Forgot your password?
typodupeerror

Comment: Re:AI is always "right around the corner". (Score 1) 553

One quibble with your reply. The rules of chess are not complicated. Yes, the pawn has several special cases, and castling is unlike all the other moves in chess, but these do not complicate the game that much. It's more complicated than checkers, but not by a lot. A complicated game is something like Star Fleet Battles or Squad Leader. Those games have hundreds of rules.

The complexity of chess is in how to play well, not how to play by the rules. That was another factor that made chess so attractive to the AI community.

Some people seemed to feel that we could take a good chess playing program and just apply it to any old problem. The techniques can be applicable to other problems, but it sure isn't as easy as some hoped.

Comment: Re:AI is always "right around the corner". (Score 5, Interesting) 553

Researchers once thought chess made a good proxy for intelligence. Not every smart person is good at chess, but it seemed every good chess player was also smart. They worked for decades to make chess programs that could beat good chess players. When that started happening, it was obvious that the programs had no general intelligence at all. They were good for chess, but had to be reprogrammed even for very similar games like checkers. When the ultimate triumph of beating the world chess champ happened, it was more of the same. No real intelligence, just faster hardware and refinements to the search algorithm.

The conclusion is that chess is not a good measure of intelligence after all. We don't have a good grasp of what intelligence really is, let alone how exactly to measure it. IQ tests have all kinds of problems, not least that the typical IQ test is very narrow. Maybe wealth or number of children or friends could correlate with intelligence, but there are lots of problems with that too. Is it smart to have wealth beyond one's present and future needs?

Comment: Re:Okay, so this has what to do with fracking then (Score 2) 154

by bzipitidoo (#47389865) Attached to: Oklahoma's Earthquakes Linked To Fracking

Is peer-reviewed data with a peer-reviewed statistical correlation really that unfair of a requirement?

Maybe. If that's demanded for proof that the sky is blue, water flows downhill, the sun rises in the east, 2+2=4, or God exists, then it is an unfair requirement. Don't ask for proof for simple conclusions that anyone can reach with Occam's Razor. Don't demand proof for the unprovable. Raising those aren't expressing honest doubts, it's playing politics, using doubt to block further inquiry, delay remedial action that might impact someone negatively.

Rather, ask for proof of the counterintuitive. It makes sense that messing with the underground will cause changes in the underground that manifest as earthquakes, and also contamination of underground waters. Prove that fracking does not cause earthquakes. Peer review the proof. I suspect it can't be done, because fracking can and does cause earthquakes.

Comment: Re:Okay, so this has what to do with fracking then (Score 4, Insightful) 154

by bzipitidoo (#47389035) Attached to: Oklahoma's Earthquakes Linked To Fracking

Standard denialist garbage. What amount of fact is enough to convince you? Think about that for a moment. What data would you have to see, to be convinced that fracking is causing earthquakes?

As to proof, how do you know anything is real? We might be living on a roughly spherical shaped object lit by a much larger nearby roughly spherical object, or we might not. We could be living in a giant simulator that is so good, supernaturally good, that we can't tell it apart from reality. God could have created the universe in 7 days. How can we tell? We can't! We understand that we can make good conclusions from observable reality, no matter whether it is real or not. To the best of our knowledge, what we observe is real, but we understand there could be a deeper reality. Whether there is or not does not affect our work.

Comment: Re:Yeah, right. (Score 1) 361

You didn't go far enough. Many of the people in control at these agencies are of average stupidity and somewhat greater than average paranoia, and they fear and distrust smart people just for being smart. They are political hacks who kissed up to the right politicians in the right ways. One of their qualities is blind loyalty to their masters. This problem was at its worst during the Bush administration. Remember how it was nearly treasonous merely to argue against going to war in Iraq? We had little choice but to watch the idiots charge into the War of Choice.

They want smart people on their side, but constantly fear that those same people might turn traitor according to a very broad definition of treason. They want those smart people thinking only about the technical details and not any larger implications. They reserve for themselves the right to think about larger pictures as long as they aren't too large, and seem to really believe that's acceptable. They and their masters do the thinking and smart people are supposed to do nothing else other than make it happen. They are blindly loyal to their masters, and expect their underlings to show the same blind loyalty to them. A smart person just thinking about larger pictures is potentially treasonous. They also want contradictory things, and will suspect inability to accomplish two opposing goals could be treachery. You could do it if you really wanted to, and you're just giving them bull, is what they're wont to think. Why don't you want to do it? They think smart people can do almost anything, particularly black hat stuff. But at the same time they constantly suspect incompetence, especially when hearing protests that something is impossible. They actually want to see smart people humbled on occasion and when they think they've seen a mistake, they jump all over it, indulge in a bit of bashing just to enjoy bringing a smart person down to their level. Sometimes they resort to threats, think that can make things happen. They totally fail to see their own double standards and hypocrisies, and that their thinking is irrational, stupid, and vicious, and drives smart people away.

Comment: Re:Longevity (Score 1) 196

by bzipitidoo (#47371681) Attached to: The lightbulb I've most recently acquired ...

