[geeks] Misuse of Java

[Dave McGuire]

On Wednesday, November 6, 2002, at 12:08 PM, Jonathan C Patschke wrote:
>>   Well I can't say I agree fully with that point though, Jon...Many 
>> programs suck mightily due to excessive "design".  Design isn't a 
>> substitute for actually writing the code...While it should never be 
>> slapped together haphazardly, there *is* such a thing as "too much 
>> design", and many "professional" programmers do it routinely.
>
> I think we should agree to disagree on that point.  I'm probably one 
> of those over-designers.  Typically, I have more documentation[1] than 
> code, and have volumes of design documents on code that I never got 
> around to writing.  However, if I ever get bored (Ha!), all I have to 
> do is grep for a project.  If I'd written code instead of docs, I'd 
> first have to wrap my brain around the code, and then jog my memory to 
> see if I could remember where I was taking it.  As it is, the 
> hard-to-remember stuff is there, and the code should just be a 
> translation of the docs.

   Ok, let's agree to disagree. ;)  My point is...when it comes time for 
the project to *do* something, the design docs don't run...the code 
does.

> However, I have yet to see code that sucks because too much thought 
> was put into the design.  I've seen plenty of code that sucks because 
> someone decided that the design of the code should be "pretty", which 
> usually means "way, way more generic than the code could ever be 
> reasonably expected to handle".  This isn't "too much design"; it's 
> just "stupid design".  There's a really good parable that illustrates 
> this mindset:  http://www.qis.net/~jimjr/eng01.htm

   Well a lot of "paper engineers" only know how to design...not 
actually build...so that's all they do.

>>   I would suggest reading up on a philosophy known as "extreme 
>> programming" (actually I'll be surprised if you don't already know 
>> about it)...its proponents say very interesting and thought-provoking 
>> things on this subject.
>
> If this is the same philosophy that ties to people to the same 
> keyboard, then I've had to suffer through three semesters of that 
> nonsense at Rice.  I'm one of those ancient people that believes 
> communication problems can be resolved by adhering -stridently- to 
> whatever specifications/interfaces the programming team agrees upon.

   Well I haven't seen any references to that, but the basic idea is 
valuing good code over good documentation.  Too many "engineers" hide 
behind design and documentation "work" and never really PRODUCE 
anything that actually works.

         -Dave

--
Dave McGuire              "So you're really not gonna bite on the
St. Petersburg, FL              transvestite bait?"    -Doc Shipley