This summary would leave any biologist apoplectic for its flaws:
1. Sex IS sharing. It's the ultimate cross-fertilization (literally). Almost all organisms, including humans, openly and enthusiastically share DNA via this mechanism. 2. Horizontal gene flow is terribly terribly limiting in its utility. Once organism becomes more complex, you can't plug-and-play like you can with a bacteria. 3. Horizontal gene flow does not foster rapid evolution in the same way that sex does. Picking up snippets and fragments from another organism is not as powerful as cross-over in sex (which does a far far better job of doing a controlled recombination of complete plans) 3. No organism in the world can resist "sharing its genome." If pirating the DNA of others was really that great an idea, then the human digestive track would contain tools for pulling DNA out of hamburger. It really would not take much cellular machinery to engulf a target cell, deconstruct it, and co-opt its DNA. The fact that horizontal gene transfer doesn't occur outside of simple organism should be an strong evidence of its limitations.
As much as I enjoy Freeman Dyson, he really lost me on this one.
One key problem with identity is non-uniqueness of names -- there can be several people named Darren Waters. Disambiguating these is nontrivial because it requires other identifiers (e.g., age, hometown, address). Often times the searcher doesn't know anything else about the target. Add spelling errors and things become even more confusing.
A second "which 'Darren Waters'" problem is role-segmentation. I, and am sure many, have multiple online persona reflect different interests, roles, and communities. Depending on the context, I'm a geek, engineer, photographer, stock trader, businessman, etc. Any meaningful search for me would need to specify which version of me they were looking for.
Searching for a person implies both uniquely defining the person and defining which aspect of the person one is looking for because "which 'Darren Waters'" is a problem with two dimensions of ambiguity.
Although price floors seem like a way to preserve the profits of inefficient retailers, I'm sure that the better retailers will figure out a way around any sort of binding MSRP. These might include:
1. Good service: extended hours, trained employees, better inventories, free shipping, free installation, etc. 2. Bundled goodies: accessories, logoed T-shirts, media, etc. 3. Extended 3rd party warrantees
And if the manufacturer says "thou shalt not bundle free stuff," then the retailer only needs to charge a nominal charge for the "separate" item -- say $0.50 for free delivery, installation, and 5 years of 24 hr in-home repair service.
Price is not the only dimension of competition and some would argue that the internet's focus on price competition is one reason retail service has come to suck so bad. The same transparency that lets current web users find the lowest price will let them find the best retailer in a fixed-price environment.
Thin flexible wings date back to the Boeing B-47. Up until this plane appeared in 1947, planes tended to have thick rigid wing structures. Advances in aeronautics, fluid dynamics, and structure design enabled engineers to create thin flexible swept wings that offered lower drag at high speed without flutter or breakage. The wings of B-47 (and B-52) were so floppy, they needed outrigger wheels to keep the wings from dragging on the ground during landings and take-offs.
Kodak has rediscovered what evolution found millions of years ago -- design a dual system such as the rods and cones of the biological eye. The average human eye has about 120 million sensitive, panchromatic rods and only 6 or 7 million color-sensitive cones (many in the central fovea). The brain merges the limited amounts of color information with the larger volume of B/W image data to paint color into the image that we think we see.
Between ever-better computer image recognition algorithms and cheap offshore labor, captchas are doomed. Morevoer, captcha's don't even solve the actual problem because the goal isn't to distinguish human from nonhuman, but to distinguish spammer from nonspammer. This means we need some mechanism to identify a registrant and be aware of their behavior.
Why don't sites band together, share data on abusive registrants, and require each new registrant to provide "references" in the form of their logins to 3-5 other sites. A person with a normal online life could easily demonstrate a pattern of nonspammy behavior. People with no prior history might be placed on probation (their posts are reviewed and may not contain any link-like data). If a registrant posts spam they temporarily (or permanently) lose their accounts on that site and all connected sites.
At some point in time, the only thing that will work is a system that tracks the identity behind the account, assigns a reputation and ostracizes miscreants.
V2.0 will need to handle 2X or 3X the current device's 100kg payload if it is to sell in the U.S. The growing numbers of "enlarged" Americans that I see using those scooters is horrifying.
Given the amazing plasticity of the young brain, the time to do this is when the kid is really really young. Ideally, a child might most effectively learn to mentally control a cursor/computer interface about the same time they learn to control their fingers and toes. At that age it really will make controlling a computer as effortless as walking or talking.
The time will come when children that didn't get "Baby's First Brain Mouse" in their first few months of life will be at a scholastic disadvantage to those that did.