There are often good reasons to go cheap. If better technology is waiting in the wings, it doesn't make sense to invest a lot of money in high quality, long lasting implementations of inferior tech. Look at the homes of the rich of the early 20th century and earlier. They have high quality obsolete tech all over the place.

For fluorescent lighting, I replaced magnetic ballasts with electronic ballasts. They are more efficient. I also experimented and replaced a 2x40W fixture with a 2x32W. 32 watt fluorescent bulbs are smaller diameter but emit as much or more light. Trouble was, it was going to take 10 years for the switch from 80 watts to 64 watts to pay itself back, and that's only if the lights are on at least 8 hours of the day. When the LED night lights arrived, we started leaving those fluorescents off at night. Now it looks like they will never be paid back. Cheaper to toss the 32w bulbs and electronic ballasts, and switch to LED.

LEDs aren't the final word. Even more efficient to just use a skylight, and go to bed when it gets dark. We as a society are hurting ourselves with far too liberal use of artificial lighting. Messes up our circadian rhythms, causing hormone imbalances and obesity.

Comment: Re:Longevity (Score 1) 196

by bzipitidoo (#47371373) Attached to: The lightbulb I've most recently acquired ...

I have noticed a high failure rate for LED night lights. I don't recall the exact numbers, but out of some 20 purchased, 4 or 5 failed after only a few months.

Another problem is high variability. Even in a package of 4 supposedly identical models, some LED night lights will be much brighter than others.

Perhaps another place to get data is traffic lights. How often do you see LED traffic lights with dead pixels, so to speak? I see that all the time.

Comment: Re:Cost (Score 1) 228

by bzipitidoo (#47364649) Attached to: Nathan Myhrvold's Recipe For a Better Oven

The point is, we can have a much better oven without an increase in price. Same amount of material, or even less material. There is low hanging fruit that is being ignored. Consider how long it took for toaster ovens to get timers. Years after the introduction of microwave ovens, all of which have timers and automatic shutoff, most toaster ovens still had nothing more than a cheap thermostat.

It's a similar story in housing. The features of the site are routinely ignored. Air conditioning coils should be placed on the east side of a building. It would be so easy and zero cost to simply flip and rotate the plans to position the coils there, but they don't. In most places, half the energy used by a house is spent on mere heating and cooling. Houses should have much better insulation. Instead, money is spend on useless bling like the unnecessarily complicated rooflines that will cost a fortune to reshingle. A simple roof would be better and cheaper. Then there is the completely stupid fireplace that was recognized as inefficient in the 18th century by none other than Benjamin Franklin. He advocated a wood burning stove. But we still put badly deisgned fireplaces in every house today. They are not serious methods of heating homes, they are entertainment devices so people can watch pretty flames. But a lot of people are fooled by them, think a fireplace can serve as heat if the furnace is out of commission.

What's with this knee jerk thinking that improvements are always costly?

Comment: Re:So....far more than guns (Score 1) 454

by bzipitidoo (#47334697) Attached to: CDC: 1 In 10 Adult Deaths In US Caused By Excessive Drinking

High bridges are magnets for suicidal people. Not only is death nearly guaranteed, it's nicely public and dramatic. Might even shut down a major route for a few hours. The Golden Gate Bridge has had problems with suicides ever since it was built. Authorities are finally taking some preventative measures, like adding netting so it's not quite so easy to throw yourself off. We have rails that make it difficult for cars to drive off the side, but human bodies slipped through the cracks, so to speak.

If high bridges are so attractive as a means of suicide, it makes sense that a tool purposely built to kill would also be attractive.

Comment: Re:I recommend (Score 2) 176

by bzipitidoo (#47334493) Attached to: Ask Slashdot: Is It Feasible To Revive an Old Linux PC Setup?

DOSBox is good.

I recommend against a native DOS setup like FreeDOS, unless you don't care about graphics and audio. FreeDOS actually works fine on modern hardware. The problem is that there are no drivers for modern video and audio. As far as I know, there are no emulation layers either-- no way to glue a Soundblaster interface to a modern audio interface with a DOS driver that DOS games can use. Graphics are worse. Without drivers, you're stuck with 320x200x256 color VGA or 640x480x16 color EGA/VGA. Ever try to use Windows at 320x200 resolution? Many old DOS games had their own graphics drivers for the most common graphics hardware, and simply will not run if the hardware they're built for doesn't exist. An emulator like DOSBox takes care of those issues.

As to well aged Linux, there's a huge dividing point at the change from libc5 to libc6 which happened in the late 90s. Binaries compiled in the days of libc5 are going to complain and crash because they can't work with libc6. Expect lots of library hell. Another big change is the still ongoing shift from 32bit to 64bit that reached a tipping point in the mid 2000s. Nearly everything from the early 2000s is going to be 32bit. Although 64bit OSes include many 32bit libraries, you'll likely find it easier to just install a 32bit OS. There are extensions for using more than 4G of RAM in 32bit mode, but it may be easier to just work within 4G.

