Dan Abel wrote:
> In article ,
> "J. Clarke" wrote:
>
>> sf wrote:
>>> On Sat, 06 Feb 2010 08:38:28 -0800, Dan Abel
>>> wrote:
>>>
>>>> I remember my first computer program. I wrote it in Fortran II,
>>>> and it ran on an IBM 1130. That was in 1967.
>>>
>>> You didn't have to learn BASIC first?
>>
>> This notion of learing BASIC first is misguided--it teaches horrible
>> habits--at least the versions that were available in the late '60s
>> and early '70s did. When I was in school they started us on Fortran
>> and in the next course we got assembler. After that we could use
>> APL.
>
> BASIC is really pretty bad as far as teaching bad habits. However,
> Fortran is often worse. Assembly is probably worse. Never used APL.
At least with Fortran you don't have to do everything with GOTOs. With
Assembler you have to construct each branch explicitly--you learn what goes
on under the hood so to speak, but you're right as far as habits go--it's
very easy in assembler to find yourself in the middle of a plate of
spaghetti.
APL has such a heavy set of array manipulations built in that you seldom
need branching at all. The trouble with it is that if you get into the APL
mindset it becomes difficult to work in other languages until you've
constructed yourself a toolkit that emulates APL. It's also interpreted, so
you can use it as a desk calculator--one line of APL can do what hundreds of
lines of other languages can do, so it's an exceedingly powerful desk
calculator. A disadvantage of this is that it's easy to create write-only
code--code that is so compact and dense that nobody can figure out what it's
supposed to be doing, including you six months down the road.
This whole business of structured programming was very new when I was
learning--object oriented and event driven were simply unknown.
>> Later on the beginner course was Pascal then students moved on to C
>> and assembler.
|
|
|