Not to detract from the joke, but this isn't true on two levels. First, shivering is the response to a low core body temperature, not the "sensing" of it. Something else in the body is senses the drop in body temperature and triggers the shivering. It may be the way that the conscious brain "senses cold" but its not the way that the body does it. Second, this protein is not for detecting low body temperatures, it is for detecting "cold" surfaces and substances. TFA says this protein triggers at 27 C which is far too cold for use in the shivering mechanism (which triggers at about 35 C).
I'd imagine that YouTube will take an increasing role in self-censoring to adhere to local government regulations. Already, YouTube complies with U.S. DMCA take-down requests, so why wouldn't they comply with a Venezuelan government requests to filter the content according to Venezuelan laws?
YouTube's ownership by Google makes it more vulnerable to coercion by governments that can threaten to turn-off Google unless YouTube complies with local content regulations.
I can't help but wonder how they will compensate for parallax between users' eyes, fingertips, and the screen objects. I'd imagine that the display panel will be relatively thick for structural reasons and that most users will be sitting far off the central axis of the table (i.e., the coffee-table scenario). This means that the image on the screen will be displaced with respect to the line sight from eyeball to fingertip. Moreover different people sitting around the table will see different parallax displacements between fingertips and the screen. Without some compensation for this, the device will be frustrating to use because it won't be intuitive as to where to touch the table top to "touch" an underlying display object. People will often touch the screen where they think they should only to find they've grabbed the object next to the one they really wanted.
P.S. Acrylic is a horrible choice of material for the top as it scratches too easily.
I hope this device actually gets a decent battery life. I've been frustrated that all the new PDAs have worse battery life than the early generation machines. The venerable (and discontinued) Psion 5-series got 25-35 hours of use on a pair for AAs and the original Palm Pilot series got maybe 15-20 hours on a AAAs. Every device since those machines has been distinctly inferior (usually getting only a few hours of real use per charge). If the new machine can't last a full day at a conference or a trans-oceanic flight, then I'll just keep using those older (and better) machines.
Codes to access an MP3 is so lame. I thought they were resurrecting the audio data file format used in early home computers that read data to and from a normal cassette tape recorder.
If they're going to go retro with vinyl, they might as well go retro on the computer formats too.
I agree with you that badly designed traffic flow patterns are a problem. Mistimed lights can encourage speeding, create unneeded backups. (Although I don't envy anyone the problem of making all the lights in all four directions pass traffic at the speed limit).
What bothers me are the people that drive 10 mph over the speed limit on a stretch of road with the lights are synced perfectly to the speed limit. These folks have to pile up at the red lights where they become obstacles to people who are driving at a light-sync speed that would have let them sail through every light on green. My point is that if you look up ahead and see the light turn red, you might as well coast to get to the intersection the instant it goes green. And if you commute on the same road day after day, you can learn the speed profile that would get your through light-after-light without ever using the brakes (if it weren't for badly timed lights and the nimrods who can't see red lights beyond the hood of their car and have to hurry to stop at every light).
I'm sure that the country would save quite a bit of gas if more drivers did what their drivers ed teacher said and "got the big picture". I'm amused, and saddened, by the drivers that shoot from red-light to red-light. So often, its plainly obvious from the color of the light and the queue of halted traffic, that there's no way that the light will turn green and traffic will move before these speed demon get to the intersection.
Their average velocity is no higher than any other driver, but they sure do burn a lot of fuel doing it.
By law, faxes in the U.S. must have a "call to remove" number. But I discovered that the number does not work via a little experiment. I called the removal line, entered a different number (a voice line that had never received faxes), and then (within a few days) started receiving fax calls on the voice line. It's just like the email spammers that use victim's unsubscribe notices to signal that they have a live recipient. I'm sure a legal-minded soul could use this behavior to honeypot the faxers, but IANAL.
I've also thought about creating an autodialer script to call the fax removal line and submit every number in the phonebook to it. A simple script could send Hayes commands to a modem to dial the removal line, wait X seconds (or punch "1" to remove or whatever), and then send another dial command to submit bogus removal numbers. Poisoning their DB of faxable numbers would make the return per dialed number much much lower.
I agree with you 100% and did not mean to imply that the goal is not worthy. Being able to search semantically or to pull out just the relevant information would be hugely valuable.
And I'm sure that next generation search engines will create clever ways of detecting and punishing ontology spam (e.g., noting the dissonance between the text and the tags)
Yes, creating a consistent ontology is challenge. But the bigger challenge is the lack of incentive for ontology truthfulness. If this type of search becomes popular, ontology spam and OSEO (Ontology Search Engine Optimization) will become a booming industry.
