Forgot your password?

Comment: Re:I don't get it. (Score 1) 541

by FreedomFirstThenPeac (#47658539) Attached to: Geneticists Decry Book On Race and Evolution
If the genetic sequence "ACGTTGTA" is correlated with a differential ability to do some cognitive task and the genetic sequence "GATACCA" is associated with the ability to grow good long hair, and the two sequences are linked (a mathematical/statistical term in this usage), then it is possible to use hair as a visible predictor of the cognitive task ("playing thrash metal", for example). Correlation may not be causation, but it can be an indicator variable.

Comment: Re:Half story (Score 1) 35

As a mathematician working on data mining where we still see lots of false positives, and with the proliferation of easy tools for fools to do data mining, I wonder how long till we see panics starting days or weeks before the government is willing to announce problems. Imagine New Orleans trying to evacuate itself while the NOAA folks think that the weather that is coming is going to be a standard low level rain event. Imagine then if it turns out that NOAA was right to be calm, FEMA was right to sleep through it, but hundreds of thousands of Wx-refugees are now sitting on freeways trying to find gas. The movie Contagion was a good preview.

Comment: The real timing question (Score 1) 184

by FreedomFirstThenPeac (#47532819) Attached to: "Magic Helmet" For F-35 Ready For Delivery
What does a physically limited human in a helmet in the cockpit bring to the battle that a physically remote human in a helmet cannot do? Is the extra "situation awareness" brought about by the kinesthetic sense and the millisecond lag caused by speed-of-signal issues worth the extra cost of making an expensive toy for pilots over making a slightly less expensive toy for armchair warriors? Do we really expect the on-site human to be able to whip that $600K helmet off, squint Dirty Harry style and squeeze of a few thousand well placed rounds into a target that cannot be seen without the enhancements of that helmet? When I was AF, I'd have raised these issues and probably been told what I was told then, the remote sensing and control technology just is not up to the task.

"Prove it" says I, and I would invoke the post WWI demonstration bombing that got Billy Mitchell in trouble.

Comment: Minnesota - Land of 10,000 Lakes (Score 1) 377

I am a member of two planning commissions in Minnesota and I find it very ironic that here in the Land of 10,000 lakes (or, in the spring, one really big lake), we are having to block ethanol plants and agricultural irrigation because of ground water and deep water concerns. Similarly we are finding that the ground water we do have is slowly being destroyed by run-in from fields covered with chemicals. It does make me an outlier in the Republican party (social liberal wing thereof) when I pose these "tragedies of the commons" arguments to the died-in-the-wool free-market libertarian types. I can show them the specific assumptions in their models that cause them to FAIL (mode critical) and as a mathematician I am often surprised that they do not see how those failures force an external, non-free market solution. But I soldier on.

Comment: Re:Astronomy, and general poor night-time results. (Score 1) 550

by FreedomFirstThenPeac (#47532531) Attached to: Laser Eye Surgery, Revisited 10 Years Later
I went with both eyes good for distance vision because I wanted to be able to see well if I fell off my sailboat and lost my glasses (since they would be off-the-shelf sunglasses or reading glasses). I have been quite happy with the results and represent a sample size of 1. On the other hand, I DO need reading glasses, but they are cheap enough that my house looks like an explosion in a glasses factory.

Comment: Re:Seems like old times (Score 1) 752

When I was teaching at the Air Force Institute of Technology, on the topic of the the math behind IFF systems, I used this example (Vincennes incident) as an example of how human factors enter into battlefield decisions in a way that can nullify the best planned algorithms. Now I am using those examples in the hospital decision making environment.

FWIW, the example problem I presented was of an airplane heading towards a base, flying with no IFF transponder, flying low and erratically. The question was whether it was a damaged friendly (no IFF, no radio) and returning to base or an enemy spoofing to look like a damaged friendly. The Army troops were unanimous, "Shoot it down; sort it out on the ground". The flyboys were not so sanguine.

Comment: Wrong question ... (Score 1) 285

The first thing you have to ask is whether a computer that passes the test has some rights that other machines don't. The test we are looking for is one that, if a program passes the test then legal protections would intervene if you wanted to shut it off and scramble the memory. Any other test is just semantics and tomfoolery, like arguing over what color is the sky. Without the actionable component (a blue sky means I don't need my umbrella to get across the parking lot to my car) the question of "best test for AI ..." is a form of mental self-abuse, without the happy ending.

Comment: Re:It's already going on... (Score 1) 353

by FreedomFirstThenPeac (#47416317) Attached to: Here Comes the Panopticon: Insurance Companies
Wrong. Wrong. Wrong.

The incentive is NOT to make the most money. It is to charge just enough that there is very little room for a competitor to slip in a lower premium for the same protections. Thus, if you are deemed to be an expected cost is $1K/year driver then your premium should be about $1.1K a year to leave them some protection against the high-side risks (protection they usually get through re-insurance, but that's another concept and we don't want to overload the wet-ware circuits). Conversely, if you are deemed to be an expected cost is $10K/year driver then your premium should be $11K a year and I bet they just hope someone else comes along and offers you a better deal.

But "increasing the total sum they get from premiums" is not even a first order approximation of the TWTWW (the way the world works).

Comment: Re:Don't mention the tree-planting thing! (Score 1) 228

by FreedomFirstThenPeac (#47371455) Attached to: The New 501(c)(3) and the Future of Open Source In the US
I guarantee that if more than some (insert your guess here)% of the economy went into barter then the IRS would start enforcing EXISTING laws that make barter transactions taxable as income. So, calling brownies money (as barter) does not make them untaxable, just (currently) untaxed. As for non-profits getting a special tax break, why not (1) eliminate all corporate taxes, (2) remove free speech from corporations and unions, (3) tax wealth more (level that entitlement cliff), (4) tax consumption more (carbon tax, anyone?), (5) tax income less (but progressively).

This list is not supportable by either party, which makes me think it could be the best set of trade-offs ever. The only improvement would be if estates of everyone who died were taxed at a rate that paid off their (proportional) share of the national debt that they let get run up on their watch (without regard to which party incurred it). Might make those 20-somethings think a little bit about how much debt they wanted their cohort to put on the books.

Comment: Re:Weather is NOT climate (Score 1) 567

As a practicing mathematician and statistician I really don't think the publicly available data is suitable for any weekend warriors to do their own analyses. The data that is available, I fear, will have been "cleaned" to ensure consistency with local dogma rather than to ensure suitability for analysis.

"There are things that are so serious that you can only joke about them" - Heisenberg