I have a dumb phone, but I only need to recharge it about once per week to once every ten days. It doesn't weigh down my pocket, and It works really well as a phone, too. For everything else, I use real computers.

One of the best firmwares out there is Tomato, and its various forks like TomatoUSB. I am running TomatoUSB on my Linksys E2000 router right now, it's rock solid. For a captive portal specifically, there is a "Tomato RAF" version by Victek. Check it out here:

http://victek.is-a-geek.com/tomato.html

I read the article, and part of the idea is that noise (radio activity) may contain falsehoods, but that silence (radio silence) is genuine and cannot be spoofed. So you first send out a hash, and then try to establish a series of radio silence periods which, when decoded, match your hash. If anything messes with this authentication, it is obvious, and the connection is refused.

See, that's where my general sense of unease with all iOS devices comes from: This is a walled garden which is primarily designed to take your time and money (both precious resources by my standards) and generate a nice profit for Apple. Spending money is made so easy it happens almost without you noticing - that is, until you get the bill. Want to power up your device? Press a button? Please register your credit card first. This is like a phishing website turned into hardware.

And all that mostly just for mediocre, (Adobe) flash-game like entertainment. Almost no productivity. The only application for the iPad that I could see is the one I am still waiting for: an iPad which is responsive and accurate enough to actually perform well as a sketch pad for artists, i.e. like a Wacom Cintiq tablet monitor.

This thread made me read up on video compression, and I can now articulate more precisely why my favorite video codec is Motion-JPEG - It uses 100% I-frames, which makes editing easy, and which makes fast motion scenes look better than codecs which use P and B frames. The only downside is that Motion JPEG doesn't offer the best compression, but it's still reasonably sized.

Assuming one could enter the tower above the turbines, wouldn't this be a nice way to launch with a paraglider?

Well, from my point of view this chip is simply a very sensitive protein quantification device. You can measure a wide range of proteins (related to cardiac disease, allergies, Alzheimer's, and many more), not just "cancer biomarkers", i.e. proteins which are suspected/proven to have a link to cancer.

How reliably can one really diagnose cancer from a blood protein test? In my opinion, cancer has so many different forms (it mutates constantly) that it is harder to find a common and highly reliable diagnostic marker for it than it is for many other diseases.

For that reason, right now a definite cancer diagnosis is still made by physically finding the tumor tissues, I think. However, there will probably be enough data to perform a high confidence diagnosis from blood tests in the near future.

The chip mentioned here could speed up research and adoption of protein tests into general medicine.

As a little mini-overview over biomarkers:

Mostly Established:

Pregnancy: hCG (human chorionic gonadotropin) is a high-confidence biomarker
Cancer: CEA (cancer embryonic antigen) was used in the mouse model in the Nature paper
Prostate Cancer: PSA (prostate-specific antigen) used to be highly regarded, now somewhat disputed
Heart disease: Troponin-I is a very specific marker of heart tissue damage

Upcoming (i.e. prospective biomarkers have been identified but need to be validated):

Alzheimers
Autoimmune Diseases
Etc. (Lots of research going on).

P.S. Here is an older publication which has been open-sourced (open access) and which shows the technology:

http://www.pnas.org/content/105/52/20637.abstract

I'm part of this research and I'm pleasantly surprised someone posted it on Slashdot. To answer some questions: The device is indeed a concentration-measuring chip (not just positive / negative, which would be simpler), and in a just-posted Nature Medicine paper it shows that the signal vs. concentration curve goes 1000x farther on the low end (and the high end too, i.e. more dynamic range) before blending in with the background than the same assay (and antibodies) used on ELISA. Plus, it is a simple device that performs identically in saliva, urine, different pH and temperatures, and which is generally rugged and not too picky about the experimental conditions. This is quite helpful too.

Another point of the publication is that this device can measure small but slowly increasing tumor marker concentrations in lab mice which are known to have cancer. The key is that these tumor markers can be measured with this chip, but are too small in concentration for the traditional platforms such as ELISA. This means you can (in mice, at least) get important early cancer growth trend information (from a blood test) which you probably wouldn't have been able to obtain before.

Just published in Nature Medicine Advanced Online publications (unfortunately requires subscription):

http://www.nature.com/nm/journal/vaop/ncurrent/abs/nm.2032.html

Technical Report abstract

Nature Medicine

Published online: 11 October 2009 | doi:10.1038/nm.2032

Matrix-insensitive protein assays push the limits of biosensors in medicine