Given that MID will be based on the McCaslin platform which will include a dual-core Stealey processor running about 600-800MHz, it should run OS X nicely. Both 10.3 and 10.4 are usable on machines as slow as a 400 MHz G3, so a tuned 10.5 should run acceptably on MID.
Yes, you are right that they are separate but related. (Your post is not rude and I hope my response is not rude, either). The lead example concerned pricing of drills. The three models (perhaps from three different makers) define different market segments as far as the retailer is concerned. Optimizing the price on the three models gives the retailer a chance to maximize both revenues and profits even if the retailer doesn't do anything to market the products differently. Similarly, markdown optimization is both a price optimization and a segmentation issue -- segmenting the "I'll pay anything to be the first person to own this" customers from the "I'll buy it later when its discounted" customers.
The classic example, that I was thinking of, is the revenue management strategies of airlines that attempt to optimize prices across presumed underlying segments of price-sensitive leisure travelers versus price-insensitive business travelers. Technically, it's the identical seat that's being sold for radically different prices (up to 10X different) depending on issues such as how the customer buys the ticket, when they buy the ticket, whether they book a saturday-night stay, etc. The result is that the business customers pay for the plane and the vacation travelers only pay for the fuel and variable costs. The ability to differentiate does benefit the business customers because the added volume of travelers means a more frequent schedule of flights, larger planes, and more destinations.
You are right, though, that true market segmentation involves much more than just price optimization.
Customers vary in their willingness and ability to pay. If a company charges one simple price for each item, it creates a situation in which some people get a great deal (they pay less than they might be willing to) and some people don't buy the product at all (because the price is more than they want to pay). But if a company can find a way to separate the customers that really value the product from the customers that value it less, then more people will be able to buy the product and the company will earn more profit. (You mathematically prove that this increases what is called consumer surplus which is the equivalent to the consumers "profit" on the purchase and the seller's profit). Both side benefit, as does society.
The amusing fact is that this is nothing more than a capitalist version of taking from the rich (those are willing and able to pay more) and giving to the poor (those aren't willing or able to pay more).
Slashdot Mirror
This summary would leave any biologist apoplectic for its flaws:
1. Sex IS sharing. It's the ultimate cross-fertilization (literally). Almost all organisms, including humans, openly and enthusiastically share DNA via this mechanism.
2. Horizontal gene flow is terribly terribly limiting in its utility. Once organism becomes more complex, you can't plug-and-play like you can with a bacteria.
3. Horizontal gene flow does not foster rapid evolution in the same way that sex does. Picking up snippets and fragments from another organism is not as powerful as cross-over in sex (which does a far far better job of doing a controlled recombination of complete plans)
3. No organism in the world can resist "sharing its genome." If pirating the DNA of others was really that great an idea, then the human digestive track would contain tools for pulling DNA out of hamburger. It really would not take much cellular machinery to engulf a target cell, deconstruct it, and co-opt its DNA. The fact that horizontal gene transfer doesn't occur outside of simple organism should be an strong evidence of its limitations.
As much as I enjoy Freeman Dyson, he really lost me on this one.
One key problem with identity is non-uniqueness of names -- there can be several people named Darren Waters. Disambiguating these is nontrivial because it requires other identifiers (e.g., age, hometown, address). Often times the searcher doesn't know anything else about the target. Add spelling errors and things become even more confusing.
A second "which 'Darren Waters'" problem is role-segmentation. I, and am sure many, have multiple online persona reflect different interests, roles, and communities. Depending on the context, I'm a geek, engineer, photographer, stock trader, businessman, etc. Any meaningful search for me would need to specify which version of me they were looking for.
Searching for a person implies both uniquely defining the person and defining which aspect of the person one is looking for because "which 'Darren Waters'" is a problem with two dimensions of ambiguity.
Although price floors seem like a way to preserve the profits of inefficient retailers, I'm sure that the better retailers will figure out a way around any sort of binding MSRP. These might include:
1. Good service: extended hours, trained employees, better inventories, free shipping, free installation, etc.
2. Bundled goodies: accessories, logoed T-shirts, media, etc.
3. Extended 3rd party warrantees
And if the manufacturer says "thou shalt not bundle free stuff," then the retailer only needs to charge a nominal charge for the "separate" item -- say $0.50 for free delivery, installation, and 5 years of 24 hr in-home repair service.
Price is not the only dimension of competition and some would argue that the internet's focus on price competition is one reason retail service has come to suck so bad. The same transparency that lets current web users find the lowest price will let them find the best retailer in a fixed-price environment.
