December 21, 2016
How does Santa pull it off? Using metrics, and analyzing what we can, could we test his tech?
Any sufficiently advanced technology is indistinguishable from magic. — Arthur C. Clarke’s Third Law
If you ever thought you were on a project with demanding work and unreasonable deadlines, I dare you to compare this to Santa’s tasks so that you’ll realize how insanely simple your work is by comparison. While I don’t claim to know the mysteries of how Santa accomplishes his business intelligence, transit and scheduling nightmares, I do know how to analyze them if we assume technology in the solution. After all, if any sufficiently advanced technology is indistinguishable from magic, perhaps therein lies Santa’s secrets.
The IT readers of this blog may believe they understand the history of computers having seen the movie “The Imitation Game” and heard about the UNIVAC vacuum tube days, then logically progressing forward. However, this totally ignores military and telecom involvement which independently created successful IT projects on an older and less public track. One can only assume the existence of yet another divergent IT technology workshop as a lab environment, a Technology Center of Excellence if you will, known as Santa’s workshop.
The naughty and nice list reminds me of giant databases. The NSA’s Utah Data Center (publicized by a man with “Snow” in his name) may have the ability to data mine various forms of communication across the U.S., but Santa has a method to determine naughty vs. nice across the entire planet.
Given the longevity of this project, I envision a legacy system, originating with vacuum tools and assembly code, before an eventual upgrade to COBOL. Hopefully a DevOps solution that folded in plenty of automation was applied early on to help with the Continuous Improvement that a project of this scope must require.
Non-magical data warehouse hurdles include updation of multiple fields including delivery address fields, name fields (due to adoptions), naughty/nice status (perhaps with some giant text fields with supporting evidence and/or attachments), individual present dropdown lists (even if GetsCoal is ticked), and the Boolean Do_they_believe field.
The fleet management concerns remind me of a worst-case Traveling Salesman Problem defined as delivery of toys to all of the good children all over the world in one night. How do you combinatorically determine the most efficient routing algorithm? Are any additional secret rules thrown in to simplify the task, the way that UPS reduced mileage, gas use and risk by never turning left? Are any smart devices – highly specialized IoT – involved to support the transportation process?
Of course, I may be overthinking this. When you have a magic reindeer delivery system, the expected non-magical boundary metrics may be easily surpassed. I should probably stop before I try to calculate average_present_weight x driver.findElements(nice_list) and determine the energy needs for the weight. Perhaps the magic, indistinguishable from technology, is where we should focus our attention, and the IT possibilities it may hold. Because high tech is where the magic happens.
What Holiday thoughts have you had which require IT and QA interpretation and calculation?