Prehacked

I got something working end-to-end for the first time today - a whole chain of grabbing frames from the camera, doing some image processing, and displaying the result on the screen.  Very satisfying!

Sometimes it gets frustrating to only have an hour or so each day to work on side projects, but I have to admit I am managing to make good progress!  I got a bit of a speedup, and am now grabbing and decoding frames from the Android camera at about 17fps, which is more than enough for what I want.

Second day in, Android development is surprisingly good.  Everything is much much easier than on the iPhone - I'm really happy with my choice to develop here first, and only on iOS once something has been proven.  I'm working with the camera, and already have it grabbing and decoding frames at a rate that I think is nearly acceptable for what I want to build (around 10fps on my LG Optimus).

I started a new project today - my first Android app!  So far it's just a "Hello World", but it was nice and easy to get up-and-running (much easier than iPhone development was).

Zoe is a few days past 9 months old.  She's spent more time outside of the womb than inside it!  It already seems to me like years have passed.

I've started working on something new:  A mobile app that helps you figure out what's causing your migraines, and how to prevent them.

It pings you regularly with short questionnaires about your recent migraine incidence and lifestyle habits, and learns over time to identify your triggers.  It gives you analytics to correlate triggers with headache occurrence and severity, and makes suggestions on how to change habits to stay symptom-free.

Eventually we're hoping to tackle all chronic conditions, but we're going to start with migraines, and then allergies.  We hope one day our app will be discovering new causes and factors, and processing data from enough people that we can give something back to the medical community and make people's lives better.

If any of this sounds interesting to you, please register here and I'll let you know when the app is ready.

I just wrapped-up my Flash pitch-detection work into an open sourced library you can use in your own web-apps:  Timber!

It's been a while since my last post!  I managed to write about my progress here every day for more than 140 consecutive days though, that's not too bad.

My wife and I have started working on a new side-project, so I'm shelving Tunologist for now at least.  I'm proud of what I did - realtime pitch detection in Flash isn't a simple thing to get working.  But I'm more excited about this new project than I have been about any other idea for a long time.

We're getting close to having a working beta of this new idea.  If you're the kind of person who loves to try out new things, send me an email (doctorbill at gmail dot com), and I'll get you access to the beta when we're ready.

I came up with an easy fix to the problem of being able to cheat by toggling timed play.  Now any time you switch that setting on or off, the game restarts - so in effect, a single game is always either timed or not timed.

I just finished the control for disabling the countdown timer.  It works fine, but I'm not totally happy yet - right now you can switch the timer on and off whenever you like, which lets you arbitrarily cheat at a timed game.  I'll have to figure out the right way to make that impossible.

I'm nearly done putting together a new mode where you can play without the countdown timer - just a few more bits and pieces to finish off, it should be done tomorrow.