Digging out and installing an old Linux distribution is going to be more trouble. If it's from the days of libc5, X won't have drivers for GeForce or Radeon graphics hardware. You'll have to settle for slow performance and low resolution from standard VESA modes, or even VGA modes. 1024x768 is nice to have, but don't be too surprised if you have to settle for 800x600 or even 640x480. Another problem is USB. An old Linux probably can't read a USB keyboard and mouse, must have the old PS/2 connectors. Then there's the hard drive. An old Linux may not be able to handle SATA let alone SAS. Has to be SCSI or IDE. The hardware may be able to help here, as the BIOS may have options to run an emulation layer, provide the old bus interfaces that the software expects. To install vintage Linux, likely need a CDROM drive, or even a 1.44M floppy drive.

Comment: Re:Time to Legislate Data Mining (Score 2) 162

by bzipitidoo (#47326119) Attached to: Hospitals Begin Data-Mining Patients

Few issues are completely one sided, but slavery is about as close as you can get. That roof over their heads is just cheap rationalising to help the masters feel good. Like patting yourself on the back for feeding someone a fish today, when you could have taught them how to fish but you won't because you want to keep control. You even go as far as stopping them from figuring out how to fish on their own, on the notion that they can't handle such dangerous knowledge. Slavery wasn't even really good for the plantation owners. Their world view was seriously warped by the prejudice they ingrained in themselves. They really believed their self justifying propaganda about blacks being inferior, latched hard onto the whole idea of the White Man's Burden. Laboring under such wrong thinking leads to systemic weakness.

The ultimate reason the Confederacy lost was that when they started the war, they were already way behind economically, and that was thanks to slavery. Slave powered economies simply are not competitive. Very static, resistant to change, and lacking innovation. They deluded themselves that southerners were more manly than northerners. Hoped that, a few other advantages like King Cotton, and most of all the advantage of being the defender would be enough to tip the scale against the Union's huge numeric advantages. But often their leadership would squander the defensive advantage by making reckless assaults, possibly out of that misplaced sense of greater manliness. Lots of battles in Confederate territory have more confederate than union casualties. General Hood was the ultimate in reckless aggressiveness, and President Davis put him in command because he wanted aggressive action. The result was that Hood got his army killed, first seriously reduced at Atlanta, then finsihed at Nashville.

What I don't like is the "blame the victim" angle of this data mining. Instead of this approach of mitigating things the consumer did, as if they might be bad for our health, why not grill the store? Like, instead of haranging the consumer about a pizza they ate, what about a talk with the pizza vendor for using too much salt or fat or whatever? One thing that the US does is pour way too much salt on our food. There is precedent. I think bars are legally constrained not to sell alcohol to someone who is drunk, and/or required to prevent them from driving away.

Comment: Re:IF.. (Score 1) 561

by bzipitidoo (#47323857) Attached to: Match.com, Mensa Create Dating Site For Geniuses

For much of the 20th century, it was thought chess made a good proxy for intelligence. Skill at chess correlated with high intelligence. Though it was clear that lack of chess skill didn't mean a person was stupid. This correlation was believed so strongly that the AI community bought into it and tried for decades to make a computer beat highly skilled human chess players. When this effort finally succeeded, it only further confirmed what many had suspected for some time, which is that skill at chess can be obatined through sheer brute force calculation. It doesn't require intelligence, whatever that is exactly, though that helps. We need better measures and definitions. And, yes, most IQ tests aren't it.

Comment: Re:Corporate Brianwashed Fools (Score 1) 710

by bzipitidoo (#47314905) Attached to: Workaholism In America Is Hurting the Economy

Be a team player

That is among the most hypocritical phrases commonly used in the workplace today, ranking up there with "show your commitment". Most of the time it's management trying to whitewash that they've just demanded something unfair or illegal of a worker, and threatened their job over it. You should buy a new car, show your commitment (and, uh, set yourself up so that if you lose yoru job you can't pay your car note and will lose it too). Of course they can't say that, so they turn to hinting. And a few times, they honestly mean be a team player, in a good way. They have to mean it sometimes, or the phrase wouldn't work as well. Be a team player, yeah!

Joe Slacker

That's another huge problem. This mentality that people are naturally lazy and have to be forced to work is wrong, but so few people believe that anymore. They accept a whipping for being a slacker as for their own good. I wonder if we'll have to fight the US Civil War over again someday, stop the new slavery.

Comment: when is our govt going to do this?! (Score 1) 69

by bzipitidoo (#47307783) Attached to: Google Building a Domain Registration Service

Buildings and lots all have addresses, assigned by the US Post Office if necessary. Highways and streets all have numbers or names or both.

We all ought to have our own addresses on the Internet. No one thinks anything of having an IP address, and everyone who knows anything about the Internet realizes an address is necessary. Why aren't names accorded the same importance and privilege? We need stable addresses, and with dynamic IP, we don't have that. I don't like such vital connectivity being in the hands of a private company no matter how good they are.

You don't have to know how the computer works, just how to work the computer.

Working...