Thin flexible wings date back to the Boeing B-47. Up until this plane appeared in 1947, planes tended to have thick rigid wing structures. Advances in aeronautics, fluid dynamics, and structure design enabled engineers to create thin flexible swept wings that offered lower drag at high speed without flutter or breakage. The wings of B-47 (and B-52) were so floppy, they needed outrigger wheels to keep the wings from dragging on the ground during landings and take-offs.
Kodak has rediscovered what evolution found millions of years ago -- design a dual system such as the rods and cones of the biological eye. The average human eye has about 120 million sensitive, panchromatic rods and only 6 or 7 million color-sensitive cones (many in the central fovea). The brain merges the limited amounts of color information with the larger volume of B/W image data to paint color into the image that we think we see.
Between ever-better computer image recognition algorithms and cheap offshore labor, captchas are doomed. Morevoer, captcha's don't even solve the actual problem because the goal isn't to distinguish human from nonhuman, but to distinguish spammer from nonspammer. This means we need some mechanism to identify a registrant and be aware of their behavior.
Why don't sites band together, share data on abusive registrants, and require each new registrant to provide "references" in the form of their logins to 3-5 other sites. A person with a normal online life could easily demonstrate a pattern of nonspammy behavior. People with no prior history might be placed on probation (their posts are reviewed and may not contain any link-like data). If a registrant posts spam they temporarily (or permanently) lose their accounts on that site and all connected sites.
At some point in time, the only thing that will work is a system that tracks the identity behind the account, assigns a reputation and ostracizes miscreants.
V2.0 will need to handle 2X or 3X the current device's 100kg payload if it is to sell in the U.S. The growing numbers of "enlarged" Americans that I see using those scooters is horrifying.
One could route the output of the tuner into a DVR, record it, and then use the DVR to buffer and skip the VOD signal.
Given the amazing plasticity of the young brain, the time to do this is when the kid is really really young. Ideally, a child might most effectively learn to mentally control a cursor/computer interface about the same time they learn to control their fingers and toes. At that age it really will make controlling a computer as effortless as walking or talking.
The time will come when children that didn't get "Baby's First Brain Mouse" in their first few months of life will be at a scholastic disadvantage to those that did.
Not to detract from the joke, but this isn't true on two levels. First, shivering is the response to a low core body temperature, not the "sensing" of it. Something else in the body is senses the drop in body temperature and triggers the shivering. It may be the way that the conscious brain "senses cold" but its not the way that the body does it. Second, this protein is not for detecting low body temperatures, it is for detecting "cold" surfaces and substances. TFA says this protein triggers at 27 C which is far too cold for use in the shivering mechanism (which triggers at about 35 C).
I'd imagine that YouTube will take an increasing role in self-censoring to adhere to local government regulations. Already, YouTube complies with U.S. DMCA take-down requests, so why wouldn't they comply with a Venezuelan government requests to filter the content according to Venezuelan laws?
YouTube's ownership by Google makes it more vulnerable to coercion by governments that can threaten to turn-off Google unless YouTube complies with local content regulations.
I can't help but wonder how they will compensate for parallax between users' eyes, fingertips, and the screen objects. I'd imagine that the display panel will be relatively thick for structural reasons and that most users will be sitting far off the central axis of the table (i.e., the coffee-table scenario). This means that the image on the screen will be displaced with respect to the line sight from eyeball to fingertip. Moreover different people sitting around the table will see different parallax displacements between fingertips and the screen. Without some compensation for this, the device will be frustrating to use because it won't be intuitive as to where to touch the table top to "touch" an underlying display object. People will often touch the screen where they think they should only to find they've grabbed the object next to the one they really wanted.
P.S. Acrylic is a horrible choice of material for the top as it scratches too easily.
I hope this device actually gets a decent battery life. I've been frustrated that all the new PDAs have worse battery life than the early generation machines. The venerable (and discontinued) Psion 5-series got 25-35 hours of use on a pair for AAs and the original Palm Pilot series got maybe 15-20 hours on a AAAs. Every device since those machines has been distinctly inferior (usually getting only a few hours of real use per charge). If the new machine can't last a full day at a conference or a trans-oceanic flight, then I'll just keep using those older (and better) machines.
Codes to access an MP3 is so lame. I thought they were resurrecting the audio data file format used in early home computers that read data to and from a normal cassette tape recorder.
If they're going to go retro with vinyl, they might as well go retro on the computer formats too.
