I've read collections much better researched that this rat bag. I could pull 100 stupider things out of my own sock drawer.
The one I remember from a book I read was an airline that offered a promotion to frequent business flyers allowing them to take a companion for free.
Then to follow up on this spiff, the decided to send letters to the residences of the business men who had taken advantage of the offer. It was a reminder many never forgot. These letters were mostly opened up by spouses(PC) who had never gone anywhere.
Marketrons just don't know how to quit, nor do they recognize the vices they facilitate.
Most of the freaking cosmos is well outside our light cone (age of universe we measure from here times mass of universe fresh off the press).
Maybe we *should* focus more of our energies with the cone of cause and effect that matters in the here and now.
Suppose we do find the "answer to the cosmos". What are we going to do next, push the starter button?
I'm sure the universe would be very impressed and all, but with a clock cycle of about one femtohertz (the speed at which the universe agrees with itself), I'm not sticking around for a pat on the back.
I hate to confirm your self diagnosis, but I have sad news to bear.
If you wish to use memory mapped IO to your file system, which has some good technical properties, you need a pointer with an address range *at least* as large as the largest possible file you might need to access, and preferably as large as the largest file system you intend to mount.
Addressibility and physical storage are somewhat orthogonal. (In theory, there is no difference between theory and practice, in practice there is.)
On a machine with 10G of memory, there is no reason for a process to use 64-bit pointers if the process doesn't require more than 32 bits of addressibility. If you look at Apache in the standard threading model, every request is managed by a different process. I doubt you need 64-bit pointers for *each* PHP instance, regardless of how much physical memory the machine contains.
On the other hand, you might be doing some kind of video stream manipulation on a 10GB file using a machine with only 1GB of physical RAM. You would require the use of 64-bit addressibility for this task if you choose the memory mapped IO model.
So yes, you are retarded, but it could be cured by thinking before you type (the post does mention memory mapped IO). There: ten simple words of advice that should apply to 2^33 members of the slashdot community.
Oh my god, I spent $5000 on a humongous television and now my expensive television is being used to display commercials I don't want to watch and I can't do anything about it? What a terrible abuse of my cheeseball spending priorities.
Hint: change the channel, or turn the damn thing off
Drinking that much coffee is not good for the body. I learned the hard way: wore out my adrenal system.
It's not so easy to quit as some people suggest.
First time I quit cold turkey, spent three days in bed with wracking headaches and no appetite for food. The headaches became less severe after three days, but my body was not yet at peace. Suffered unproductively for the better part of two weeks and then started drinking coffee again to get on with my life. But a lot less than before.
Another iteration of quiting and unquiting got me down to about two or three large cups a day.
Then I had a prescription medication that interacted badly with caffeine and I had to quit again. Still had the headaches for several days, but this time my life didn't stall completely. A month later I still couldn't function at full intensity, so I started drinking one cup each morning.
At one cup of moderately strong coffee, I can quit anytime without a headache. At 1.5 cups per day, missing a day is risky. At 2 cups per day, I'm fully addicted. It can vary over a wide range from one person to another.
After many hard fought battles, I figure it takes the best part of three months for the body to fully adjust to a different caffeince consumption level. People forget that coffee has hundreds of other alkaloids, not just caffeine. Decaf coffee affects cognitive structure (not in a good way) without causing the same vascular effects.
Now I stick to about one cup a day, the level where I know I'm not addicted. Can miss a day with only a little blah to deal with.
Tea never worked at all as a caffeine substitute for me, nor do any of the colas. It's not just the caffeine you have to live without.
The best trick I learned was to change my brewing methods.
First, use a high quality dark roast with intense flavour. Dark roast has less caffeine, because some of the caffeine is destroyed in the roaasting process. If the roast is good, I find I'm less tempted to cheat on the ratio.
Don't use a French press. I love the body of a French press, but it comes at the price of extracting in triplicate. I switched to drip, which was (un)depressing at first, but I got used to it.
Grind your own beans. Some roasts can be ground a lot finer without losing flavor or becoming bitter. A fine grind with a quick brew cycle will extract more flavour relative to the amount of caffeine. Don't ask me about the physics, I don't understand it either.
Brew in smaller batches. I used to use brew length as an indicator for the quality of a roast. If the roast can be extracted in a French press for more than four minutes without becoming nasty, the roast is really good. With a French press, the coffee tastes better if you pour from about ten inches above the top of the Bodum in a slow drizzle. I could never figure out why this worked, but then I learned that this is just enough time for the water temp. to drop below 200 degrees. Water right at the boiling point does something nasty to coffee beans. But, oh, I was saying don't use a French press only the memories are too good.
