Even though we’ve been heads down fixing bugs and getting ready for the next Vista beta (yes, my work-life balance is somewhat out of skew lately), I’ve found that I’ve been spending a lot of time just thinking to myself about applications with regards to how people use them. Even with tons of usability studies and personas, I still believe that most software developers just don’t understand users.

One notion we’ve been using in Windows Calendar is that people don’t like a variety of dialogs popping in their face to get a simple task done. The most common activity when using calendar software is the creation of new appointments, so everything you need to do that task is directly in front of you.

Another thing that’s been bothering me for awhile is the consistency vs. experience issue. I’ve mentioned it before, but what do you do (as a software engineer) when the common controls just don’t cut the mustard? Or even worse, what if the common control doesn’t even exist (have you seen how many color picker "buttons" are in various windows apps?) Do you just live with it, emulate another design (see Ribbon Control) or do you reinvent the wheel? It’s a hard problem, and something that I personally grapple with frequently.

I wonder if most software suffers from "death by risk-aversion"? Perhaps it’s as simple as Dori says, we should do something really interesting and break something.

Most software lately just seems to subscribe to the notion of Furrygoat’s Law. This is of course, because everyone loves a pony.