Last week I had the enormous pleasure of attending a meetup in London to chat with some other iOS developers. My motivation was largely to get some early reactions to ther CardBoard app. One of the most helpful people I ran into was a UX specialist called Don, who is known somewhat ominously as the ‘code breaker’.
He quickly grasped how the application works and then he did something I’ve never seen before. He started stress testing the app by working it far harder than I’ve done to date. Rapidly adding dozens of sections, deleting cards, adding boards, adding cards to boards and so on with stunning rapidity. In some ways it was like watching a deranged toddler running amok in a playroom.
It’s a style of testing that ideally suits a mobile device. Interacting with a desktop app is intermediated via a keyboard and mouse, and web apps sometimes have a built in delay as pages get downloaded from a remote server. It’s also a style of testing that proved to be highly effective – three bugs found, two of which were so minor that I was able to fix them on the way home, one slightly harder that I fixed this weekend. With any luck I will bump into Don the Code breaker again and see if there’s anything else he can ferret out. If CardBoard can survive him, I expect it can cope with most people.
The whole experience with Don reminded me of a comment a collegue made about a web application I was involved with over a decade ago. He said that although positive comments give you a warm fuzzy feeling – it’s actually the negative comments that help. Telling me that flipping cards over is fun is nice, but telling me that the settings card background has mysteriously gone black is helpful.
Of course it also makes me feel like a complete fool, but ‘why stop now ?’ I always say.
I have a couple of minor issues that I hope to sort out prior to giving this to the first of the alpha testers. All but one is me simply wanting to tidy up architecture and tidy up rough edges that worry me – but are invisible to users. The visible issue relates to devices dropping off a network and it’s simply a piece of boilerplate code that hasn’t been copied over from the prototypes.
I’m gritting my teeth again this week and putting these in prior to adding beta features – which are the ones that make this a really useful app.