Even with a drip, the extraction cycle is important. The problem is that if the coffee tastes like crap, my first instinct is to fix the problem by tossing twice as many grounds in the filter basket.
Drip coffee makers don't scale: the length of the extraction cycle varies with the amount of water processed. Shorter extraction cycles are better for getting good coffee with less caffeine.
For my small Braun drip, anything over half a pot creates difficulties with balance. I drew a black line at the fill level which produces an optimum exrtraction cycle: it works out to two 10 ounce cups.
Even with the black line, I had a constant battle with an expanding miniscus. Some days I could make that miniscus so large, I could squeeze an entire third mug out of the deal.
The stroke of genius was to throw the caraffe away. Now I brew my coffee
I'll take the factless drivel from the Inquirer over the factual contents of quarterly balance sheet any day.
At least the factless drivel from the Inquirer is spread more or less equally in all directions.
The people I know who complain the most about the Inquirer are the types who rely on quoting others to support their arguments, instead of thinking for themselves.
Yes indeed, the Inquirer is a useless bag of filth if you don't think for yourself.
This particular article began well, then spiralled insanely out of control. The guy must have downed too many jujubes while he was scrambling to make quota.
I also don't believe that Microsoft's bag of dirty tricks has failed to work. It has failed to work *yet*. I still believe they are plotting something diabolical in the IP sphere.
Five years from now, MS might own a $40 billion patent portfolio. Where is open source then? Before we write MS off, we need to remember that Rome wasn't burned in a day.
Every Debian user soon finds out what it feels like to be years behind the curve. Hell, I've still got potato running in the basement because I'm afraid to mess with my software RAID configuration. On my woody desktop I've been backporting and unbackporting on a regular basis lately trying to get a few recent packages to peacefully coexist, and it still mostly hangs together.
Applying the term "good working order" to Win98 boggles my mind. At the peak of sanitation, it's still too awful to use.
Jobs' comment that "a legal alternative to stealing music hadn't been invented until six months ago" takes "arguable" to new heights.
Despite the nose bleed, this article taught me something: the secret of Jobs' marketing genius is to equate instant gratification to a constitutional entitlement.
First he names the company after something you stick in your mouth, and twenty years later he is still trying to compel people to lick the visuals. It's a view of the American constitution through an infant psyche.
There are portions of our wetware that have been rendered into a format that can computed with digital computers, and the results match human competence (accoustic processing neurons, Kutzweil talks about this at length himself).
If the analog signal you are referring to is the number of neurotransmitter molecules to cross a synapse, I think the computer can represent that quantity quite well, using plain old integers.
Kurzweil is as nutty as Penrose, but he is nutty with a much better grip on the facts that matter. Kurzweil is more interesting to me being wrong about the right facts than Penrose is being right about the wrong facts.
The scary thing about Kurzweil is that he might not be as wrong as one would hope or expect.
Lesson to be learned: don't let a man who is simultaneously as smart and stupid as Roger Penrose take control over the definition of philosophical terms such as "consciousness" or he'll convert common sense into a Klein bottle.
The universe has a lot of structure, at many levels, which is hard to explain. What's more, Turing computable information systems are capable of generating a lot of structure we can't explain either, by any definition of "explain" that renders the nature of consciousness a useful question.
It costs money to test, identify, locate, describe, and report bugs.
Reporting bugs in MS products to MS before releasing the bug report to the public amounts to working for MS for free, while MS makes huge profits foisting substandard, crappy products on their customers in the first place.
Until MS demonstrates a proper respect for their end customers, their privacy and their personal data, and ceases to expose their customers through entirely unnecessary software defects, I see no reason why MS or the reputation of its products should benefit from unpaid private disclosure.
If I believed MS had made a fair calculation up front about the balance of features vs the risks devolved to their user base, I wouldn't take this position. There has to be a feedback loop somewhere in the system to punish MS for the consequences for the unfair balance they chose to pursue.
Arguments that amount to this don't impress me: "millions of people use MS products, and these people are all being held hostage by possible exploits of defects created by MS, therefore it's the messenger's fault".
When MS offers a $10K bounty for every verified bug reported ethically by a bug researcher, and fully discloses the number of bounties paid, and for which bug fixes, then I will believe that MS has regained a moral position to demand this concession from the bug research community.
My only motivation in discovering and reporting a bug in IE would be to help create a corrective force to end the business practices which created this situation in the first place. How does offering my services to MS for free accomplish that goal?
Gauss is the worst counter-example you could have picked. He filled many journals in his late teen years when his mind was exploding, and didn't manage to publish all the results therein over the rest of his career. I think he had the foundation in place for most of what Abel "discovered" 50 to 100 years later (IIRC when Abel lived).
