I wish that ISPs would cleanup the bots, but they won't because it would be too expensive. How would an ISP neuter a zombie without disrupting the idiot customer's PC? And if they sever the net connection of bot-infested machines, then who pays for the customer service costs of telling customers that it's the customer's fault that their PC was knocked offline. I'm thinking that each cranky bot-infested customer will cost the ISP $10-$30 in customer service costs (= long calls to explain why they were knocked off the net, what they need to do about it, etc.) for each and every time the customer gets pwned (some customers will probably get infected several times a year or more. Not to mention, the first ISP that does this will get hammered by crime syndicates that don't want there botnets disrupted. Something tells me that the cost to an ISP for letting bots survive is much much lower than the cost of trying to kill them.
It may feel good to blame the idiot that let their machines get infested, but that doesn't pay the salaries of the customer service munchkins (even at low India call center prices).
1. Post "obscene" material to a person's blog, forum, etc. 2. Grab snapshot of said post as proof 3. Wait until past the filing deadline 4. Extort the site owner with proof they had obscene material 5. Profit (no "...." needed)
Because most site owners won't have reported, they face the choice of paying the blackmailer's "reasonable" fee or "up to $300k" in Federal fines.
As much as I like the idea, they've tacked the wrong problem. It's not the car that needs designing, its the manufacturing systems that need designing. Until they can manufacture 1,000,000 of their cars for under $20,000 ea (if they want middle-class buyers in developed nations), or 10,000,000 for under $10,000 ea (if they want worldwide volume), or 100,000,000 for under $5,000 ea (if they want to pre-empt the environmental nightmare of 1 billion new cars in China & India), they've done nothing to address the problem of transportation's contribution to global environmental problems. Form may follow function, but manufacturing defines what form you can make and sell.
As cool as their renderings and open-source specs are, they do nothing to address the real problem. And before someone claims that this is only a concept and that manufacturing can come later, they need to know that 80%-90% of the cost of something is baked in during the design phase (the figure comes from companies such as Volkswagen and Lucent). If manufacturing is an afterthought, there's no hope of getting the costs down because it's too late. Maybe a few stock-option millionaire geeks will be able to spring for the vehicle, but it will never hit a price point that sells the volume that makes a difference.
I hope they switch the focus of the effort to make a breakthrough in manufacturing systems. That would be really cool!
I'm not going to argue over whether the total money spent on science is going up or down or whether it should go up or down.
Instead, I'd like to focus on productivity. If there's any field of endeavor that should have benefited from the last 20 years of information technology, it's science. All the new better tools for simulations, data analysis, experimental equipment, and knowledge sharing mean scientists of today are more productive. Just look at the strides made in genetic sequencing. What used to be a 10-year multibillion dollar "Manhattan Project" venture to sequence one animal's DNA is fast becoming something that an ordinary consumer (or cash-starved bio lab) could afford. I look at the research people can do these days with genetic knockouts, gene splicing, and magical laser-MRI-spectrometric instruments and realize how a scientist of today can do so much more science with a given unit of resource.
I wonder if the "OMG science budgets are dropping" is like saying "OMG the budget to buy desktop PCs is dropping." People don't need $7,000 to buy a PC ($10,000 if adjusted for inflation) -- we can now get a desktop machine that's state-of-art for 1/10 the price. I'm not saying that science follows Moore's Law (twice the bang from half the buck every couple of years), but I'd wager that we can now do a lot more science for the same dollars or the same amount of science for less dollars than in the past).
The point is that the inputs to research ($ in the fiscal plan) don't matter, it's the outputs that matter. Increases in productivity mean that we can do more science for less. Furthermore the same tools that let use do more science for less money also mean we can convert scientific knowledge into valuable engineered products.
Re:No way! (OK, Setup several IDs)
on
The Case for OpenID
·
· Score: 4, Informative
Any website switching to openID exclusively will lose my business
There's no need to abandon a place just because they use openID. Why not setup multiple IDs with different user names, passwords, and email addresses? (I assume that's possible under OpenID?).
I agree that a single collection of IDs (all-eggs-one-basket) represents a dangerous single point of failure. But just because someone implements a new potentially better basket doesn't mean you have to put all your eggs in that basket or avoid using sites that use that type of basket.
HDHP + HSA is the way to go if you are at all heathy and fiscally prudent. Low deductible insurance is a money loser. With a low deductible, you are all but guaranteed to pay more in premiums each year than you would by saving the money and paying from savings. The tax-deferred/tax-free nature of the HSA makes this even more true.
