Both behaviors are documented and "if not midnight" adds a fair bit of complexity, probably a good idea to know exactly what happens when you write that.
In 2007 a Czech art group hacked a webcam and incorporated a nuclear explosion at the location it monitored. It later got showed on a weather news segment.
Check out Open Refine. Has a feature that clusters similar strings and unifies. I remember last time I looked at this data set... 4 letter acronyms spelled 12 different ways, it's unbelievably messy.
Taleo makes a great job making it as painful as possible to search and apply for jobs. I find it pretty funny a "hack" of how to look through taleo listings is in an article about unadvertised jobs.
He is apparently involved with Intellectual Ventures, so while probably not directly involved in killing the reform, this isn't the most noble of positions.
I think it means "#1 son". Bender and Flexo follow the same numbering pattern. Makes sense if you assume all bending units' serial numbers are expressible as sums of two cubes.
I've been working on another one of those apps. It's a real pain in the ass to get accurate location. Both NER or regex location extraction and geocoding services can be slow and inaccurate in variety of ways. The way that brought me the most success is regex and a tabulated file with city names, coordinates and some other info. In either case, there's manual labor involved in finding false positives and negatives.
Coursera has a class as well. Martin Odersky tweeted a while ago about a possible advanced course in the fall but it doesn't look like it's happening yet.
[1] https://docs.python.org/2/library/stdtypes.html#truth-value-...
[2] https://docs.python.org/2/library/datetime.html#time-objects