The "Men of Mathematics" by E.T. Bell (IIRC) has much of this story, as do most of books about the history of mathematics.
I must be a brick short of a full load. This thread has turned into a sceed fiesta. Where is the scientific evidence that sighting a naked women turns a young boy into a mass murderer, or a pimp, or a dealer? Is there is some basic social knowledge that is mysteriously lacking from my genetic repetoire? Where does this knowledge come from, and why don't I have it?
Maybe this goes right back to genesis. From the time we first discover sex, we feel guilty about it. Then we all rush around as adults to protect our children from becoming the horrible depraved adults our generation has become, and our parents generation, and all the begatters to the beginning of time.
No wait, they have done research on masturbation, the majority of adult males have an urge to clean the pipes the day before being reunited with their lust interest. According to research, it improves semen quality to call up a fresh set of reserves. Who would have ever guessed that sexual reflexes and instincts were associated with reproductive fitness?
I don't children in this society need to be told much about the birds and the bees. Perhaps it would be better to update our schools with a dopamine education class. Here is dopamine: human folly boiled down to molecular dimensions. Vonnegut wrote a novel about a Martian army with transmitters embedded in their minds to control them (Sirens of Titan). That seems like an unnecessary redundancy. We are already under the control of a hostile molecular force. Perhaps there is a better way to educate children that scrubbing internet connections and pretending we have not yet discovered the molecular snake.
The O(N) shadow statement is a sufficient statement of O(N) behaviour for the big O pedants. I looked at the graph, and I vote we keep the wording as it was.
O notation is overrated. Sorting is always described as O(N*log N), but for any practical architecture using a radix sort with L1/L2 cache locality, replace log N with the constant factor of 3 or 4. A million cache local buckets can radix sort 10^30 elements in 3 log N time.
Using all of main memory as your bucket store, I'd guess you could sort every proton in the known universe in 8 passes. So what exactly is that log N term trying to tell us?
Wake up and smell the bacon, people. The techno-utopianism of Wired when it was boosting the dotcom era into orbit has proven itself a poor match with human nature on all fronts.
The benificient fathers of the internet made two horrendous design decisions concerning the final destination of a global internetwork: excessively strong anonimity and a near zero cost for dumping pollution into public media.
Privacy, openness, spam-free: pick any two.
For anyone who looked into ECC yesterday, you might have noticed that RSA has ideal properties for preventing some of this mess: expensive to sign a certificate, cheap to verify, and the ratio becomes worse as you scale up.
If every spam artifact was signed with an anonymous RSA cert (anyone could make as many of these as they wish), as soon as one spam is confirmed, every other post signed by the known-spam cert could be instantly revoked.
This would force the spammers to create a new anonymous cert for every spam instance. Yet with RSA certs, the computational cost to generate a cert is vastly greater than the cost to verify the cert.
As an added step, the cert could require the IP address of both endpoints to be embedded inside (the server would reflect back the IP source address it sees, and then ask for an anonymous cert to be generated at a desired RSA key size).
We won't have to damage anonymity very much to vastly increase the cost of dumping pollution.
In this respect, weblogs would be a good place to start. This is a relatively new technology that could be retrofitted at one percent of the cost of a global e-mail infrastructure upgrade. It really doesn't matter if you inconvience a few bloggers working out the kinks, these people have not much useful to do in any case.
I personally think that broadcast media is where the 20'th century made a wrong turn. You see it in the quality of the programming produced, the consumeristic culture that buys anything a celebrity endorses, the power of organizations such as RIAA, the rise of golf as a spectator sport (what better way to fill up all those empty broadcast hours?), and the way an average person can't name a single active ingredient in any soap product they use (unless you count adultery and denial).
I think at best the personal use rights on broadcast content is a Faustian bargain. I think of the broadcast bit as meaning "don't consume this". It should be regarded as a non-so-subtle hint that for the next century we should explore non-broadcast alternatives. It doesn't take a very close look at the kinds of people being elected in America to see that broadcast media was a huge mistake to begin with.
With his superior knowledge, he should contribute to the tuning documentation so that everyone can benefit. No wait--that's already been done.
I've seen so many benchmarks devote many boring pages of text to the hardware and software configuration used. For once I'd like to see a benchmark that discloses what standard documentation the benchmarker bothered to read before pontificating on his results.
If the system clearly documents a limit of 4000 foofroms and then the person testing the system exceeds this limitation, it's the person using the system who has failed, not the system itself. Correct behaviour of a system is to function as documented. If you can't be bothered to read that documentation, you aren't qualified to apply the word failure to anything other than your own actions.
