They can't validate a scale using unintervened progression or existing treatments, then pretend it says ANYTHING about a new/unknown treatment. The whole point of a new treatment is to alter the progression of the disease in a new/different way; the whole point of clinical trials is to determine the NEW course of the disease using the NEW treatment.

The claim made here is: a better tool to predict the time progression of headaches treated with aspirin (or beer or sex) can better predict the time progression of a headache treated with some yet-uninvented drug, so we needn't test the new treatment as thoroughly to characterize it. That's like saying "the more predictable sex with your partner is, the more you know about sex with a different partner"

And yes, I AM a physician and molecular biologist.

When I was in medical school (decades ago), we had a lecture by one of the pioneers of endoscopic gall bladder surgery (cut some 1-2cm slits and use long-handled tools and a tiny camera to cut/remove/etc) which I well knew was already preferable to the "open procedure" that slashed the patient open (classic surgical proverb: you can never have too much exposure) so you could have the working space to reach in and do it with your big mitts)

I was a big fan, but as a student of both philosophy and the history of science I had to ask how he justified performing the procedure *before* (until) he got the complication down to the level of the standard open incision. He was outraged (as were my classmates) and tersely stated that he had gotten consent (not knowing that I'd done a thesis on the inadequacies and inherent ludicracy of getting "informed consent", especially based on information from the surgeon who wishes to do the procedure).

It was a sincere question, one that I felt could not answer to my own satisfaction (his answer didn't help; he'd simply been looking to "the medical advance" and had never been trained in genuine ethics), but despite that, I feel that he had done the right thing, and that tens of millions have greatly benefited since.

Though not all would-be 'medical advances' end so salubriously, the sad fact is, we don't know any better way -- and I'd wager that we'll have workable fusion generators long before we have a better usable method for making medical advances. "First, do no harm" was a simplistic principle suited to the era before Christ when a doctor was as/more likely to do harm as/than good. (Note that the Hippocratic Oath forbids surgery outright)

We are now skilled enough that some of our advances seem "too good to deny to all comers" without full data -- but where are we to get that data, except by trial (and error). We are not yet advanced enough that MOST of our attempts at medical advance are so beneficial, nor are we advanced enough to have a much better alternative to "try it and see".

No, "critical mass" is defined as the amount that can create sufficient runaway escalation of the fission rate by capturing the energetic byproducts of fission to stimulate more fission in a positive feedback loop.

I'll tell you what: I'll give you 1 gram of the radioactive isotope of your choice, if you can stop its fission. You can't. You can only moderate its rate somewhat. Clearly it's self-sustaining on every level down to the individual atom

RTFA, which was a study done in Radon, and cited confirming studies with other isotopes.

That said, I personally think it is AT LEAST as likely that both solar processes and terrestrial decay rates are being influenced by some outside factor. This is supported by other decay rate correlations with e.g. the Earth position in it orbit relative to the galactic center.

Under this years new "first to file" patent reform, prior publication, scientific or otherwise is not enough to void a patent, since the standard is explicitly NO LONGER "first to invent", no such claim need be made.

One hopes the courts act sensibly and close this loophole, but given the history of the matter (owners of IP-abusing patents settle or fold if they suspect a case will be ruled against hem, rather than risk a sensible precedent that would weaken their other IP holdings -- or deep-pockets third parties offer them cash behind th scenes to fold) I suspect that the first few lower court precedents to go all the way to a published decision will almost necessarily be silly ones (earnest plaintiffs can't afford to drop their cases even if they might lose) that will have to be reversed (perhaps in an unrelated future case) by higher courts or even the USSC.

The above applies only to the US, of course, but MOST matters of law are jurisdiction dependent

A Falcon 9 can lose not just one but two of its engines *after* launch and still complete its mission.

It's just not going to take off when it spots a problem when it's still on the ground. The Saturn V couldn't spot such a problem, much less abort half-a-second before lift-off. It would have been past the point of no return

How is that 10%?

Nukes are measured in kilo-or mega TONS, 1 ton = 1000kg, so 1 kt = 1,000,000 kg not 1000kg.

Yes, I know the original article screwed up on this point. I wrote that off as a typo, but let's not extend the error here, where future junior-high geeks will find it via search engine and contaminate their intuitions

It was ironic back *in* 1984.

I was an Apple developer pre-Mac, and THAT is when it began looking like Big Brother. The Apple II/III had been open (not quite as open as today's open source, but the principles hadn't yet been firmed up) but everything Mac, hardware and software was locked down. Apple forbade early Mac owners from using ANY hard disk for 18-24 months until they developed their own. Even an external HDD voided the warranty which you *needed* because Mac's expensive non-user-replaceable power supplies blew out every 6 mos, even without the load of an HDD

I'd think pole-strung Fiber would need steel strands for structural strength in high winds and other potent weather -- underground fiber has less need of structural strength.

The steel strands, however, happen to be conductors which need to follow proper isolation procedures.

After a year of increasing interest, I'll be buying a reader or tablet this week. I almost pulled the trigger 3x this weekend, but each time found a better deal on a better model -- though ultimately, *any* of the the three, in hand by the end of next week, will be adequate

I was excited to read about this release. It felt like a serendipitous alignment until I realized that I wouldn't have actual possession of ANY file, just a 'service' feeding me a page at a time -- and Google is quite clear that it logs each page I read and when (it touts this as a feature, saying they record it so I can pick up on the same page of each of my ebooks on any other device).

Do I want to be cut off from all my eBooks in wifi or wireless outage? No. That's when I'll want a book or manual most --- during an outage, in a plane, in the woods, in a lab or shielded room... Do I want anyone monitoring and recording exactly what pages I read or re-read and how often, tech or fiction? Nope.

