notes on a site design

I really envy web designers who seem to fling out an entirely new design for their home sites every few months, designs that are spectacularly beautiful, original, and creative, and that employ all the latest standards and software. And they do it all while retaining day jobs.

I hate these people.

I used to be able to do rapid design iteration, back when my entire web site had six pages and there were only four HTML tags to choose from (H1, P, A, and BLINK). These days: haha hah. If I get a site redesign out every four years or so I consider it a job well done.

My problem is not intent. I have ideas. I love playing around with software and design and technology. In the last three years — since the last iteration of this site — I’ve worked on a dozen or more revisions that never got done. On the contrary my issue is focus and follow-through. To get things done I need to cease playing, put butt in chair, and apply a consistent set of ideas to the entire site.

I managed to finish the site this time because I established the redesign as an actual project, with goals and a prioritized to-do list and a deadline (which. er. I didn’t make. still.) I had three major goals for this redesign:

• New and better CSS. The last time I did my site I didn’t understand a whole lot about CSS, and it showed. The styling was crude and buggy. I had worked around the simpler bugs over the years but there was a lot I didn’t think I could fix without a tear-down. It was time. I also wanted to spend a lot more time learning and understanding CSS in general, especially cross-browser issues. The current design still has some problems and inefficiencies but it’s a lot better. I have a lot I want to say here so I’ll put it into another post.
• Integrate my entire site into a content management system. My blog has always run under Movable Type. The core “www” part of my site — essays and fiction, primarily — was a whole pile of raw HTML files. This had turned out to be kind of unmaintainable. I wanted to put the entire site into a database-backed content management system, split it into modules, and structure it so that the next time did a redesign I could just write some CSS and hit “publish.” I looked into a couple packages for doing this including WordPress and Drupal but decided that I didn’t want to redesign the entire site AND learn CSS AND learn a new CMS at the same time so I stuck with Movable Type for the entire site. In retrospect: even with the pain it would probably have been a better idea to migrate to something else. I also have a Movable Type complaints post coming up.
• Upgrade the Movable Type software, redo all the templates, fix untold bugs, take advantage of features I had previously ignored (categories), and add new features (tags, new feeds, dynamic publishing). You’ll notice that there aren’t actually any categories or tags on the site right now. That’s because after three months of working on the site I got worried that I was working on the Windows Vista of web design — bloated, buggy, late — so I started dropping features off my todo list. Goal: ship it. I’ll add categories and tags in a dot release. (sooner than another four years, I hope).

Along the way of course there was going to be a new visual design. There were actually three new visual designs along the way because I can’t leave well enough alone (sigh). I did play around a lot with gradients and animated icons and more colorful backgrounds and schemes — more modern web 2.0 kinds of things. For a long time the site had a black background so the blue headings looked kind of neon. I even had ajaxy javascript-based menus that opened and closed once. But I’m not a very good visual designer and there was a little voice in my head telling me “but you’re a writer. its all about the content.” So I kept stripping things out. This design is kind of plain but its readable.

Now that I’m done with the design I can get back to writing. Yay.