I've never run across this sort of FreeBSD user. And when I have heard people constrasting FreeBSD with Linux, the subject usually turns on the impression that the FreeBSD development path is somewhat more controlled than the pell mell evolution of Linux.
I've already sensed the message from Linus that post 2.6 kernel, the pell mell phase of Linux kernel evolution will end. He thinks future work will largely be driven by the needs of supporting advanced hardware.
The main difference I've experienced is that I feel comfortable managing FreeBSD without consulting word on the street.
With Linux, I find myself checking google a lot more often to make sure I'm not wasting time configuring some subsystem already slated for obsolescence.
I wonder how preference between FreeBSD and Linux would correspond to people with and without IM clients. (What, there are people who don't have an IM client running 24/7? How do they live!)
The IM users I know are the kind of people who take the Linux churn completely for granted.
At least I found one three word expression in that article I could agree with.
In fact, I had an original fat Mac once and I gradually grew to hate the interface. Am I the only person around who cares more about the content area of what I'm trying to work on than the glitzy frame around it?
The Mac window model gives you all kinds of controls over the window *frame*. What I wanted was a quick way to indicate "I need to see this patch of text here and that patch of text there on the screen at the same time." I wanted to be able to set up several of these viewing relationships concurrently and toggle among the various exposures as my workflow dictated, with a single keystroke if possible. I certainly didn't want an efficient way to spend my day as a window management tugboat.
I still don't have the GUI that does this the way I want it, but at least I now have two large monitors. I still expend far to many mouse strokes manipulating windows so that I can gain access to window manipulation controls. There's something wrong with that, which we would soon figure out if it wasn't so damn pretty.
Beauty is in the eye of the beholder. In this beholder, the visual inputs connect directly into a highly impatient "getting something done" neuron cluster, which then connects directly to twitchy "can't type fast enough" fingertips.
Sure we all care about aesthetics, but my aesthetics are kinematic aesthetics (eyeball to fingertip) that aren't satisfied by dancing gobs of fruit gelatin.
And no, I don't need mood enhancers from my window management system. I can be a cranky bastard with no help at all. Put that in your fat ass scroll bar and smoke it.
Soekris will soon have a new PCI crypto accelerator, the VPNL401 Encryption at 400Mbs. That ought to be enough encrypted bandwidth to map every prospective Starbuck's franchise in the Virgo Galactic Cluster.
Even without hardware crypto, any modern 1GHz CPU can saturate a fat pipe using AES or Blowfish as the cypher algorithm. Quit blaming sftp and find a way to make sftp work properly.
I've never flipped this keyboard over until just now. My windows machine has an IBM keyboard with 1984 stamped on the bottom. Did they already F1-F12 in three sets of four along the top, or is that just the copyright date? The rest of the label is impossible to read.
That said, I think including keyboards in this survey is tantamount to including desk lights and desk chairs. Good grief, the first Selectric typewriter dates back to 1961. By 1984 IBM already had 23 years of experience to draw upon in keyboard design.
How about a real example of protecting the family jewels the technology of yesteryear?
Three days ago I bought a used Asus P2B-S with Pentium II/300 for an Amanda backup server. I threw it into a relatively expensive case (twin 80mm drive fans, 120mm rear exhaust fan, 80mm blowhole fan) and slapped in a decent Enermax 365 power supply. The tiny CPU fan on that giant heatsink could fail entirely and the machine wouldn't skip a beat. 45 loonies for the mobo and CPU, 170 loonies for huge case plus power supply. Installed FreeBSD 4.9 RC2 onto an old 4G SCSI server drive and some giant refurbed IDE drives as my Amanda backup spool. The box is surprisingly snappy. The brand new Intel EEpro network card cost me more than the mobo and CPU. I'm playing around with an old Archive Python tape drive as a way to learn how to configure Amanada. Soon I'll add a modern tape drive or a modern DVD burner, or both. Call me crazy, but I've always liked the steel/watt ratio of those old Pentium II slot-1 cartridges.
Closest I can date the P2B-S is August 1998. The point here is that this mobo is five years old, yet I was still willing to wrap what was basically a new machine around it to perform an important function.
But oh yes, I agonized a lot more about continuing to use my circa 1984 IBM keyboard. I really lost a lot of sleep over that one, sacrificing two whole decades of exponential improvements in keysize reduction.
Slashdot Mirror
I've read collections much better researched that this rat bag. I could pull 100 stupider things out of my own sock drawer.
The one I remember from a book I read was an airline that offered a promotion to frequent business flyers allowing them to take a companion for free.
