carpeaqua

In my last post I admitted that I love to ship software. In fact, in the past few months I’ve shipped 3 new iOS products and several updates for each one. Part of this is because I was coming out of my self imposed iOS boycott and had too many ideas I wanted to get them out of my head. The other part, however, was because the middle of 2010 was filled with many false starts, experiments and mindless refreshing of my Twitter feed.

I am a habitual project starter. If you took a peek in my Elements, you’d see a folder of “App Ideas” that has dozens of ideas for applications I’d love to pursue someday. Getting ideas out of my head and into my laptop or iPhone has long been my digital equivalent to jotting down an idea on a napkin.

Once the idea gets out of my head and into Elements, I usually start sketching out user interface ideas to envision how it works. I’ve got notebooks filled with app idea sketches and notes.

In some cases, I even start working on these projects to assess their financial or technical feasibility. More often than not, the projects fail.¹ That’s perfectly acceptable to me, because with each one of the failed projects, I learn something about a new technology or am able to extract code I wrote for it to use in another application.

I liken my explorations to sampling a book on the Kindle. When you get a sample from Amazon, it gives you just enough of the beginning of a book to decide if you want to outright purchase it. Once I get around 10-20% of the way through a project, I can determine how feasible or interesting it is for me to pursue.

There’s a subset of the technology sector who thinks that ‘million dollar ideas’ are hard to come by and that every idea you share needs to be prefixed with a lengthy NDA. Not me.

I went through my ‘Development’ folder the other day to clean out the failed projects of the past few years and thought it’d be an interesting exercise to share some notes about each one:

AudioNinja: As part of my podcasting and radio hobby, I use the now defunct AudialHub every week to convert audio from MOV, MP3 or AAC into WAV. AudialHub does a great job of it, but since it’s no longer being developed I started pursing building my own replacement. I ended up shelving the project about 20% through because dealing with ffmpeg and the variety of different codecs proved to be too much of a headache.

VideoNinja: What AudialHub was to audio, VideoHub was to video. Likewise, VideoNinja was a subset of the AudioNinja code. I was designing the UI so that it could share quite a bit of code between both projects. Again, way too many headaches for me to pursue further.

Vida: This was by far the oldest project in my box. Back around the time I was building PocketTweets, Robert Andersen and I started looking into building an application that would track everything you do on your Mac so you could determine trends in your computing usage: how long you spend in Safari, on email or refreshing your Twitter feed. Robert took a job at Apple and I got bored with it, so it went to the deadpool. Since then, a few similar apps have popped up, though I’m not sure how successful they are.

Soapbox: Soapbox was by far the latest I’ve ever scrapped a project. I’ve never found what I consider to be the perfect Twitter client for the iPhone. I wanted something with the style of the now defunct Birdfeed, a few features of Tweetie and (most importantly) support for filtering your tweets based on keywords or their source (Foursquare, Gowalla). I spent a good chunk of the late winter and early spring of this year getting the application into a beta state. When Twitter bought out Tweetie, I made the difficult decision to scrap the entire project. Going into the iPhone Twitter market was a bad idea before, but with Twitter now offering their own clients, it’s a suicide mission. I did end up open sourcing the code on Github in hopes that it’d inspire Twitter developers to add the filtering features into their own applications. I continue to hold out hope.

Confessions: I went through a phase where I was really interested in things like Postsecret, messages in a bottle and similar sites. Confessions was an iPhone application that would take your current geolocation and show you the anonymous ‘confessions’ of other users of the service within a given proximity to you. You could also post your own anonymous confessions that others could read all while never knowing who exactly posted it. I ended up scrapping the project because I began going down the rabbit hole of what awful things people might post on the service and got paranoid about liability.