Posts by FelixReg 100 publicly visible posts • joined 7 Nov 2007
"When I see someone say there needs to be a retraining program they often ignore the fact that many do not have any real skills for the new jobs."
Huh? Retraining programs are *based* on the assumption that "many do not have any real skills for the new jobs".
Another base assumption of retraining programs is that people can be trained. Apparently, you think training is "likely to be a complete failure." Perhaps you're right. But, others may disagree.
Half that gig. The "parameters" would probably be 32-bit floats, but there are ways to cut that down, too.
The "parameters" are generally multipliers stashed in a GPU's memory and divvied up in parallel to however many computation units the GPU has (in the 1 to 3 thousand range nowadays for a single, good GPU card).
Have a ball scanning the "patent citations" in this thing. They start in 2003. Apparently email was new then.
And what a cluster of jokes!
They seem to be like the infamous Compton's patent, one of the first software patents. Compton seemed to take a typical system description document, such as big organizations make before beginning any project, and put it in patent-ese form and language. Viola! A novel idea. Which, hey, maybe it was novel to the talent at Compton's, apparently unfamiliar with computers. Or unfamiliar with an "electronic computing device", which back in the '70's or so was what you sprinkled your description of the wheel with to make it novel and patentable.
Wanna speculate on how may patent applications there are right now sprinkling their software descriptions with "machine learning" or some such blather to make blatantly obvious ideas able to support the livelihood of patent lawyers?
Sprinkling patent applications with accepted buzzwords is one of a plurality of novel methodologies my (patent pending) automated patent generation system employs. My novel buzz-sprinkle method, utilizing machine learning technology, will be for sale as soon as it passes review. As will my novel neural-net based buzzword identification and extraction method used to support the buzzword sprinkle ... blah, blah, blah.
"Mindless extrapolation..."
Quite.
Which is why one tries to keep the mind alive when doing this sort of thing.
Personal experience says that such extrapolations can predict with uncanny accuracy. The uncanny part comes from:
1) The curves are exponential. The mind does not do exponential well, normally.
2) Though others see apples and oranges, the prediction is about fruit - fruit-ness that others somehow miss.
In the '70's the turn-around time for running a program was often 24 hours. A decade ago, 30 years later, mechanical engineers could "build" a thing using CAD/CAM and 24 hours later the physical thing would magically appear on their desks.
If the curve for such custom, physical builds follows that of computer programs, we're talking Star Trek replicator tech in the mid '30's or so. 3D printer enthusiasts may nod their heads and say, "Duh".
There will be racing tension between those who build on-site and those who ship finished goods.
The future is fun if you're not frozen in a hand-wringing, Hollywood angst world.
The paper seems to be behind a pay-wall, so it's not clear whether the evidence of generalization (e.g. the bees pushing a closer, differently colored ball than what they've seen pushed) is truly generalization or whether it's that the bees just don't notice the "closeness" or "color" of the ball.
Put another way, how do you distinguish cluelessness from deep thought?
Good idea. That way you make sure to count the votes of those who don't care. Say, giving them two votes is even better! They are, after all, less likely to be influenced by opinion news, and fake news, right?
Or.
Instead, go for the highest quality voters rather than the lowest. Allow voters to save their votes. If you don't vote this election, next election you have two votes. And so on.
Yep, DrStrangeLug. Scan through Datamation magazine in the late '50's, early '60's era. You'll find an article telling managers COBOL is going to let managers program the computer directly without needing those odd-ball, troublesome, different-from-you programmers.
The article was right! Just try to distinguish a post-50's corporate data processing programmer from a '50's manager in thoughts, understanding and attitude.
We should not forget that our opinions about these two are a distillation of what we see and hear through the media. And, with the Internet making it easy for even right wing people to live in an echo chamber, much of what we "know" is bad-mouthing by the candidates' opponents.
So instead of a choice between two very capable, successful people whose plans for the US administrative branch may not align well with our personal opinions, we have a choice between Eva Peron and Biff.
MD Rackham: You're promoting child labor, aren't you?