Then to follow up on this spiff, the decided to send letters to the residences of the business men who had taken advantage of the offer. It was a reminder many never forgot. These letters were mostly opened up by spouses(PC) who had never gone anywhere.
Marketrons just don't know how to quit, nor do they recognize the vices they facilitate.
Most of the freaking cosmos is well outside our light cone (age of universe we measure from here times mass of universe fresh off the press).
Maybe we *should* focus more of our energies with the cone of cause and effect that matters in the here and now.
Suppose we do find the "answer to the cosmos". What are we going to do next, push the starter button?
I'm sure the universe would be very impressed and all, but with a clock cycle of about one femtohertz (the speed at which the universe agrees with itself), I'm not sticking around for a pat on the back.
I hate to confirm your self diagnosis, but I have sad news to bear.
If you wish to use memory mapped IO to your file system, which has some good technical properties, you need a pointer with an address range *at least* as large as the largest possible file you might need to access, and preferably as large as the largest file system you intend to mount.
Addressibility and physical storage are somewhat orthogonal. (In theory, there is no difference between theory and practice, in practice there is.)
On a machine with 10G of memory, there is no reason for a process to use 64-bit pointers if the process doesn't require more than 32 bits of addressibility. If you look at Apache in the standard threading model, every request is managed by a different process. I doubt you need 64-bit pointers for *each* PHP instance, regardless of how much physical memory the machine contains.
On the other hand, you might be doing some kind of video stream manipulation on a 10GB file using a machine with only 1GB of physical RAM. You would require the use of 64-bit addressibility for this task if you choose the memory mapped IO model.
So yes, you are retarded, but it could be cured by thinking before you type (the post does mention memory mapped IO). There: ten simple words of advice that should apply to 2^33 members of the slashdot community.
Oh my god, I spent $5000 on a humongous television and now my expensive television is being used to display commercials I don't want to watch and I can't do anything about it? What a terrible abuse of my cheeseball spending priorities.
Hint: change the channel, or turn the damn thing off
Spend ten billion dollars to go someplace with no real plan for what to do once you get there.
Drinking that much coffee is not good for the body. I learned the hard way: wore out my adrenal system.
It's not so easy to quit as some people suggest.
First time I quit cold turkey, spent three days in bed with wracking headaches and no appetite for food. The headaches became less severe after three days, but my body was not yet at peace. Suffered unproductively for the better part of two weeks and then started drinking coffee again to get on with my life. But a lot less than before.
Another iteration of quiting and unquiting got me down to about two or three large cups a day.
Then I had a prescription medication that interacted badly with caffeine and I had to quit again. Still had the headaches for several days, but this time my life didn't stall completely. A month later I still couldn't function at full intensity, so I started drinking one cup each morning.
At one cup of moderately strong coffee, I can quit anytime without a headache. At 1.5 cups per day, missing a day is risky. At 2 cups per day, I'm fully addicted. It can vary over a wide range from one person to another.
After many hard fought battles, I figure it takes the best part of three months for the body to fully adjust to a different caffeince consumption level. People forget that coffee has hundreds of other alkaloids, not just caffeine. Decaf coffee affects cognitive structure (not in a good way) without causing the same vascular effects.
Now I stick to about one cup a day, the level where I know I'm not addicted. Can miss a day with only a little blah to deal with.
Tea never worked at all as a caffeine substitute for me, nor do any of the colas. It's not just the caffeine you have to live without.
The best trick I learned was to change my brewing methods.
First, use a high quality dark roast with intense flavour. Dark roast has less caffeine, because some of the caffeine is destroyed in the roaasting process. If the roast is good, I find I'm less tempted to cheat on the ratio.
Don't use a French press. I love the body of a French press, but it comes at the price of extracting in triplicate. I switched to drip, which was (un)depressing at first, but I got used to it.
Grind your own beans. Some roasts can be ground a lot finer without losing flavor or becoming bitter. A fine grind with a quick brew cycle will extract more flavour relative to the amount of caffeine. Don't ask me about the physics, I don't understand it either.
Brew in smaller batches. I used to use brew length as an indicator for the quality of a roast. If the roast can be extracted in a French press for more than four minutes without becoming nasty, the roast is really good. With a French press, the coffee tastes better if you pour from about ten inches above the top of the Bodum in a slow drizzle. I could never figure out why this worked, but then I learned that this is just enough time for the water temp. to drop below 200 degrees. Water right at the boiling point does something nasty to coffee beans. But, oh, I was saying don't use a French press only the memories are too good.
Even with a drip, the extraction cycle is important. The problem is that if the coffee tastes like crap, my first instinct is to fix the problem by tossing twice as many grounds in the filter basket.
