This clock points out the ultimate irony of QR codes which is that they are a technological convenience that really isn't convenient. In order to read this clock, the user will require some sort of QR scanning device which is guaranteed to have a time-telling function built into it already.
So, that's why I made it. Hilarious, right?
See how the clock ticks in this video:
The clock has been designed and built by electrical engineer Michael Ciuffo (pronounced "Choof-oh"; thus "ch00f") who was inspired to do it by seeing so many, completely pointless QR codes on a four-day drive across the USA:
The real problem with QR codes is that they force people to use a machine to translate what could easily be human-readable information. Rather than a QR code, why not just show the text "Chevy.com" or even better just the word "Chevy" because everyone knows how to google stuff? Seriously, I've seen QR codes on highway billboard signs as if anyone is going to whip out their smartphone while they're pulling 65.
Despite his conviction that nobody uses QR codes (as corroborated on this website) Cuiffo set out to build an attractive self-contained clock. The challenge was given there are 1440 minutes in a day storing the number of QR codes needed would have required too much memory. Hence the solution of generating them in real time which required an intricate, although highly repetitive, circuit.
The obvious solution of using a microcontroller has its problems too in that it takes a lot of memory to store the QR codes. In this case an ATMega328 with 32K was more than enough to generate the QR codes on the fly, i.e. in software. Even so the coding to generate the codes wasn't easy.
His site has full details of the schematics and layout - and yes you have to agrree that the resulting board is beautiful. He also shares the problems and gotchas encountered during construction. So if you fancy trying your hand at a similar project his write up is very instructive.
Neural networks achieve great things, but it is slightly worrying that we don't really know how they work. Inceptionism is an attempt to make neural networks give up their secrets by showing us what t [ ... ]