[oclug] Form and Function
bb at L8R.net
Tue Jun 17 10:17:01 EDT 2003
On Tue, 17 Jun 2003 09:50:30 -0400
Brad Barnett <bb at l8r.net> wrote:
> On Tue, 17 Jun 2003 09:29:46 -0400
> "Francis J. A. Pinteric" <linuxdoctor at linux.ca> wrote:
> > On Tue, 17 Jun 2003 08:53:35 -0400 (EDT)
> > Phil Tanguay <ptanguay at magma.ca> wrote:
> > >
> > > This function over form business is a crock. Unless you can
> > > guarantee that you won't have any bugs AND that you'll stay with our
> > > company for the next 15 years, form is VERY important.
> > >
> > Both form AND function are both extremely important in software
> > development. What many people don't realize is that simply getting
> > the job done is not enough, but being able to efficiently track down
> > bugs and maintainability of the software by others than the original
> > author(s) is also important. Maintainability (not to mention
> > documentation) is crucial in software that is to be used for critical
> > purposes, especially embedded systems where they control machinery.
> You're missing the point though, Francis. Without function, form is
> useless. It is nothing. Therefore, form is primary, and function is
Obviously I have a typo above. ;)
"Therefore, function is primary, and form is secondary."
> Again, as I mentioned in another email, this does _not_ reduce form's
> necessity. It simply means that without function, form has little use.
> It also means that inhibiting function with form is equally useless, and
> that form must _always_ remain secondary.
> > To favour function over form is like saying that the means justifies
> > the ends, and the slippery slope that implies. Similarily, favouring
> > form over function is like dogmatism over practical experience. There
> > is need of, and room for, both.
> Yes, there is need and room for both, and they are very important.
> However, function is always primary. I can still get to work in a
> rusted out bucket of a car, but I can't in a car with a perfect body
> job, but a blown engine.
> Equally so, code that looks like crap and runs is better than code that
> looks good and does not run. The crappy looking code is certainly in
> need of good form, but it still functions, thereby proving that form is
> Or are you suggesting that ugly people can't think? ;)
> OCLUG general discussion list
> OCLUG at lists.oclug.on.ca
More information about the OCLUG