I'm amazed /.ers take this so lightly

So much for serendipitous fortune. This reader is off my list, until it's hacked to keep Google OUT unless invited

Discovered? This behavior was discussed in great depth in my sophomore cell bio text 30 years ago.

That course inspired me to get a degree in molecular bio, and I've posted about this behavior often, here and elsewhere. It's remarkable and inspiring in many ways, but any reference to farming -- or sudden surprise that a microbial organism is capable of doing anything but grazing to death is ... sad and ill-informed

I've been experimenting with this for quite some time now, using off-the-shelf parts from Linear Technologies, specialty micro transformers from Coilcraft, and standard junkbox parts. Both companies offer free samples, but the chips and associated coils are each under $5 each anyway.

I found the LTC3108 (or the LTC3801-1 variant) best suited most of my projects, but the LTC3588/LTC3588-1 is better for capturing energy from ambient sound or vibration via a piezoelectric transducer. (Their evaluation kit, which includes all the parts and a selected PZT, is a bit pricey for a hobbyist, so just get some free sample ICs and roll your own)

Their online specs, designs and datasheets provide everything you need to build your own test rig. These chips even come with built-in support for auxiliary capacitor/ultracap storage, including bucking/boosting the ultracap voltage to the programmed output voltage, which I'd expected to have to implement externally. It's all there in one cheap package with a minimal of external components.

Why must a dictionary be electronic? A paper dictionary eliminate the wireless issue.

Why is an unfamiliar calculator such an issue? Are tests really written in a way that requires maximal calculator throughput? If so, doesn't that penalize students who don't work hard at maximizing their calculator speed during their less-strictly time-constrained homework problems? If you tell them they can only use a calculator (which you provide, and make available for prior inspection) with add/sub, mult/div, roots/powers, logs/exps and standard trig/inv-trig functions and algebraic entry (i.e. not RPG or stack based) then they have all the info they need.

I encourage accommodations, but inevitably, at some point, one is merely drawing a line between the failed and the all-but-failed (even if "failure" is failure to reach the threshold for an A). No reasonable-length college exam can discriminating to 1% accuracy consistently. Period. Our best studied, and slowly evolved multiple choice exams with sample sizes in the millions don't pretend to 1% accuracy/precision (and measurably fail to achieve it)

Will anyone will ever accept the "unfamiliar calculator" defense for a bridge failure? If a bridge, probe or key presentation failed due to picayune calculation/computer issues, it's always been unacceptable, not excusable

You want an A or a Pass? Show that you deserve one, not that under other circumstances you may have squeaked by

Just so you know, you can build a circulator and precision controller for quite a bit less than the lab equivalent. A couple of years ago, I built exact that kind of sous vide immersion cooking unit for under $30, plus $70 for two controllers, because I was interested in comparing a factory-made PID controller and J-type thermocouple (~$40 incl. S/H on eBay) with a microcontroller and thermistors (~$30). Result: either alone would've done the job just fine.

Actually, since I do electronics/hardware tinkering as a hobby, the entire experiment only cost ~$50 (most of that being the commercially manufactured PID controller), because I already had almost all the parts in my junkbox. If I hadn't *specifically* wanted to compare the PID controller+thermocouple, it would have cost me $20 out-of-pocket. The [repurposeable] controllers I built, and lessons I learned, were well worth the added expense

I like the forms feature. I most particularly use it [on comuters under my physical control] to keep track of my login names on sites I may not visit frequently. I don't want to clear my form history to clear my search history, as they advise and have good reason to believe it would help much if I did. Perhaps the Waltham MA Google center is running some sort of experimental "improvement", but they seem to track more than cookies and browser history

Though I have loved Google for over a decade, I dislike Personalized Search strongly. I've always minimized logins, and clear cookies frequently, but it has been increasingly skewing my Google results for at lest a year.

I know how to craft a search, so why not give me the results I request without adding your proprietary invisible weighting? Or let me set the weighting: I'd been doing that on a wide variety of search engines for years before the Web or Google existed, before Google's search experts (definitely smarter than I) had graduated college, but instead of increasing my options, they've silently decreased/disabled many search refinement options

I can name one good reason, BTW: ease -- not my ease, but their computational load through pre-indexing, etc.

At one time, I used "relatively virgin computers" (e.g. one I normally use for analog data acquisition/analysis) to get impartial results, but in the past year, even fresh OS installs haven't gotten me the same results I get when I travel. I strongly suspect Google is keying their indexes by IP, but changing my Google links to pass through a public proxy may only introduce a different bias, more diffuse, but still not geared to my requests!

Here's a novel idea: search for what I ask for, and give me the options that will help me narrow my search. This sort of "I know better than you" helpfulness is one of the fastest ways to turn me off a system.

IMHO, a lot of the "what-ifs" people are positing have already begun to happen. For example, I run a modest website on the side, under 1.5 million posts, a tiny peripheral mote in the Dust storm of the Internet -- but whether I am logged in or not, I often get results from my own site on my first page of results, even when the topic is less than 24 hours old, is a rare/unique topic on that site; and *is* covered widely by many other sites, larger than my own. That may tickle some webmaster egos, but it's terrible behavior, and I can't shut it off, even with a fresh install.

I'm especially sick of getting the same already-read results on the first page, then finding the answer I wanted on a search from a hotel, cafe -- or a colleague's workplace. (Do I seem as great a fool to them as they sometimes seem to me, just because Google presents information prominently to one, but not the other?