Drip coffee makers don't scale: the length of the extraction cycle varies with the amount of water processed. Shorter extraction cycles are better for getting good coffee with less caffeine.
For my small Braun drip, anything over half a pot creates difficulties with balance. I drew a black line at the fill level which produces an optimum exrtraction cycle: it works out to two 10 ounce cups.
Even with the black line, I had a constant battle with an expanding miniscus. Some days I could make that miniscus so large, I could squeeze an entire third mug out of the deal.
The stroke of genius was to throw the caraffe away. Now I brew my coffee
I'll take the factless drivel from the Inquirer over the factual contents of quarterly balance sheet any day.
At least the factless drivel from the Inquirer is spread more or less equally in all directions.
The people I know who complain the most about the Inquirer are the types who rely on quoting others to support their arguments, instead of thinking for themselves.
Yes indeed, the Inquirer is a useless bag of filth if you don't think for yourself.
This particular article began well, then spiralled insanely out of control. The guy must have downed too many jujubes while he was scrambling to make quota.
I also don't believe that Microsoft's bag of dirty tricks has failed to work. It has failed to work *yet*. I still believe they are plotting something diabolical in the IP sphere.
Five years from now, MS might own a $40 billion patent portfolio. Where is open source then? Before we write MS off, we need to remember that Rome wasn't burned in a day.
Because the world is full of stupid shoppers.
Scratch Dan Heatly from the list of people looking forward to the 1000 horsepower car. Did anyone else see his face on ESPN today?
I guess tjstork is too busy oozing to know what's going on around him.
Every Debian user soon finds out what it feels like to be years behind the curve. Hell, I've still got potato running in the basement because I'm afraid to mess with my software RAID configuration. On my woody desktop I've been backporting and unbackporting on a regular basis lately trying to get a few recent packages to peacefully coexist, and it still mostly hangs together.
Applying the term "good working order" to Win98 boggles my mind. At the peak of sanitation, it's still too awful to use.
Jobs' comment that "a legal alternative to stealing music hadn't been invented until six months ago" takes "arguable" to new heights.
Despite the nose bleed, this article taught me something: the secret of Jobs' marketing genius is to equate instant gratification to a constitutional entitlement.
First he names the company after something you stick in your mouth, and twenty years later he is still trying to compel people to lick the visuals. It's a view of the American constitution through an infant psyche.
There are portions of our wetware that have been rendered into a format that can computed with digital computers, and the results match human competence (accoustic processing neurons, Kutzweil talks about this at length himself).
If the analog signal you are referring to is the number of neurotransmitter molecules to cross a synapse, I think the computer can represent that quantity quite well, using plain old integers.
Kurzweil is as nutty as Penrose, but he is nutty with a much better grip on the facts that matter. Kurzweil is more interesting to me being wrong about the right facts than Penrose is being right about the wrong facts.
The scary thing about Kurzweil is that he might not be as wrong as one would hope or expect.
Lesson to be learned: don't let a man who is simultaneously as smart and stupid as Roger Penrose take control over the definition of philosophical terms such as "consciousness" or he'll convert common sense into a Klein bottle.
The universe has a lot of structure, at many levels, which is hard to explain. What's more, Turing computable information systems are capable of generating a lot of structure we can't explain either, by any definition of "explain" that renders the nature of consciousness a useful question.
How did 21.0 mebibits turn into six megabytes? I think he meant mugabytes.
Offtopic, Inflammatory, Inappropriate, Illegal, or Offensive comments might be moderated, and messages that are too short might be downrated.
It costs money to test, identify, locate, describe, and report bugs.
Reporting bugs in MS products to MS before releasing the bug report to the public amounts to working for MS for free, while MS makes huge profits foisting substandard, crappy products on their customers in the first place.
Until MS demonstrates a proper respect for their end customers, their privacy and their personal data, and ceases to expose their customers through entirely unnecessary software defects, I see no reason why MS or the reputation of its products should benefit from unpaid private disclosure.
If I believed MS had made a fair calculation up front about the balance of features vs the risks devolved to their user base, I wouldn't take this position. There has to be a feedback loop somewhere in the system to punish MS for the consequences for the unfair balance they chose to pursue.
Arguments that amount to this don't impress me: "millions of people use MS products, and these people are all being held hostage by possible exploits of defects created by MS, therefore it's the messenger's fault".
When MS offers a $10K bounty for every verified bug reported ethically by a bug researcher, and fully discloses the number of bounties paid, and for which bug fixes, then I will believe that MS has regained a moral position to demand this concession from the bug research community.
