I am fed up with my Garmin GPS units. Not so much with their hardware, but with their closed software. Some people have cracked how to make maps for Garmin a while ago and turning open data into open maps was good fun for a while. But so many things remained difficult, impossible or just a pain in the back. The new units have plenty of space and reasonable processing power, so why not be able to have satellite images on the units for places other than the US? Why not better ways of managing tracklogs? The pains have become just too much.
So I am embarking on turning a Android phone into a GPS unit. I have done enough programming to know that my programs will just be as buggy as those of others, but for once I will have the option of fixing all my code. My GPS will not be as feature rich but they will be the features that I use and want. Plus I think Garmin's software developers have somehow stopped innovating. My Colorado has a better screen and a better chip than the old etrex, but that is hardware. The software is certainly no improvement.
How difficult will it be? That is the big question. A lot of stuff is already built into Android, but a lot of stuff is missing. This will be real-life experiment on how mature Android is, how difficult it is to use and how good it can become.
There are other platforms out there, winCE, openmoko, the iphone...so why Android? Well, Google has done pretty good software so far and this is something a lot of money rides on for them. So I hope they will put a lot of effort into it and they have made it open source. And it will be available (in due course) on many devices. So this is a good time to start. Let's get rolling.