Re: Some Laws

From: Laconic2 <laconic2_at_comcast.net>
Date: Fri, 1 Oct 2004 06:58:33 -0400
Message-ID: <gISdnXfLLZTtpcDcRVn-rw_at_comcast.com>


"Lemming" <thiswillbounce_at_bumblbee.demon.co.uk> wrote in message news:g28pl0lq0aboe98ef3qetpvuc0u5dhubv3_at_4ax.com...
> >There's nothing wrong with BASIC that can't be cured by learning how to
> >program! ;<)
>
> Although I realise your comment was tongue in cheek, I'm surprised to
> read it from you. From "Real Programmers Don't Use Pascal": '... the
> determined Real Programmer can write Fortran programs in any
> language".

The comment was definitely tongue in cheek. But it refers indirectly to the mission that BASIC had when it was first invented in 1965. At that time, the first generally useable timesharing system (CTSS) had been installed on several campuses, but the only thing non programmers could run was TYPESET, RUNOFF, and a few games. And learning FORTRAN was painful for some sociologist with an interesting prolem in statistical analysis.

The solution? Create a language that's easy for a non programmer to learn. That's actually brilliant. Dartmouth BASIC had built in operators like matrix multiplication. I daresay most geeks don't know how to use matrix multiplication, let alone when to use it. But in the hands of someone that understands matrix multiplication, it's a powerful tool.

Some very good programs were written by "non programmers" in Darmouth Basic.

Here's the rub. Programming in BASIC was so easy that it concealed the existence of another craft: the craft of programming well. Someone who already knew how to program could write good programs in it. Someone who learned Basic, and the learned some programming skills, could write good programs in it.

But someone who learned Basic, and never went on to learn programming, remained a beginner for a lifetime, making the same classic mistakes over and over. Hence my comment.

>
> The language doesn't define the programmer; the programmer is a good
> programmer or a bad programmer because of what s/he produces. A good
> programmer can produce useful programs any OS, language, whatever.
> The poor programmer is bound to a specific language.

Agreed. Like I said, there isn't anything wrong with BASIC that can't be cured by learning how to program!
We're really saying the same thing!

>
> I've only ever made my living writing COBOL. I've dabbled with OO
> languages - Java, C++ - and have moved rapidly away from them because
> it seems they make you "talk funny". If a junior COBOL programmer has
> a problem he can't quite get sorted, he goes to an elder COBOL
> programmer and asks some questions. THe elder says "Hmm ... ok, let's
> have a look at your code ..." and "... ah, are you checking that your
> file is open before you read from it?" ISTM that a junior C++
> programmer goes to another C++ programmer and says "I'm, a bit stuck"
> and the elder C++ programmer says "Ah, have you considered the
> mountain?". The junior says "huh?" and the senior says "The snow
> stays only in the winter, consider the lillies in the field, they do
> not reap, neither ..."

I skipped COBOL. But let me ask a dumb question... Why doesn't the compiler or the run time alert the COBOL programmer to that error?

The closest I ever came to COBOL style programming was Datatrieve, and Datatrieve wasn't a "real programming language".

>
> And quite sensibly, the junior makes his excuses and leaves.
>
> I would like to learn C++, yes I really would, but I don't think I can
> do the whole "OO koan" thing, and that's what puts me off. I want to
> write programs that do useful things (and I can do that in BASIC,
> Pascal and COBOL) I don't want to join some dopey religious cult.
>

What is the sound of one object clapping? Received on Fri Oct 01 2004 - 12:58:33 CEST

Original text of this message