My only motivation in discovering and reporting a bug in IE would be to help create a corrective force to end the business practices which created this situation in the first place. How does offering my services to MS for free accomplish that goal?
Gauss is the worst counter-example you could have picked. He filled many journals in his late teen years when his mind was exploding, and didn't manage to publish all the results therein over the rest of his career. I think he had the foundation in place for most of what Abel "discovered" 50 to 100 years later (IIRC when Abel lived).
The "Men of Mathematics" by E.T. Bell (IIRC) has much of this story, as do most of books about the history of mathematics.
I must be a brick short of a full load. This thread has turned into a sceed fiesta. Where is the scientific evidence that sighting a naked women turns a young boy into a mass murderer, or a pimp, or a dealer? Is there is some basic social knowledge that is mysteriously lacking from my genetic repetoire? Where does this knowledge come from, and why don't I have it?
Maybe this goes right back to genesis. From the time we first discover sex, we feel guilty about it. Then we all rush around as adults to protect our children from becoming the horrible depraved adults our generation has become, and our parents generation, and all the begatters to the beginning of time.
No wait, they have done research on masturbation, the majority of adult males have an urge to clean the pipes the day before being reunited with their lust interest. According to research, it improves semen quality to call up a fresh set of reserves. Who would have ever guessed that sexual reflexes and instincts were associated with reproductive fitness?
I don't children in this society need to be told much about the birds and the bees. Perhaps it would be better to update our schools with a dopamine education class. Here is dopamine: human folly boiled down to molecular dimensions. Vonnegut wrote a novel about a Martian army with transmitters embedded in their minds to control them (Sirens of Titan). That seems like an unnecessary redundancy. We are already under the control of a hostile molecular force. Perhaps there is a better way to educate children that scrubbing internet connections and pretending we have not yet discovered the molecular snake.
The O(N) shadow statement is a sufficient statement of O(N) behaviour for the big O pedants. I looked at the graph, and I vote we keep the wording as it was.
O notation is overrated. Sorting is always described as O(N*log N), but for any practical architecture using a radix sort with L1/L2 cache locality, replace log N with the constant factor of 3 or 4. A million cache local buckets can radix sort 10^30 elements in 3 log N time.
Using all of main memory as your bucket store, I'd guess you could sort every proton in the known universe in 8 passes. So what exactly is that log N term trying to tell us?
Wake up and smell the bacon, people. The techno-utopianism of Wired when it was boosting the dotcom era into orbit has proven itself a poor match with human nature on all fronts.
The benificient fathers of the internet made two horrendous design decisions concerning the final destination of a global internetwork: excessively strong anonimity and a near zero cost for dumping pollution into public media.
Privacy, openness, spam-free: pick any two.
For anyone who looked into ECC yesterday, you might have noticed that RSA has ideal properties for preventing some of this mess: expensive to sign a certificate, cheap to verify, and the ratio becomes worse as you scale up.
If every spam artifact was signed with an anonymous RSA cert (anyone could make as many of these as they wish), as soon as one spam is confirmed, every other post signed by the known-spam cert could be instantly revoked.
This would force the spammers to create a new anonymous cert for every spam instance. Yet with RSA certs, the computational cost to generate a cert is vastly greater than the cost to verify the cert.
As an added step, the cert could require the IP address of both endpoints to be embedded inside (the server would reflect back the IP source address it sees, and then ask for an anonymous cert to be generated at a desired RSA key size).
We won't have to damage anonymity very much to vastly increase the cost of dumping pollution.
In this respect, weblogs would be a good place to start. This is a relatively new technology that could be retrofitted at one percent of the cost of a global e-mail infrastructure upgrade. It really doesn't matter if you inconvience a few bloggers working out the kinks, these people have not much useful to do in any case.
I personally think that broadcast media is where the 20'th century made a wrong turn. You see it in the quality of the programming produced, the consumeristic culture that buys anything a celebrity endorses, the power of organizations such as RIAA, the rise of golf as a spectator sport (what better way to fill up all those empty broadcast hours?), and the way an average person can't name a single active ingredient in any soap product they use (unless you count adultery and denial).
I think at best the personal use rights on broadcast content is a Faustian bargain. I think of the broadcast bit as meaning "don't consume this". It should be regarded as a non-so-subtle hint that for the next century we should explore non-broadcast alternatives. It doesn't take a very close look at the kinds of people being elected in America to see that broadcast media was a huge mistake to begin with.
With his superior knowledge, he should contribute to the tuning documentation so that everyone can benefit. No wait--that's already been done.
I've seen so many benchmarks devote many boring pages of text to the hardware and software configuration used. For once I'd like to see a benchmark that discloses what standard documentation the benchmarker bothered to read before pontificating on his results.
