Of Crunch and Codeblock

I am archiving older pieces I have written on other sites, making this the definitive home for all my work. This is one of several I am porting over from my GameDev.Net user journal. Enjoy!

Code block - you know, when you sit at the keyboard, IDE open, mind primed and ready to get to work. Code block is a myth. Code block is evidence of insufficient upfront design work. If your design has gone from general guidelines to detailed technical specifics, you won't have code block. The code will almost write itself.

Crunch, on the other hand, unfortunately is not is myth. I'm in crunch mode on TickerWatch. This coming Friday, August 20, is my last day at this job. I need to deliver the application in a usable state on that day.

I'm not being paid for TickerWatch, and I probably will never use it after I leave the company on Friday, but it's partly a calling card of sorts (employee A: "Wow, where did you guys get this amazing tool?" employee B: "Remember that tall, gorgeous black guy who worked here last summer? He wrote it in his spare time !") and partly to help me regain the discipline necessary to deliver applications.

After TickerWatch I take one week off, and then I start another project, this one to last 11 months! A lot of reading and learning will be involved with that project, though, so it won't be quite the same. I'm even considering a second project as a stress reliever from the first, and that second project will be a game, so I'll have something pertinent to yak about here.

On Data Design

While I did sufficient upfront design of the user interface and user interaction model, I didn't quite do enough data modeling for TickerWatch. Consequently, I've been writing some seat-of-my-pants code to represent internal data which I'll probably have to go back and refactor later.

Famous last words.

If anyone happens to know some interesting affordable data modeling tools (I can't afford Rational Rose, okay?), please drop a comment.

Back to the code mines...