The Raspberry Pi will not replace a regular Windows machine- but I find it hard to think of any easier way to have something that also allows the user to have the freedom to poke around, modify files, even destroy the filesystem, and the system can be restored in a few minutes. SD cards are cheap enough that part of the lab setup costs could include a huge number of cards that each user could take their setup with them in their pocket.

An Android powered stick is a great choice for someone who already knows Linux, but the learning curve is way too steep for an educational environment. The Android stick doesn't have anywhere near the install base (throwing in Android phones/tablets is a red herring) or the documentation in a hacker sense- how many well supported Allwinner A10 Android distributions can you just write to a SD card with virtually any windows machine with a SD card reader? I'm not saying RPi is perfect, just that it would be a good alternative in an educational environment- far, far easier to re-write a SD card than to do even a Live-CD based install.

I was thinking about this- in the long term, you really want to encourage people to play around with things rather than run a static collection of applications. If people want to just browse the net and search, Raspberry Pi is a poor choice, but for learning about programming- the Raspberry Pi is great because if things get messed up, just flash a card with the reference image- system is re-imaged. Cards are cheap enough that you could hand them out to people and they would end up keeping their entire setup with them.

If I'm at a counter and the person behind the counter is just reading things off a screen to me, what's the point? If the person adds nothing to the transaction, what I really should be talking to them about is what they are going to do after their job is eliminated.

Yep, gigabit to the home would be cool, and I would score massive geek points, but in terms of an individual user, what use would it be? A big pipe makes a lot of sense when you're aggregating traffic from a bunch of different sites, but a normal residential customer (torrents aside) is going to be pulling most of their bandwidth from a small number of sites at any one time. Of course, this is for the near term, and I would expect that we are a pretty long way from putting a 2 way gigabit connection to use.

On the other hand, I expect that TWC already has plenty of experience in delivering one way multi-gigabit bandwidth- digital television.

Since SD cards are the standard storage medium for the Raspberry Pi, what about a speed comparison for that?

On all parts I've seen, if the datasheet says 100K PE cycles minimum, is at the maximum rated temperature for the part. The physics of flash works out that the worst case is when hot. Without a temperature bound, the number of PE cycles is meaningless.

I'm not sure what you're saying is badly wrong- 40C is pretty cool for most laptops. If your drives are 31-34C, you have *very* good cooling available. a temperature sensor on the front of the hard drive (if the ambient air is coming from the front) only tells you the lowest possible temperature the rest of the drive could be. On most servers I've seen in data centers, internal temperatures in the 30's would be an indication of a very light load.

The temperature dependence is a very strong factor that does seem to be missing from the analysis- to add to what the AC parent said, my experience is that the minimum number of erase cycles is when the device is at maximum temperature, take it down to room temperature, and the typical number of erase cycles goes up by an order of magnitude. Most computers have an internal temperature of over 40C when run in a normal environment,

Your drive will fail, SSD or HD. You must be prepared for that.

Why does it matter at all? It doesn't matter the relative processing power of my phone, there's a hell of a lot more to it than processing power. Like the simple fact that my phone isn't on a different planet.

What I don't see in TFA is something that describes how the one big hurdle of this type of design will be overcome- the IMMENSE costs! The speeds that processors and RAM runs is so high you can't just drop it down on a board and expect it to work- you're in a long loop of simulate, build, test, repeat, and each iteration is extremely costly- we're not talking Arduino here- in reality, these boards (populated) in mass production will cost hundreds in just BOM costs, not counting the assembly. If the biggies in the industry are truly willing to foot the bill, great, but no matter what, these boards will still remain expensive, and likely still in the NDA wasteland as the individual parts that make it up are unlikely open up their documentation (or distribution chains) any time soon (yes, Broadcom, I'm talking about you).

With the talk of stalkers and such, what is the maximum range we can read the badges? Passive RFID has to power the badge from the reader- at long range at the frequencies that these run, you're going to need a powerful transmitter and a big antenna, which makes it pretty hard for the stalker to hide at least. I do understand that advancements are being made all the time, but we still can't break the physics and basic information theory with respect to power and signal-to-noise.