On Sat, 2010-04-24 at 07:14 -0400, David McGlone wrote:

> On Friday 23 April 2010 07:45:16 Michiel Sikma wrote:
> > On 23 April 2010 13:15, David McGlone <da...@dmcentral.net> wrote:
> > > Is there a good strategy to studying PHP?
> > >
> > > For instance, is there a way to break everything down into small
> > > managable topics?
> > >
> > > --
> > > Blessings,
> > > David M.
> >
> > Just build what you want to build and look things up when necessary. The
> > biggest part of learning any language is becoming able to creatively solve
> > problems within its realm of possibilities, and that's something you can't
> > really learn just by reading and solving textbook problems. Hands-on
> > experience is a very important key to understanding.
> >
> > Keep an open mind, ask people for answers if things get difficult, always
> > assume that there's a better way to do something, and in a while you'll be
> > able to write perfectly decent programs.
> This is the route I have been following for the past couple years and I have 
> always assumed there is a better way to do something and that is, in a way 
> what makes me lose confidence.
> I always feel like my code is crap and sometimes even ashamed to even show 
> any 
> of it.
> -- 
> Blessings,
> David M.

For personal sites, I think as long as the code works, and doesn't have
any glaringly obvious security holes in, then it should be fine.
Personal sites tend not to have a huge amount of traffic, so speed
doesn't really become an issue unless you're doing something totally
crazy (like generating all of your header images on the fly with PHP for
example without any caching)

For commercial sites, more time should be spent on security and testing,
and working over things in your head about how the site is going to get
used. If an area of the site that you feel is going to be very popular
will be busy, then is creating a lot of large objects and making many DB
calls such a great idea?

I look at some code I did a while back and shudder. It started off well
enough, but then feature creep set in and now it looks like a creation
of Frankenstein! I think it happens to all of us at some point, it just
depends how well we deal with it. I'd rather have less then stellar code
than have to tell a client it's going to take more time to add a feature
and then have them go elsewhere.


Reply via email to