GI in Japan after surrender of Japan in WWII, picked up gut bacteria and whenever he ate carbs he got drunk.

That may be the one my parents told me about, back in the '60s or so (but as a war story which probably puts it in WWII.

In the one I heard about the GI was thrown in the brig and put on bread and water - which of course made him even more intoxicated. Then they mounted an investigation to see how he was getting the booze smuggled in. That finally showed it was a medical problem.

Turns out he had diverticulosis - one or more failures in the intestinall muscle wall where the gut membrane bubbles out into a little appendix-like pocket and is prone to infections - and one of these became home to a culture of brewer's yeast.

Not only does it cost a LOT to port this stuff and risk errors in doing so, but the cruftier it is the harder (and more expensive and error-prone) it is to port it.

If, instead, you can get the new machines to run the old code, why port it? Decades of Moore's Law made the performance improve by orders of magnitude, and the behavior is otherwise unchanged.

If you have an application where most of the work is done in a library that is largely parallelizable, and with a few tiny tweaks you can plug in a modern multiprocessor-capable library and run it on a cluster, you get another factor of almost as-many-processors-as-I-decide-to-throw-at-it, with small effort and negligible chance of breaking the legacy code.

What a deal!

And it's one less reason to touch the tarbaby of the rest of the working legacy code.

Let the COMPUTER do the work. People are for setting it up - with as little effort as practical - and moving on to something else that is important and can't yet be automated.

Eventually somebody will teach the computers to convert the Fortran to a readable and easily understandable modern language - while both keeping the behavior identical and highlighting likely bugs and opportunities for refactoring. Until then, keeping such applications in the legacy language (unless there's a really good reason to pay to port them) is often the better approach - both for economy and reliability.

Not at all. The size (spacing) of the pixels relative to the wavelength limits the angle through which you can (first-order) diffract the beam,

A hologram of a scene where only a small region, very near the spot the reference beam is aimed at, i.e. one that leaves most of the light nearly parallel to the reference beam on reconstruction, is a very low-resolution, big-blobby thing.

60 Hz would be terrible. Let's imagine they can IQ modulate 64 symbols (6 bits per symbol) with a samplerate of 60 Hz, that gets you 6 * 30 (Nyquist) = 180 bits per second.

That's per pixel in the modulator. Now multiply by the number of (mega) pixels.

The modulator is essentially the guts of an LCD display so you get the same bandwidth as the video image it could display. What's new is:
  - You're modulating the(many) modes of the cavity itself, rather than shaping the beam after it emerges.
  - You're modulating the modes rapidly, rather than by hand-replacing components (call it one symbol per hour if you're good at it).

My impression is that, because the mirror is "replaced" with the LCD, the LCD is inside the cavity, with each pixel modulating either the Q or the polarization of a particular chunk of the cross-section of the cavity. This amounts to adjusting the gain of the various modes of the cavity and thus switching which one(s) oscillate and consume the energy from the amplifier in the cavity.

Though the modes that are selected would not be mapped one-to-one onto the pixels, , you can control a lot of modes with the ciquid crystal display - probably all of them available, or up to the number of pixels in the liquid crystal device.

You can also switch them as fast as the liquid crystal switches. With modern drivers (which remember the previous state of the liquid crystal in each pixel and temporarily overdrive those that must change more in order to switch them rapidly, rather than just letting them settle passively into the new state) you can switch it at 60 Hz or better.

You might use holographic techniques to change the angle of the beam, or emit a number of beams of various intensities in various directions. Result: Scanning and image formation without moving parts (other than the molecules in the display).

I think the computation to turn it into a (one-color) projector would be pretty much a straight 2-D FFT times a nonliinear tweak to deal with energy-stealing among modes.

I'd like to see versions of this with array-of-Kerr-cells in place of the liquid crystal device (for more rapid modulation, at the cost of high voltage drivers), or digital light processors for the mirrors (though the latter are more on/off than continuously adjustable so they might be more limited on what beams they can form).

for the life of me I don't know why Cisco, Microsoft and other big players just don't pay up to get at least some insight into how these guys are finding exposures in their systems

I would assume that VUPEN would refuse to sell to Microsoft and Cisco on account of it diminishing the value of the zero-days they're holding.

