The cloud is a revolution in billing. Not only do you sidestep all the HR/legal faff, you can bill the hosting cost to the department that uses the application instead of IT, it's impossible to price match competitors since you don't know how much it's going to cost until after you start the app. If someone writes crap code it auto spins up more instances and the hosting bill goes up, rather than someone having to write a document justifying additional expenditure.

You seem to have forgotten the birth certificate requirement for passport applications.

Unless your site catches fire and the fire department insist you cut the power before they'll enter as happened to ThePlanet.

http://www.datacenterknowledge.com/archives/2008/06/01/explosion-at-the-planet-causes-major-outage/

"The fire department is not allowing the company to run backup generators, so the facility has been without power since the incident occurred."

Adequate fire protection doesn't help that much if the electrical room explodes with enough force to remove three walls.

I have a Core 2 Duo based router, it has 100 peers with 8 full routing tables. It's currently passing around 300,000 packets per second with, about a gigabit of traffic at 10% CPU load.

That's within one order of magnitude of the Vyatta result, on much slower hardware. It's within one order of magnitude of the 3mpps performance you quote is '4 orders of magnitude greater'.

By the way, the Core 2 Duo had 4MB of cache, Nehalem comes with 8MB. The nature of cache is such you don't have to get your entire data set in it, if memory access is correlated it will speed up access. You'd get correlated requests in the unlikely scenario that a tcp packet destined for a given IP address was followed shortly afterwards by another one going to the same IP address.

http://www.vyatta.com/about/press_releases.php?id=78

20Gbps and 3,000,000 pps on commodity hardware with CPU to spare.

The parent is clearly unaware of interrupt coalescing or PCI-E.

When the upside-down-ternet was slashdotted it originally had more images in the page, but the 100Mbit upstream on the server filled up, so I took some out.

We have custom built rack shelves for Mac Minis that neatly hold the power supplies and minis.

http://www.mythic-beasts.com/macminicolo.html

14 minis in 5U including the power supplies. You need to allow some rackspace for the masterswitches and switches too.

Recording at -20dB isn't 100% FAIL. Suppose you're recording outdoors and it's windy. The difference between the ambient noise floor of the wind (say 40dB) and the spoken voice (say 90dB) gives you 50dB of dynamic range. Setting it to peak at -20dB puts the wind noise at -70dB, compared to the digital noise floor of -90dB (16bit) or -120dB (24bit). That's a perfectly sensible optimisation which gives a huge safety margin on clipping the signal whilst having the digital noise floor so low it doesn't matter. Remember when setting up you have a much better estimate of the minimum background noise level than the maximum peak - e.g. should your interviewee suddenly shout and you're recording with peaks a 0dB in normal speaking you're going to look very stupid.

In general a professionally mastered recording should peak at 0dB to use the full dynamic range, but sometimes computers are used with original content before it's been mastered.

So electric cars with a 500 mile range are problematic if

1: You live somewhere very isolated with very unreliable electricity supply and life threatening weather

2: When you plot a 500 mile journey you're completely unable to go past a quick charge point to compensate for (1)

I'd suggest that when you go to Heaven, you don't have a conversation with Mr Darwin and his survival of the fittest theory.

Here's a potential solution.

At present we drive a big lorry full of petrol from the refinery to the petrol station every so often?

How about we drive a big lorry full of batteries from the power station every so often and take the old batteries back?

Or how about we put a great big power lead into the fuel station. The sort of lead they use for hospitals, data centres, aluminimum factories etc.

You only need to supply replacement batteries to people who are making >500 miles journeys without an overnight stop so the number of people stopping and drawing energy from the fuel station is comparatively low. In the UK you could get away with less than a dozen stations in total on the major motorways (M5,M6,M1,A1M,M62,M25,M4, somewhere in Scotland) and you'd cover almost every non pathological 500mile journey you could do. Really, that's replacing the entire petrol station network in the UK with less than 12 battery stations.

Actually that's probably overkill. Starting from London, Aberdeen and Inverness are the only cities that's far enough away to need a recharge (535miles/560miles). It's entirely possible that the number of people needing a battery swop on the way would be so low it would be uneconomic to make the batteries swoppable or to fund a battery swop station.

Besides, in the UK you're strongly recommended to stop for 30 minutes after every three hours driving. At most that's 210miles of energy every 30 minutes, that's a delivery rate of 180kW to the car while it charges. That's difficult (in particular getting the battery to accept the power without overheating) but may not be impossible.

This is a very silly line of argument in being neither correct nor generally applicable.

It is legal to cycle on roads where I (and every poster other than you) live.

The roads were designed for roman legions, pedestrians and horses. Bicycles came later. Cars came even later.