The first job I ever automated was my own
In 2007 I was a space console operator with the 7th Space Warning Squadron at Beale AFB — missile early warning and space surveillance. Part of the job was being ready for upcoming satellite passes, and the way we stayed ready was five-plus little push-button alarms, each one punched in for a different pass. When one went off, you figured out which one it was and what it meant. It worked, in the way that a row of beeping alarms technically works.
I got a little over-eager and set about automating my entire job.
Five timers, one screen
On my own time I wrote a C# app I called the Space Console Accelerator. One screen: five satellites, each on its own row, laid out across the next ten hours. A clock thread ticked every second and kept the board scrolled to the current Zulu time, with the day handled in Julian days the way the ops floor actually thinks about time. Buttons to jump back or ahead an hour, or ten. A dialog to add a bird, a dialog to remove one. Instead of five timers going off at random and making you do the detective work, one glance told you what was coming and when.
One detail I’m still glad I got right: the copy I kept runs entirely on made-up data. The app generates fake pass times and fake statuses so the workflow could be built, tested, and shown off without a single real number in it. The tool was mine; the real schedule stayed on the ops floor where it belongs.
The demo the brass liked
I pitched it to every officer an E-4 could get to sit still for a demo. They were genuinely impressed. And it still went nowhere — because getting software adopted on a military ops floor takes acceptance, integration, and a process, and an airman with a working demo is only the first step of a long walk nobody was in a hurry to take. I was a bit ahead of my time when it came to automating tedious jobs in the military. It was fun to work on, and it is pretty darn cool if I do say so myself. I wish I had been allowed to keep working on it.
What it taught me
I have spent the years since automating other people’s tedious jobs: cattle operations, security dispatch, paint retail. The lesson from 2007 held every single time — a good tool is not the finish line. Adoption is. The software has to meet the people and the process where they are, or it ends up a cool demo on somebody’s hard drive. Mine sat on that hard drive for almost twenty years, file dates still reading July 2007.
Not sure anyone has a use for an app that tells them where a satellite is and will be at any time overhead — but maybe I can make a cool little Python tool based on it someday and relive my old space console operator days.
I'm Joseph Dattilo — engineer-founder in Lansing, Michigan, author of the FleetHarbor suite, and founder of Date Palm Media. More about me · More writing · Get in touch