Developers: The myth of starting from scratch


During my recent web site redesign I decided I would start from scratch. I diligently worked and worked and finally had a ‘beta’ version of code that I deployed. It came out terrible.

I couldn’t believe it, how could I have gone so far wrong?

I had the positioning looking great, the images looked great, the screen effects looked great - yet there was no mistaking it. When IE got a hold of the page it looked like all the components were dumped in a pile in the middle of your screen.

It was about that time that I took a hold of the default Kubrick layout that comes with WordPress and began to hack it into my own creation. Many lines were excised, and many were not. Cloning Kubrick allowed me to get the central positioning and “browser resize messing up my content” issues addressed quickly.

My question was: Why did I need to start from scratch?

I’m not a web site designer, I don’t go from ex nihilo to something amazing. I don’t have a Photoshop-generated mock-up to follow. In short, there was no credible reason for me to start with an empty file called “style.css”. Yet I did it anyway.

2 weeks later I had a mess, 2 weeks after that, going the modifying Kubrick route, I had a working site that looked about the way I had imagined it.