Well yeah, that's kind of what the whole paper is about - the fact that they can analyze cache behavior to detect network and mouse activity on the system.

Is it the same hardware? This implies that it's not identical: The Ubuntu version of the Compute Stick has as a similar CPU -- if the CPU is different, how much of the rest of the chipset is different?

RTFA:

Because if you want to ensure your security and privacy, you shouldn't use the web.

If you understand the CPU architecture, any program that can control what happens within its address space can manipulate data moving in and out of the CPU cache.

It's still a valid complaint -- why give the Ubuntu device half the ram and 1/4 the storage? Even if Ubuntu *requires* less resources than Windows, the applications that people want to run may not. Chrome, in particular, seems to grow to consume all of my RAM whether I run it on my old 2GB laptop or my 16GB desktop. And the Windows device has 19GB of usable storage -- more than 3 times the total amount of storage on the Ubuntu stick, Ubuntu users store data too, especially on a device well suited to be a media player.

Getting the wired network to the laptop is not a problem -- most laptops in the office get to the wired network through the same cable they use to plug in to the monitor, but that wired network doesn't come for free, my company paid $50,000 to wire up cat-6 for an office that we only plan on being in for 2 years - and it already constraints where we can place desks. This doesn't include the $40 - $50K spent on access switches in the server room.

We have Wifi, which works well for phones, tablets, and laptops in conference rooms, but it's no substitute for the wired network since when the graphics guys are saving gigabytes of photoshop files to the file server, the rest of the network suffers.

That sentence was valid with "their": "Hell, the handlers dogs, they want to please their handler."

The same thing they do now -- use RF Wifi.

Everyone in my office plugs their laptop into their large monitor at their desk (which is why I said to put the receiver on top of the monitor, just as they now get their wired connection through the monitor), though a laptop may still be able to get good optical signal with a receiver built into the top of the display. Tablets and phones tend to have lower bandwidth needs than laptops and desktops (few people are editing uncompressed TIFF files on a tablet), so they could still use traditional Wifi (which will have a lot more bandwidth available when the heavy users are using Optical links). This technology wouldn't never replace RF Wifi (since it's a lot harder to ensure optical line of site than to ensure RF signal), but supplement it.

This is as close to IrDA as RS-232 is to ethernet.

This technology purportedly creates small one meter hot-zones of light, so instead of an AP having one (or a few) 5Ghz channels shared by everyone in range, it can have dozens of separate hotzones so each , and an AP in one room won't interfere with one in the next room.

I could see this being very useful in offices -- instead of spending tens of thousands of dollars pulling wire to each desk back to a central wiring closet, a few AP's can be hung on the ceiling with receivers on top of everyone's monitor with much better total throughput and less interference than RF.

Even if they used visible light, no one is going to see the flickering of a multi-megahertz flickering light.

Watering a 30x30 ft lawn with 1/3 inch of water is around 180 gallons, so if you're spreading 50 gallons over that lawn, you're only getting around 1/10th of an inch of water, barely enough to penetrate the soil.

Where do you live that you pay $0.20/gallon for water? In San Francisco, the water+sewer rate is closer to $0.02/gallon. The city with the highest cost for water + sewer in this chart is Atlanta, GA ataround $0.026 per gallon.

If you're refilling your 50 gallon barrel 2.5 to 4 times a month with rainwater, you probably don't need much water for irrigation anyway, sounds like you're already getting regular rain.

I don't see how you get so much use out of a single 50 gallon barrel -- a lawn needs around an inch of water per week, so a small 10 foot by 10 foot lawn is going to need around 62 gallons of water per week. Yet you manage to water your lawn, wash your cars, and bring up the water level of the pool with a single 50 gallon barrel.

You just suggested building rainwater collection and 6,000 gallon holding tanks into new construction, if consumers don't pay for it, who will? Putting the 24 ton tank on top of buildings sounds like it's going to drive up building costs substantially (especially in an earthquake zone). Seems like it would be better to bury the tank and pump water to where it's needed.

But what's the cost/benefit between small scale rainwater collection and large scale municipal water collection? For most people, water only costs a penny or two per gallon (including sewage costs), is it really worth $60 - $100/year to build a 6000 gallon water collection system into your home? Greywater recycling might be more affordable since it doesn't need a huge tank.

I don't know if such a system is workable -- each acre would collect 27,000 gallons of water from one inch of rain, that's a lot of water to store. So collecting 10 inches of winter rain to use for summer irrigation is going to take a 270,000 gallon tank for each acre (which is a 100x100x4' tank)

Plus, preventing rainwater from seeping into the ground is just going to increase the amount of water that needs to be added through irrigation.

In any case, the answer for why it's not done is "It's expensive -- more expensive than the subsidized water" it would replace.

Nearly $1M per home sounds like a lot even by Marin standards, assuming that the cost of land is not included in that $200M figure.