Research in the area is definitely ongoing, but the article summary presents such a basic overview that it makes it sound like subthreshold as a basic premise is some kind of new idea.

Now a subthreshold FPGA, that's nice if the interconnect is kept at the same low voltage. But then it would take a *really* long time to communicate. Although I guess you can throw in some sense amplifiers. But if the interconnect is high voltage, then it remains the major source of power loss. Is your Prof John Lach? I can't get to the seeming paper on my IEEE student account.

A more important property of the brain is that it is not as precise as a computer. The brain, and many other biological computations, perform their calculations in an analog manner that usually gets them "close enough" to the right answer. Digital designers think they need every bit of precision in a 64-bit floating point computation and they engineer the circuit to require it--this involves a lot of "over engineering." Of course, the really cool thing is that biology has "digital" circuits as well, when it needs it!

I'm not really an expert, I've just attended a lecture by Prof. Sarpeshkar at MIT. He's built some cool DSP devices modeled after the human ear. He has a new (text)book, "Ultra Low Power Bioelectronics: Fundamentals, Biomedical Applications, and Bio-inspired Systems."

It's really fascinating stuff, and I think is where low-power computation is really going to go: into a hybrid analog-digital domain. Subthreshold is cute in SPICE, but process variation is a huge barrier to using it on any kind of microprocessor or SoC sized chip.

You would be closer to say PNP = -NPN. Especially if you carefully define the - operator.

Let's assume that you're not using Google's recursive DNS server (because you're obviously and rightfully afraid of them). Instead, say, you're using OpenDNS.

You want to go to www.google.com, but you need to resolve the domain name. You're request goes to OpenDNS. They get to see your IP. They always have. Then OpenDNS goes to google' authoritative DNS server to figure out the IP for their webserver. Under the proposal, the authoritative server would get to see some of your IP address, so okay, Google knows where you are, omg. But then you get the DNS query back and your web browser shows the Google homepage. OMG, their webserver just got your IP address again! So Google would know your full IP address anyways.

On the other hand you may want to go to www.cnn.com. Again OpenDNS gets your query and your IP. Under the proposal, the cnn.com nameserver would get to see some of your IP address when answering OpenDNS's query. But then again, cnn.com would get your full IP address later when you actually go to the site. ****And Google Would Know Nothing Of Your Visit To CNN, Even Under This Proposal**** baring CNN using Google analytics on their webpage, which they very well might, but this proposal has nothing to do with that.

Web sites already know where you're coming from. They have your IP address. Every single one of them, unless you're using a proxy. The problem is they can't easily redirect you to the server closest to you once you've already resolved their address. The only in the whole system who do not know your IP when you're browsing the web is potentially the authoritative DNS server; the usual case is the same people who run the authoritative DNS server also run the web server, so while they don't get your IP when you do the DNS lookup they will when you eventually land on the site.

The proposal says they would only use the first three octets. And users could just use a different DNS server if they had a restrictive servers that blacklisted Iran or whatever.

Certainly 17 is overkill. But (at least) 4 or 5 years of those are wasted by teaching to the lowest denominator.

My physics teacher used to talk about how the equations would change. del dot B would equal some measure of magnetic charge density rather than zero, while del x E would equal the partial of B wrt t + some measure of magnetic current density.

Basically the equations just become more symmetric; electric charge has monopoles after all. Certainly there will be a wide range of implications.