THE PROBLEM IS DEFINED. But just try 
> and
>  use
>  a design methodology like applied UML modeling and after a couple of months
>  designing, management will come down on you for not coding and some
>  developers will do the same.

And I personally think that both the management and the coders
are right. We do _not_ know enough to actually plan that well.
It is simply hubris to think we do.

At a certain point because we are so ignorant we must simply
start, make mistakes, correct and muddle through. It is one
hell of mess, based on ignorance. True ignorance, i.e. nobody
really knows how to do it.

>  IMO, these two things have more to do with why software is so often over
>  budget and problematic at best if it is delivered. That and the fact that 
> there
>  are almost no software architects in the market since it is not taught or
>  encouraged.

I agree that the solution, if one exists, lies in making
software architecture into a much more serious subject. I
do _not_ see that it is there yet. I would be happy if you
would enlighten me as to the best knowledge base on the
subject. It has been a while since I looked into it.

>  So did I miss something, cause this thread/posting seemed to come out of the
>  blue?

The original post was prompted by my frustration with the
very brightest guys on this list being bogged down (my perception)
in endless discussion of the details of a variety of general
purpose computer languages, e.g. C++, Python, Java, etc. The
details are relatively easy to discuss and the discussion like
much shop talk serves some purpose if only to allow people to
vent but some sense of deeper understanding is completely missing.

Hence the "Wise men and the Elephant" subject line.

What is a general purpose computer language really supposed
to do? Is is nothing more than a collection of tools for
specific purposes? If so then what tools? If not then what
is it?

BobLQ "Still frustrated"


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to