Or at least not sell them the best stuff.

Obviously, if Cisco, Microsoft, etc. were going to buy this service, they wouldn't do it (only) as themselves, acting directly. They'd do it through a front, to insure they got the same things the bad guys were getting.

Just as a startup did, about a decade ago, when I was designing a next-generation routing chip, and we needed to obtain equipment from Cisco for testing it for function and compatibility.

It took two half-rack, 3/4 megabuck, top-of-the-line Cisco routers to drive it properly. We bought them through another company on a very hush-hush basis, just to be sure Cisco wouldn't be tempted to send us defective or gimmicked equipment, not support it properly, or hold up shipment and slip our schedule.

Anybody knows how the new commercial space launchers do in comparison?

Don't know about the current crop. But back in the late '80s AMROC controlled their launches without the classic room-full-of-custom-consoles. Instead they hacked up their "consoles" as a GUI on one instance of the state-of-the-art windowing interface computer of the time - a Macintosh (what they'd now call a "Macintosh Classic").

I hear that, when they showed up at Goddard for their test shot, the usual control room crew was standing around with their jaws dropped as the whole thing was run from the little screen on the little box on the single desk. B-)

If you never heard of AMROC: They were the ones that were working with the hybrid rocket: Solid fuel (synthetic tire rubber), liquid oxidizer (liquid oxygen). Controllability of a liquid fuel, complexity halfway between solid and liquid fuel (only ONE set of plumbing, not too that must be synchronized), safety better than either (turn off the LOX and the fuel just smolders and goes out.

They lost their mover-and-shaker founding-team member days before their first flight attempt. Then, though the engines had many successful tests, the actual flight attempt failed in about the worst possible failure modes for a hybrid: The LOX valve iced up (due to ambient moisture) and stuck at about 30% open: Not enough thrust to get off the pad, but enough slow burn energy to destroy much of the rocket and pad equipment, and they couldn't either launch or shut down. Then they went bankrupt, so there wasn't a second shot. (Their tech was sold and some of it is used in space ship one.)

Now I can record even more TV that I'll never get around to watching. Technology is great and progress is divine.

Ha ha, but seriously....

Letting the automation pull the the needles out of the hundreds-of-channels vast wasteland haystack is a classic example of using automation to do the drudgery, leaving you to do the interesting stuff.

There's two hundred channels of crud and 20 minutes per day of stuff of interest? Let the computer watch the junk sieve out the jems for you. That way you don't need to be rich enough to hire an army of interns to do the same.

Or better yet, once in court and under questioning invoke the 5th Amendment. It does work once there. It cannot be used to trump a subpoena and skip out on testifying as a witness.

Until they give you a grant of immunity. Then it doesn't work anymore.

There are two kinds of immunity: Transactional immunity (witness can not ever be prosecuted for crimes related to his testimony) and Use immunity (they can't use your testimony, or any evidence they gather based on its information, against you - but if they find other evidence you're fair game.)

Unfortunately the Supreme Court has ruled that Use Immunity is enough to extinguish the 5th Amendment right and federal prosecutors rarely offer Transactional. Some states have more stringent laws, constitutions, or constitutional interpretations and Transactional immunity must be offered before their courts may compel testimony.

Of course prosecution is not the only life risk for a witness. I know of at least one person, here in the disarmed-citizen utopia of California who has stated an intent to "not have seen anything" unless granted a perpetual concealed carry license.

Quite some time back the people administering the low-income (and gang-ridden) housing projects in south Chicago decided to search all the units for guns. The NRA and the ACLU sued (successfully) to block this unwarranted search of the residents' homes.

they all start with some common underlying mamallian hearing genes and then they tweak them to develop echolocation.

Actually, a lot of animals that aren't credited with using echolocation actually use a variation of it: Sounds from their own motion (such as footsteps) create echoes, which their hearing system processes into a map of nearby objects.

People, for instance, do this. That's why you can "feel" the nearness of walls and objects in the room (especially those near or immediately behind you) without looking, when you're moving.

