Cameras in the stands what could possibly go wrong?

For several months I have been checking on the on the official LHC schedule:

https://espace.cern.ch/be-dep/...

and had been wondering about the notation "leap second" on the Wednesday of week 27 of 2015. This posting now makes me understand what is going on. I can imagine lots of consequences for the both the machine (think of how far in the lab frame a 7.5 TeV proton goes in 1 s) and the enormous (at least enormous if you aren't Amazon, Apple, Google, or Microsoft) pile of HTC servers on 6 continents we use to reduce the data into Nobel prizes (hopefully prizes plural...).

Fred

The first computer I used was DEC PDP-8E. It had 12k of 12 bit core. My high school had bought it for $40,000 in approximately 1974. It I/O system involved punched paper tape and teletypes. I guess this will not be the oldest system someone mentions but it's pretty old school. And technically I didn't own it. The first computer I owned was a Gateway first generation Pentium and that machine had 2 MB if I remember correctly.

I marked the picometers entry but I was really using it to mean picoseconds (since the speed of light in a given medium is a constant the two are equivalent). I work on the TRT a subdetector of the ATLAS experiment at LHC. We are definitely sensitive to changes in the timing base for the time measurement made by the TRT when the LHC machine timing signal changes by ~100 ps. I think that a crazy good level of accuracy but than one of my friends pointed out that 100 Gbps network that we are now using to move around the ATLAS data has digits that are ~10 ps wide which implies a a few ps rise time for those digits in a 100 Gbps data stream. I don't about other slashdotters but having struggled to get ~100 ps timing stability, I find 10 ps wide digits with a few ps rise/fall time to be mind-blowing...