Can't speak for AC, but I do. It really bugged me that my kids could not legally work, AKA "Do things for other people." It was both cute and sad when one daughter would regularly ask for a job when we were out and about. She was pre-teen and didn't understand why she could not do what she saw others doing.
Sure, most classic kid jobs are long gone. Especially those matching our weird images of "child labor" - working 27 filthy, dangerous hours a day for a bowl of gruel. But get rid of these antiquated laws to find out whether there are age appropriate jobs that kids *can* do. Kids would be much better off.
To be fair, it's hard to spot a HIPPA violation in your 2014 story.
HIPPA is written-out common sense with some oddities regarding federal prisoners. But, HIPPA is not interpreted in the spirit of common sense. Bureaucrats are not fired for making work for other bureaucrats.
"cost insensitive business"? Let me fix that: "cost enhancing business". See the previous paragraph for one minor example.
One more fix: "solution to this will be" -> "solution to this was". Past tense.
Or is it better just to add another character to your password?
It might be interesting to test this. Have a site that forces 1/3rd of the users to jump through the special-case-sensitive-character hoop, 1/3rd of the users to enter what they want, and force 1/3rd of the users to enter 1 or 2 more characters than whatever the middle third are required to enter. Then look at the passwords chosen by each group. Which are more crackable? By a machine.
Something that's often forgotten is you don't have to create a file with a perfectly matching MD5 or SHA1. All you need is a file with hashes that match at the beginning and end, and for enough of the other hex digits to *look* ok.
Though semi-matching *two* independent hashes would be a neat trick for the bad guy to pull. I'd worry that MD5 and SHA1 are not particularly independent, though. They are algorithmically close.
Use two USB drives to back up the master, local copy of the data. Keep one drive at a friend or family member's place. Swap them every so often.
USB hard drives are as cheap or cheaper than optical discs and 100% handier. And 100% more likely to have their contents transferred to more modern media as time passes - which is the only way the data will survive, long term.
You can use rsync or rdiff-backup to copy to the drives. The --compare-hash or --compare-full options to rdiff-backup may help guard against bit-rot. Though, if we're talking pictures, movies and the like, how bit-perfect does the backup need to be?
To put "long term" in perspective, get a credible doctor to say you have about a year to live.
The article should probably say how much data you want to save.
Alan, are you sure about the prices. It appears HD's are about $30 a T. I don't see any 4x, $120 SSDs out there. SSDs still look to be in the 10x range many years after Register commenters said HDs were dead.
Sigh, from a guy who would like to justify a little 2T+ SSD on his own machine to handle most of the non-media stuff.
"The presumed goal of advanced roboticisation of society is to eliminate work as the primary occupation of human beings, The leisured society."
Well, I think "work" is a four letter word meaning, "What you do for others." And, so far as I can see, that's roughly how the universe defines it.
Eliminating "work" does not seem a worthy goal.
Consider this goal: To raise the value of humans compared to the things around them.
That sure looks like what the industrial revolution is doing, goal or not. And this rising value of humans has caused all sorts of problems when attitudes about the value of humans have lagged reality.
It seems that, while it's important to remember that things people pay for are very, very likely to have positive value, and that pretty much all our gadgets and services are from other people, we should not forget a huge value stream we tap in to: That new people are born every day.
At, say, a million bucks a pop, that means a million new people are a trillion smackers added to the balance sheet. Life, itself, is pretty valuable, moola-wise.
Language. Hence, humans.
Tool making (e,g, the bag and spear for starters). Hence able to go anywhere on land and sea. Many more humans.
Animal domestication and breeding. Hence domination of other animals.
Agriculture. Hence domination of plants. Many more humans.
Literacy. Hence ability to organize on large scale.
Industrial revolution (which we are in the center point of right now). Many more humans.
Arranging them in importance is like arranging in importance all the turns you make in a road trip.
But, then, hey, there's always the battery. Implying use of the electro-spectrum.
One wonders whether there might be a way to total up a current national net worth.
Like start with counting the net worth of all a nation's warm bodies, including the dough they have "in the bank", though comparatively speaking, that dough might be chump change.
Add the value of "infrastructure".
Maybe subtract spending on or investment in things that, in a perfect world, might warrant zero spending. Security things: cops, bars over windows, etc, come to mind. Such subtraction might help account for "social capital".