I agree with you that badly designed traffic flow patterns are a problem. Mistimed lights can encourage speeding, create unneeded backups. (Although I don't envy anyone the problem of making all the lights in all four directions pass traffic at the speed limit). What bothers me are the people that drive 10 mph over the speed limit on a stretch of road with the lights are synced perfectly to the speed limit. These folks have to pile up at the red lights where they become obstacles to people who are driving at a light-sync speed that would have let them sail through every light on green. My point is that if you look up ahead and see the light turn red, you might as well coast to get to the intersection the instant it goes green. And if you commute on the same road day after day, you can learn the speed profile that would get your through light-after-light without ever using the brakes (if it weren't for badly timed lights and the nimrods who can't see red lights beyond the hood of their car and have to hurry to stop at every light).
I wonder how many people will give up their secret security freeze number to phishers?
I'm sure that the country would save quite a bit of gas if more drivers did what their drivers ed teacher said and "got the big picture". I'm amused, and saddened, by the drivers that shoot from red-light to red-light. So often, its plainly obvious from the color of the light and the queue of halted traffic, that there's no way that the light will turn green and traffic will move before these speed demon get to the intersection.
Their average velocity is no higher than any other driver, but they sure do burn a lot of fuel doing it.
By law, faxes in the U.S. must have a "call to remove" number. But I discovered that the number does not work via a little experiment. I called the removal line, entered a different number (a voice line that had never received faxes), and then (within a few days) started receiving fax calls on the voice line. It's just like the email spammers that use victim's unsubscribe notices to signal that they have a live recipient. I'm sure a legal-minded soul could use this behavior to honeypot the faxers, but IANAL.
I've also thought about creating an autodialer script to call the fax removal line and submit every number in the phonebook to it. A simple script could send Hayes commands to a modem to dial the removal line, wait X seconds (or punch "1" to remove or whatever), and then send another dial command to submit bogus removal numbers. Poisoning their DB of faxable numbers would make the return per dialed number much much lower.
He probably muttered "kill -9" and the cops thought it was a confession.
I agree with you 100% and did not mean to imply that the goal is not worthy. Being able to search semantically or to pull out just the relevant information would be hugely valuable.
And I'm sure that next generation search engines will create clever ways of detecting and punishing ontology spam (e.g., noting the dissonance between the text and the tags)
Yes, creating a consistent ontology is challenge. But the bigger challenge is the lack of incentive for ontology truthfulness. If this type of search becomes popular, ontology spam and OSEO (Ontology Search Engine Optimization) will become a booming industry.
Given that MID will be based on the McCaslin platform which will include a dual-core Stealey processor running about 600-800MHz, it should run OS X nicely. Both 10.3 and 10.4 are usable on machines as slow as a 400 MHz G3, so a tuned 10.5 should run acceptably on MID.
Imagine losing your car keys and having someone else find them with Google Earth. Imagine someone without a warrant keeping track of your car keys.
I don't usually wear a tin-foil hat, but this idea has exploit written all over it.
Yes, you are right that they are separate but related. (Your post is not rude and I hope my response is not rude, either). The lead example concerned pricing of drills. The three models (perhaps from three different makers) define different market segments as far as the retailer is concerned. Optimizing the price on the three models gives the retailer a chance to maximize both revenues and profits even if the retailer doesn't do anything to market the products differently. Similarly, markdown optimization is both a price optimization and a segmentation issue -- segmenting the "I'll pay anything to be the first person to own this" customers from the "I'll buy it later when its discounted" customers.
The classic example, that I was thinking of, is the revenue management strategies of airlines that attempt to optimize prices across presumed underlying segments of price-sensitive leisure travelers versus price-insensitive business travelers. Technically, it's the identical seat that's being sold for radically different prices (up to 10X different) depending on issues such as how the customer buys the ticket, when they buy the ticket, whether they book a saturday-night stay, etc. The result is that the business customers pay for the plane and the vacation travelers only pay for the fuel and variable costs. The ability to differentiate does benefit the business customers because the added volume of travelers means a more frequent schedule of flights, larger planes, and more destinations.
You are right, though, that true market segmentation involves much more than just price optimization.
Customers vary in their willingness and ability to pay. If a company charges one simple price for each item, it creates a situation in which some people get a great deal (they pay less than they might be willing to) and some people don't buy the product at all (because the price is more than they want to pay). But if a company can find a way to separate the customers that really value the product from the customers that value it less, then more people will be able to buy the product and the company will earn more profit. (You mathematically prove that this increases what is called consumer surplus which is the equivalent to the consumers "profit" on the purchase and the seller's profit). Both side benefit, as does society.
The amusing fact is that this is nothing more than a capitalist version of taking from the rich (those are willing and able to pay more) and giving to the poor (those aren't willing or able to pay more).