The difference between “testing” and “using” is pretty amazing. I spent 6 months working on my Good Nanny parenting app and I squashed every bug I found. Then I would delete the app and start again. Now this is a good way to kill bugs. It’s time consuming but you end up with a rock solid application. Again, and again I ran through this process. Finally, I thought I was pretty solid and submitted to apple.

I didn’t delete my application and used it as I meant it to be used. A log for events — in this case for my son on the days I was watching him. It worked great, but in actual usage I quickly learned that I needed to change some things:

1) The email log should be for all the events in a day, not just the last ten. It turns out that just the last ten items is just an afternoon. This defeats the purpose of the email log.

2) The email log doesn’t include the duration or the completion time. I just like to know when something happened; but my wife immediately asked how long (a nap) took.

3) Events need a milestone switch. Some of the events are really just a notice. This is a switch.

4) The extras are nice, but I really need a baby monitor. I was visiting my brother and we forgot the baby monitor. They exist in the app store, but they are relatively expensive. I already have a code base for recording and one for dialing. I should combine them together as another extra.

5) The lack of a start-up image really matters. Good Nanny starts up at the same speed as most other iphone applications (I’ve timed it). But since it doesn’t have a start-up/loading screen it seems like it takes a lot longer. I knew this, but I downplayed it mentally. A mistake.

6) I should finish the web service. It’s all present and I’m supposed to be pretty good on this web thing. Google Charts API would be a nice little add on. I’ll add it to a jQuery dialog so its all embedded in a single page.

I’m committed to this application so most of these things are already being added to the version 1.1. Its going to be better and better quickly.