It's clear to me that plasmas give better quality image but I still choose LCD. The plasma issue of burn-in is the main worry but they're also more power hungry and heavy too. Plasmas easily beat LCDs for black levels, colour accuracy, response time and viewing angles but LCDs are good enough. Even if my kids didn't spend hours playing video games I know somehow there would be burn-in and then I'd want to buy a new set ... which is just a waste. Plasma being the losing technology is not all down to marketing.

The fundamental engineering problem here is that cars float and submarines sink. Ballasting that car with enough weight so it's close to neutrally buoyant will ensure it performs nothing like a sports car on the road. This is the kind of issue that made lead acid batteries such a great choice for submarines in the first place.

The best approach is going to involve minimising the volume where water is excluded, i.e., ensuring that as much of the vehicle is flooded by water as possible when it dives. At least, as a sports car, the interior is very small so they may have a chance of making it work.

Just because a file is shared with Dropbox doesn't mean that accesses involve a network round-trip to their servers. The files are still stored locally (on an SSD if you have one) and only synchronised when a change is made on another machine. Dropbox is not the same as a Windows "network drive" over SMB/CIFS or Linux NFS.

Apparently the trick in progress here is that people already gave their email to someone else, namely their service provider. The legal logic is that this borks their expectation of privacy, see http://en.wikipedia.org/wiki/Katz_v._United_States from 1967. One might hope SOTUS will revisit their decision in the light of the current state of technology but until they do you're stuck relying on legislative protect rather than constitutional.

Relevant: http://articles.latimes.com/1989-10-05/news/mn-913_1_soviet-union. Those who don't study history are condemned to repeat it.

For copyright purposes both physical and electronic mail is the property of the sender or, more precisely, the author.

Freefall strictly speaking means 9.8m/s/s which, after 228 seconds, multiplies out to 5000mph. That's an order of magnitude more than Baumgartner's speed. Wikipedia explains:

"The example of a falling skydiver who has not yet deployed a parachute is not considered free fall from a physics perspective, since they experience a drag force which equals their weight once they have achieved terminal velocity (see below). However, the term "free fall skydiving" is commonly used to describe this case in everyday speech, and in the skydiving community."

Still, terminal velocity for a human at sea level is about 120mph which is 4.5 times slower than the quoted 536mph. Taking the square root gives an atmospheric pressure 2.1 times less than normal which translates to him popping the 'chute at about 25,000. Actually he had a pressure suit which would probably slow him down so it could have been higher than that.

Since we can predict the next (2013) close approach very accurately we're very confident it will be a miss. Therefore that approach doesn't rate a mention in the table.

The trouble comes in that, while we know the 2013 approach distance will be greater than 0km from the surface (>6400km from the centre) there's still some uncertainty. The earth is massive and the close approach will cause a relatively large change in the orbit of DA14. The size of the change is inversely proportional to the square of the approach distance. Thus even a small uncertainty for 2013 results in a large uncertainty for subsequent approaches. Celestial billiards at work.

In my view an important property of any ballot is that the great majority of people must be able to understand the whole process. That's the only way for people to have confidence that there's a reasonable chance of detecting and preventing rigging. It also rules out pretty well any form of electronic voting. Internet security involves very serious maths that very few people can handle.

Around here we still write numbers in squares on pieces of paper and drop them in the ballot box. It works. The cost is tiny compared to the cost of government. I just can't see the advantages of more automation being worth the risk.

People might think it weird that an IT guy would have this luddite view but I think, on the contrary, I'm better placed than most to know what could go wrong.

You do realise they do course corrections? Yes, the maths they do is neat, especially when they need to take relativistic effects into account. There's a limit to how accurately the spacecraft can set its course so they plan to fine-tune multiple times during a mission.

It doesn't sound like this falls into the offshoring category to me. Since the military is involved I guess they demanded the source to assure themselves that there were no backdoors. It doesn't seem an unreasonable step for any government (even/especially in the US) to take before using your software in a security context.

The fun is in considering what recourse Symantec has. If they didn't have some really expensive penalty clause in the non-dislosure agreement that will have been involved here they'll be kicking themselves right now. They'll also be wishing they gave themselves some way to identify the source of the leak. Their smart move would have been to insert some minor changes, e.g., to indentation or comments, to make each version released to third parties unique and therefore traceable.

I propose that, for the people to trust their democracy, they must be able to understand all aspects of the voting system. This rules out pretty well all automated systems, especially computers with cryptography and hashes. Just go back to people writing on paper and ballot boxes.

Sure counting the ballots by hand is expensive but it's tiny compared to the cost of travel and time for the voters. The risk of serious, undetected fixing of results can't be eliminated with automated systems.