[Gllug] More Microsoft patents

[Jason Clifford]

On Wed, 12 Feb 2003, Doug Winter wrote:

> I think there is a fundamental difference in perl, which is to do with
> the implied data storage.  Someone else (I can't remember who) made the
> analogy with FORTH - because all the changes to data happen "off the
> page" (in the case of FORTH, on the stack) you have to "run" the code
> in your head. 
> 
> Most languages explicitly avoid this problem by only allowing named
> variables, and using strict scoping rules.  These give the languages are
> more verbose "narrative" quality that makes it far more maintainable.

Exactly as you do in perl with the use of the "Use strict" pragma.

I don't know of any serious perl coder who would write anything without 
it.

> Comments are, IMHO, a bit of a red herring in my view.   Code should be
> understandable without them - comments are there to provide depth to
> understanding and explain what is intended (which is not necessarily
> what is achieved).  

Indeed and for that very reason good comments make for maintainable code 
while undocumented code that works but doesn't produce expected results 
doesn't.

It's quite possible to have code that is correct but contains a logic 
error leading to errors as you no doubt know. Having a comment in place 
saying "this bit should to this..." helps to identify these - particularly 
for the person who later has to fix the original problem.

> If you have to put a line of comments for every line of code to explain
> what it means, then that code is bad.

And this is relevant how? Nobody is advocating that.

Jason Clifford
-- 
UKFSN.ORG		Finance Free Software while you surf the 'net
http://www.ukfsn.org/			Sign Up Now


-- 
Gllug mailing list  -  Gllug at linux.co.uk
http://list.ftech.net/mailman/listinfo/gllug