No he did not replicate an $86 million project in 57 lines of code
When an experiment with existing open source technology actually makes a good prototype for a multi-million dollar enterprise application
This morning Medium update brings me this very provocative title : How I replicated an $86 million project in 57 lines of code by Tait Brown.
Its a very interesting reading. Tait definitely did some cool development and tests. The result is, to say the least, very impressive.
What he developed doesn’t include a 24x7 national-wide support infrastructure.
What he developed isn’t packaged in a hardened, impact resistant, easily movable and mountable casing…
What he developed isn’t connected to a back-end server that keeps the information stored, and triggers alerts by taping into the relevant law enforcement systems to inform the officers in the car if the current driver could be an armed and dangerous suspect, or if the car is “just stolen”…
What he developed isn’t “57 lines of code”. It leverages hundreds of thousands of lines of Raspian Linux, multiple libraries for image capture, optical character recognition, storage, network connectivity, display… stuff which needs to be developed for the embedded solution using probably much more rugged code that most likely has to be custom to the embedded chipsets used…
The openalpr library used totals actually, just itself, over 300 files in multiple languages including XML, Java, C, C++, YAML… and over 62000 lines of code. (Using the counter of lines of code script described here.)
What he developed?
What he developed is the (very cool) functional equivalent of a prototype for the $86 million project… in 57 lines of code, leveraging off-the-shelf software components.
And that’s OK. The team who did the $86 million project most likely did something similar when they were prototyping their project.
Don’t confuse a hobbyist development, a prototype (which can be very close to a hobbyist project — some skunkworks teams work that way), and a full scale enterprise project.