That's the strategy I heard as a kid about what the new supermarket chains did in the '30's.
Which thinking sort of explains the '30's in a cute way. That is, what with thinking that's what happened, it's no wonder those people kept a depression going for so long.
Count the number of times someone eliminated the competition and raised prices. And then count the number of times prices kept going down.
Or just use common sense. Are data storage prices going up? Ever? And is Amazon going to drive data storage competition out of the market? Ever? With prices that pay for an equivalent new disk drive every 3 months?
Didn't know about that ship/buoy difference. Makes sense.
Yep, if you ever see raw data these climate guys have to work with you sure can understand why the group picture at that site shows a big group. Lotta work to clean that historical stuff. Heck, back in '80 I saw real-time temperature readings from all around the world. Lotta typos.
Another more subtle problem is, how do you know you've improved the data when you fix it? Correlate the fix-up program's output with something else, generally. That process gets knarly when the only "something else" you have is your expectations and common sense. Dang, my program says the winter temperature in Nome is 124 degrees! Time to find a bug. But if that temperature is -20, well, move on. Other stuff to do. And maybe you miss a bug.
Looking at some of the charts at the sidc.be site, a first thought is any sunspot *count* would be a very fuzzy value. Overlapping spots could be counted in so many ways
Total sunspot area could be used as a measure for modern values. I wonder how much of the historical data includes pictures from which area could be computed.
And, I also wonder whether the whole sun is included in the values. Or whether the values come from samples of the sun visible at a particular time.
Guaranteed, someday a sweet girl will be flying in from New York and find the suburban house and street lights far below her showing a moving-dot display of "Beth, will you marry me?"
And then we'll know all this work we've done to make electronic brains available to every person of every age, in every walk of life, will have been worth it. Yes, worth it, indeed.
Matter of degree on that last suggestion.
Your objectives are smooth and aware flow, while cutting the number of times cars do things like pull out right in front of you and kill you. Because if you're a bike (or, probably, a cute, unthreatening Google car) they *will* do that.
Optimally, your pay-attention-to-me driving wants to wake up the sleeping guy who is just a circumstance away from "pulling out in front of you". At the same time, such driving wants to be effectively invisible to the aware drivers and those who present no threat. This is a subtle thing. Your "reckless" driving is not crazy. It's picking which side of the lane you're on at the moment. It's moving your head in a way that another driver will pick up unconsciously. In the case of you moving toward the car in back of you, you want just a moment of attention from that car. Enough to get the driver's foot off the gas and, maybe, on to the brake.
I didn't mention that you can read and adjust for things like brand-of-car, driver body language, and exoskeleton (car) body language. Others have pointed out that you can sense when someone will be turning by their exoskeleton body language (wobble in lane, move to the turn-side of the lane, etc). That guy who plowed in to the Google car didn't look like a turner. That means he's less likely to slow down and more likely to want to hurry thru the intersection, depending on his age, sex, etc.
And what about the other cars? Are any of them doing things likely to take that guy's eyes off your approaching bumper? Remember, when you're stopped, you are competing for the attention of the guy coming up from behind. If you want to operate at a high level, think about how you manage the guy behind him. You don't want the guy behind you to be rear-ended, either.
Driver-less cars are probably acquiring this sort of knowledge. So a town filled with driver-less cars will feel a *lot* different from a human-driven town.
Questions come to mind.
Was this one of those itty-bitty Google cars? if so, welcome to the world of motorcycles. The big cars are out to get you. If they even see you. Which they often won't.
You must defend yourself.
How?
Well, first, blink the brake lights until the guy behind responds.
Second, even-out the crash. You can be moving forward slightly when he hits you. Always leave room to do so.
Third, you may be able to dodge him by bailing to the gutter. That can backfire if he steers for the gutter to avoid you.
Fourth, if you're a big car, let the guy behind you see the cars in front of you so he can see them stopping. Monitor whether he is watching them - or if he's just following your taillights. If the latter, lose him.
Fifth, don't be invisible. Wobble in the lane. Vary your speed and acceleration. Create uncertainty in the other driver if you can't rely on him. Scare him if he's a bozo. He'll wake up quick. You want to do the opposite of how you lead a drunk driver to safety.