There's at least one recorded instance of a totally blind child who learned to ride a bicycle and avoid objects, by making clicking sounds with his mouth to provide excitation for this system.

(The hearing system of things like mammals is evolved from the lateral line of fish - which both detects other nearby fish by direction-finding on the sound from their muscle twitches and other sound-reflecting objects by detecting the echoes of muscle twitches of the fish doing the listening. (A flat surface, for instance, would produce an acoustic mirror image of the fish every time it twitched, identifying the return as an echo of the fish itself.) It would not surprise me if the processing for echolocation in other animals is just a revival or slight remapping of this same mechanism.)

please, please ... pass on this advice to a Progressive

While you're at it, point out that a lot of their prescriptions INCREASE risk while purporting to reduce it. It's doubly annoying when they work so hard, throwing money, effort, and restrictive laws into trying to solve a problem when the effort and sacrifice actually makes it worse, in a positive feedback loop.

Progressives have no monopoly on this, either. Neocons, consdrvatives, and even Libertarians do it as well. It's easy for all to do things to attack a problem and not see that the indirect effects of the effort cause more harm (even in terms of the problem being attacked) than the first-order effects help.

Some examples:

  - Gun control: Private ownership and carrying of guns REDUCES crime, violence, victimization, and death, while citizen disarmament increases them.

  - Attempts to police the world produce "blowback", creating new and/or motivating existing enemies, increasing, rather than decreasing, the risk and costs to the US from war.

  - Drug prohibition creates more drug use and criminal enterprises, rather than reducing drug use, and harms the drug users more than the drugs do. Its component programs often have counterproductive pathologies of their own. One example is the D.A.R.E. program, which attempts to use peer pressure to encourage kids to ignore peer pressure, and has been shown to increase drug use.

  - Grabbing advances to any program, rather than considering whether achieving goals in the wrong order makes things worse rather than better. (A Libertarian example: They want both open borders and an end to government wealth distribution such as welfare programs. Unfortunately, opening the borders first leads to an influx of social program dependents, making the overall problem worse (and increasing the voting block to preserve and expand the programs), when fixing or eliminating the programs first would remove most of the downsides to opening the borders later.)

If we extrapolate this curve and assume everything else remains constant, DOOOOOOOOOM!!!!

Darned right. (The authors seem to think the battery makers won't respond to a market for more batteries by building more batteries. Duh!)

As I understand it there's NOTHING blocking production ramp-up for Li-Ion batteries except lack of customer demand, which the auto industry is now rectifying. There's nothing exotic or rare in their composition.

Pretty much ditto for NiMH (except maybe for the price of nickel).

Henry Ford built a bunch of infrastructure to supply his auto company with necessary materials - including building his own steel mills, power plants, and soybean warehousing and processing operations (for early plastics). Any bets on whether Elon Musk would build his own battery plants if the current industry doesn't make him enough (or gouges him on the prices)?

Sure it can. A process can generate a lot of some material which nobody currently needs. The manufacturer will then go and look for a market which can use this material and try to develop that market.

That was the case long ago for gasoline, it was a useless by-product for a long time there...it was actually thrown causing some environmental problems, till they could finally figure out a use for the stuff.

Other examples:
  - Plastics
  - Asphalt
  - "Coal-tar" dies.
  - Liquified Natural Gas from remote oil fields (like the Middle East).

One of my favorites: Stove Pellets: They're made of sawdust from lumber mills, which used to be disposed of by burning it on site. They can sell them very cheap and still be far ahead of spending money to get rid of them (especially after EPA regs made burning them pricey). As a result, my house heating (in a mild climate place where shipping raises their price substantially) costs me maybe $300-400/year, vs. several thousand if I were still using natural gas. (It's carbon-neutral, too.)

Ping time can geolocate? Within 10 feet. I'm about to piss on myself laughing. Ping is more determined by quality of network than distance, and varies according to the route each ping takes. You assumptions assume every route is the same, or is on a cell phone. They are not.

Go play network expert somewhere else, and get off my lawn.