Phoenix is not the best place for kite flying. If we happen to get a breeze at all, it's often too light or too gusty to support kites. I've been trying to catch the perfect wind to fly my little dual line delta kite, but I keep missing the best hours of the day. I decided to build a little alert system that can help me recognize when the best conditions are present.

A few days ago I managed take note of the wind speed that I needed to launch. Somewhere around 14mph is the minimum for my kite, anything over 24mph is probably too harsh. I looked up those wind speeds on the closest private weather station to my apartment using wunderground.com. Turns out that Weather Underground also provides xml feeds for each of those stations.

Using C# I built a little application that downloads the latest weather xml and parses out the wind speed plus a few other details. I included options to set the wind's optimum and minimum velocities as well as the refresh rate. The current wind speed is compared to the velocity settings and scaled to a percentage value. 0% means you have no chance of getting off of the ground, 100% means the current wind matches or exceeds your optimum. Now I needed a way to show off the data.

A notification pop-up or ding on my computer would have been sufficient, but I wanted something more symbolic. I hooked up a small servo to a BS2 and added code to my program to output that percentage value over serial every time it refreshes. The basic stamp watches for an update and moves the servo according to the value. Mounted on the end of the servo is a wire with a little paper kite its end. As the the wind approaches the optimum speed the kite rises into the air. Pretty symbolic huh.

Now go fly a kite!

P.S. I'll let you know when and where the application and example microcontroller firmware is available for download just as soon as I work out some bugs.

I've been using a service called Dropbox for the last couple months to help keep the files on my netbook in sync with those on my desktop. I tend to do most of my work while I'm at home but it's always nice to be able to travel around with that data. Thumbdrives used to be the answer, but they're a little hard to keep track of and keyring versions get too bulky. Not to mention you have to think ahead about the files you might be needing. Dropbox uses a cross platform application to continuously synchronize files within a folder on each computer. Any file you place in that folder will be synchronized with any other system you use the service on. Plus there's a web interface for access on systems that don't have the app installed. It also keeps backups of versions and deleted files for up to 30days using the free account. Oh yeah, and you get 2GB of storage for free.

I'll be posting a full gallery of photos from the trip soon and expect to see some interesting extras.  I've been working hard over the past week to deal with the huge collection of GPS data I logged over the three weeks I was in the UK.  My AMOD AGL3080 GPS logger was running (nearly) every day of the trip and collected over 550MB of nema text logs.  At the moment I can't find a program capable of managing all of that GPS data successfully.  I'd like to extract some elevation profiles, walking mileage, speeds, etc.

While I continue searching for a tool to do that, I've been working on my own piece of software.  Using Jeffery Friedl's Geoencoding plugin for Lightroom, I tagged all of my images with the GPS coordinates of the location where they were captured.  Over 1900 images now have location information embedded in their metadata.  The next step was pretty obvious, drop all of those photos into Google Earth and view them on the globe.  There are a few paid software packages out there that can handle the generation of a Google Earth kml or kmz file, but none of them supported the kind of customization that I was looking for.  I laid out the plans for my own program and started working.  I'm nearly finished with the first version and so far it's working great.  There are a few bugs that need to be squashed, and maybe some improvements to the overall functionality of the program, but all-in-all I'm pretty proud of my work.  Expect to see my Google Earth kmz posted here in the next few days.