Up-vote for the Grey Goose scenario, noem,
Except a slight change should be made to better match the case at issue: It's not "which outcome would you rather...". It's: Is the second employee's response illegal?
Those who argue against Amazon in this case are saying, "In response to the customer's 'gray geece' query, the clerk (who is not human and is therefore expected to never make a mistake) must either stay silent or tell the customer, 'Our store does not have any Gay Juice'."
The article doesn't tell whether the Amazon search engine in this case had any knowledge of a product and/or brand name. I would expect not.
Break up the huge banks. To the extent that the resultant little banks share environments, they'll flock.
So instead of 1 big bank tanking, 100 little banks go off the cliff together.
They all share their regulated environment, for starters. And, given the nature of money, more than that part of their environments are shared.
Isn't this what was true in '07? Seems like they all had to make liar loans, for instance.
"Whereas the act of procreating in an overcrowded world is the ultimate act of selfishness"
So, living for yourself and letting other peoples' kids support you when you're old is "selfishness"? Interesting. Seems more like knowing how to work the system. And quite canny in being able to apply a veneer of sanctimony to it.
Remember, as Tim has pointed out, the US system is mostly a one-payer system, Medicare. Dying old folks. You could slash overall costs by pulling the plug on anyone who matches some approximate pattern of "gonna die soon". Easy enough to stop life expectancy numbers shining a light on what you're doing.
And the US system is a cost-maximizing system. Is that true of other places, too?
Why cost maximizing? Maybe because it would be uncaring, mean and inhumane to not provide the very best to ourselves and any other people with whom we feel charitable. So spend more, get more.
Which provides measurable proof the US system is the best in the world. Four times better than Singapore! :)
In the US some systemic lies are exposed to all. Several people have commented about US medical billing. These caustic comments don't even scratch the surface. A US medical bill can be assumed to be an arbitrary number, often wrong by any measure and on many levels.
The question is whether closed, single-payer systems hide such things. Comments here seem to indicate such is the case. Do UK people generally know that they have "rationed" health care, subject to budgetary and other oddities? Here is a thought of why people in the UK like the NHS despite the things commented on in discussions like this. Same reason "Jaws" was scary and "Earthquake" was not. No. Not the director or general quality of the movie. :) No, it's that sharks eat you. Alone. Just you. But we're all in an earthquake together. Secure. Warm. Cared for.
In the US you know you're on our own. Alone. You must count your fingers after shaking hands with any part of the medical system. That's not pleasant knowledge. It's not like we're talking about a competitive system, here. They are completely safe from you, legislated, like banks and phone companies.
Am I the only one suspicious of those percent-of-GDP numbers repeated endlessly in discussions like this?
In some cases, I'd guess the numbers come from the fox guarding the hen house. Sorta like Soviet economic growth numbers. In the US case, I'd guess the numbers are maximized because someone thinks the real numbers are "too high", whatever the real numbers may be.
And "life expectancy" numbers for different modern-world areas? They look like statistical bobble.
"Europe or africa"
DragonLord, that got me wondering. Cape C and Disney World are at the latitude of southern Morocco. Miami is across from the country of Western Sahara.
Never really thought about it, but the eastern hemisphere "Florida" is in the Sahara desert. "Europe" is across from Canada, not the US. Weird geographic things. Like how South America is south of Greenland, not North America. :)
Pete2 - I like your suggestion.
Nifty - "cube radio?" I see a gob of "cube radio", but didn't see one that works by turning it. Link? Might be handy for 89 yr Mother.
What's left out of these ageing-boomer discussions is: Sure, you're 69, you're not going to be able to pull the hours of 30, 40 years before. But does that mean you can't do any useful work? I think not. Well, I hope not, anyway.
A ton of water is a cubic meter. So 20 gigatons would a line of 20 billion such cubes of water. Or, in one, big cube, the 3rd root of 20,000,000,000, 2714 meters on each side. A cubic mile or two. That's a lot to drink, but adding it to the Earth's oceans would be like adding a rain drop to a small lake.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