If the system clearly documents a limit of 4000 foofroms and then the person testing the system exceeds this limitation, it's the person using the system who has failed, not the system itself. Correct behaviour of a system is to function as documented. If you can't be bothered to read that documentation, you aren't qualified to apply the word failure to anything other than your own actions.
A strawman strawpoll. How interrrresting.
I've never run across this sort of FreeBSD user. And when I have heard people constrasting FreeBSD with Linux, the subject usually turns on the impression that the FreeBSD development path is somewhat more controlled than the pell mell evolution of Linux.
I've already sensed the message from Linus that post 2.6 kernel, the pell mell phase of Linux kernel evolution will end. He thinks future work will largely be driven by the needs of supporting advanced hardware.
The main difference I've experienced is that I feel comfortable managing FreeBSD without consulting word on the street.
With Linux, I find myself checking google a lot more often to make sure I'm not wasting time configuring some subsystem already slated for obsolescence.
I wonder how preference between FreeBSD and Linux would correspond to people with and without IM clients. (What, there are people who don't have an IM client running 24/7? How do they live!)
The IM users I know are the kind of people who take the Linux churn completely for granted.
What is it about this comment that gives me a tremendous urge to turn on the TV and watch a solid hour of M*A*S*H reruns?
At least I found one three word expression in that article I could agree with.
In fact, I had an original fat Mac once and I gradually grew to hate the interface. Am I the only person around who cares more about the content area of what I'm trying to work on than the glitzy frame around it?
The Mac window model gives you all kinds of controls over the window *frame*. What I wanted was a quick way to indicate "I need to see this patch of text here and that patch of text there on the screen at the same time." I wanted to be able to set up several of these viewing relationships concurrently and toggle among the various exposures as my workflow dictated, with a single keystroke if possible. I certainly didn't want an efficient way to spend my day as a window management tugboat.
I still don't have the GUI that does this the way I want it, but at least I now have two large monitors. I still expend far to many mouse strokes manipulating windows so that I can gain access to window manipulation controls. There's something wrong with that, which we would soon figure out if it wasn't so damn pretty.
Beauty is in the eye of the beholder. In this beholder, the visual inputs connect directly into a highly impatient "getting something done" neuron cluster, which then connects directly to twitchy "can't type fast enough" fingertips.
Sure we all care about aesthetics, but my aesthetics are kinematic aesthetics (eyeball to fingertip) that aren't satisfied by dancing gobs of fruit gelatin.
And no, I don't need mood enhancers from my window management system. I can be a cranky bastard with no help at all. Put that in your fat ass scroll bar and smoke it.
Even without hardware crypto, any modern 1GHz CPU can saturate a fat pipe using AES or Blowfish as the cypher algorithm. Quit blaming sftp and find a way to make sftp work properly.
I've never flipped this keyboard over until just now. My windows machine has an IBM keyboard with 1984 stamped on the bottom. Did they already F1-F12 in three sets of four along the top, or is that just the copyright date? The rest of the label is impossible to read.
That said, I think including keyboards in this survey is tantamount to including desk lights and desk chairs. Good grief, the first Selectric typewriter dates back to 1961. By 1984 IBM already had 23 years of experience to draw upon in keyboard design.
How about a real example of protecting the family jewels the technology of yesteryear?
Three days ago I bought a used Asus P2B-S with Pentium II/300 for an Amanda backup server. I threw it into a relatively expensive case (twin 80mm drive fans, 120mm rear exhaust fan, 80mm blowhole fan) and slapped in a decent Enermax 365 power supply. The tiny CPU fan on that giant heatsink could fail entirely and the machine wouldn't skip a beat. 45 loonies for the mobo and CPU, 170 loonies for huge case plus power supply. Installed FreeBSD 4.9 RC2 onto an old 4G SCSI server drive and some giant refurbed IDE drives as my Amanda backup spool. The box is surprisingly snappy. The brand new Intel EEpro network card cost me more than the mobo and CPU. I'm playing around with an old Archive Python tape drive as a way to learn how to configure Amanada. Soon I'll add a modern tape drive or a modern DVD burner, or both. Call me crazy, but I've always liked the steel/watt ratio of those old Pentium II slot-1 cartridges.
Closest I can date the P2B-S is August 1998. The point here is that this mobo is five years old, yet I was still willing to wrap what was basically a new machine around it to perform an important function.
But oh yes, I agonized a lot more about continuing to use my circa 1984 IBM keyboard. I really lost a lot of sleep over that one, sacrificing two whole decades of exponential improvements in keysize reduction.