Re: The Last Word on Comments (was Re: Hungarian notation)

[shane@wonk.demon.co.uk (Shane Badham)@wonk.demon.co.uk (Shane Badham)]

In article <4p3dad$mia@charm.magnus.acs.ohio-state.edu> fulton.52@osu.edu (Mike
Fulton) writes:
>In article <4p2jau$hkp@seagoon.newcastle.edu.au>, peter@tesla.newcastle.edu.au wrote:
>>Clark L. Coleman (clc5q@mamba.cs.Virginia.EDU) wrote:
>>
>>>The real problem in academia in this regard is the continual assignment
>>>of projects that will only take 1 to 3 pages of final source code.
>>
>>This is indeed the problem, and I don't know how to solve it.
>>Those of us with experience in real-world projects know that
>>there's a radical difference between the techniques needed to
>>write a big program and those needed to write a "Hello world"
>>program.  Most students don't understand this, and frankly I
>>don't know how to get the message across.
>
>I'm a soph at Ohio State, just finished my second modula-2 class and I agree 
>with the two preceding points 110%.  I started college at the age of 28 and 
>have coded in everything from 6809 assembler to MS VBA (but never modula 
>until now), starting at the age of 12.  But, this is my first time to learn 
>correct structure--wish someone would've told me this stuff when I was 12 :)  
>So, maybe my perspective is odd, but here's my idea.
>
>Why don't schools design projects/labs that build upon themselves.  Every one 
>of my projects could have been done without doing the preceding projects.  
>When we implemented Abstract Data Types (ADT), they gave us the .def files and 
>told use how to use the procedures.  Why not have us build those 
>implementation modules first.  Start with what the language gives you and then 
>extend it.  Create an ADT from the ground up.  Then have us use the ADT we 
>made to implement other ADT by layering them over our first ones.  After we 
>build enough structures and routines, have us use them to create a real 
>project.  Something that would be difficult to build from scratch, but 
>moderately easy using OUR tools.  This would truely show us the value of 
>starting with good data structures.  And it would force us to write readable 
>code and use good commenting, since you have to use today's code 1 or 2 months 
>later, or even next quarter.
>
>I realize this would cripple the student who had trouble in the beginning.  Or 
>maybe it would bring the problems to light sooner (by force) so the student 
>could get back on track.  I think you must understand the basic structures or 
>there's no point in going further anyway.
>
>Just my thought :)

Mike,

Thats good input.  We did something similar on our Ada programming assignments.
 However, the tutors did not have us do all the bottom up packages.  Instead
they
_provided_ many of the packages, pre-written.  We then used these in our
application and wrote some ourselves.  This was the older Object based Ada,
similar to Modula-2 in concept.

The effect was that of belonging to a larger team on a much larger project. 
The scope of the project was large, not "itsibitsy" apps.

Regards,    Shane.
shane@wonk.demon.co.uk
"A closed mouth gathers no feet."