Also, the HSA regs give you tax advantaged savings based on the money you put into the HSA (not the money you take out of it). Check with your accountant, but I believe that nothing in the IRS regs says you must pay for all healthcare expenses with HSA money. Yes, you can't use HSA money for anything but healthcare (unless you are over 65 or disabled), but that doesn't imply that you can't use non-HSA money for healthcare costs. An HSA is a great way to build more tax-deferred savings if you've hit the limits on other tax-deferred savings programs.
Software such as MS Word exemplifies the one-choice-for-all model of software. The result is bloatware when a single piece of software must support a diversity of users.
We all agree that Word is 90% bloatware, but we can't agree on which 10% of functionality to keep.
Although this research is nice, it does not address the worst vulnerabilities of the current internet. Botnets, ARP poisoning, DNS poisoning, pwned routers seem to be a more dangerous risk than mere unreliable components. Cyberterrorism and criminal exploitation of the internet means subverting the system rather than just breaking pieces of it.
The original internet design carried the naive assumption that all the devices on the net could be trusted -- all the devices assumed the validity of all control data, responses to protocols, etc. In the original model, devices had two primary states -- "unavailable" and "available" where "unavailable" might cover both damaged or overloaded components (a slightly more sophisticated version assesses capacity or latency as gradations between the binary unavailable/available dichotomy). In this one dimensional two-state model, disruption tolerance means routing around "Unavailable" or overloaded components.
Yet the rising threat is from malicious entities that want to subvert the network's functioning, not just disable it. Spam, phishing, click fraud, and extortion depend on twisting a functioning network, not just poking holes in the network -- all the parts remain "available" but their data and responses become deceptive. Thus future fault-tolerant networks will need to distinguish between trustworthy and untrustworthy components. This suggests employing techniques such as cryptographic signatures, polling systems, blacklisting, FOAF, firmware integrity checks, and device-to-device secret questions.
Designing a more robust internet is a laudable task but we need to spend more effort on securing against the true threat of untrustworthy components rather than unavailable components.
A friend gave me an old iMac G4 because the HD (Quantum Fireball 13 GB) was fried. The HD's motor driver chip had a nice burn mark where the chip had spewed it's magic smoke. I yanked the circuit board of a similar HD (Quantum Fireball 10 GB) -- the circuit boards "look" identical -- and the Frankenstein HD worked. My friend got her data back and I got to keep the iMac.
The point is that electronics problems with HDs (but not mechanical problems) can be fixed by swapping circuit boards.
You might also consider specific subcultures, including corporate cultures, in your study. Look at the book "the HP Way" for an example of a once strong corporate culture than affected computer products. I've talked to HP employees from different countries and they all felt more affinity to HP's culture than to their own geographic culture.
The larger point: don't forget self-selection biases. The people that make computing products are not random samples drawn from their respective cultures. They are people that gravitate to computing because computing resonates with them in some way. As such, the people that make computers may be more like each other than they are like their fellow citizens or birth-culture.
I always bring an extension cord (a 12' cord with 2-prong plug and 3 outlets) to 3rd-space workplaces. Not only does it let me work from a table that's a little further from the outlet, but I can often convince outlet-hogs to let me plug in my cord and let them plug their stuff into my extension cord.
GeoTrust TrustWatch caught 99%, but had a 32% false positive rate.
I'd be interested to know about these false positives. I'd bet that some legitimate sites use designs that are are hard to distinguish from phishing sites. I would argue this is bad.
Perhaps GeoTrust is right and the false positive sites are wrong.
Is it possible you're the victim of a dictionary attack? These days spammers are sending junk to $RANDOMNAME@knowndomainname.com. I've seen this on both big national ISP domain names and dinky domain names that I own. If your user name a common name and letter, it might be getting hit at random without any need for compromising your account.
Given all the issues with "liquids" and flaming laptop batteries, I doubt this will be allowed on aircraft. (We'll see if methanol fuel cells pass TSA muster). I guess that's just another example of the terrorists winning their goal of keeping people out of the 21st century.
You don't really think the bank is going to create money to pay for the losses, do you? Make no mistake about it--banks, like every other convenient, abstract legal fiction--don't pay for anything. Individuals pay for things.
Exactly true in the short-term, but not true in the long-term because customers can chose which bank to do business with. Banks still compete and the ones that can levy the lowest fees because they have the lowest phishing related losses will get the most business. The interesting issue is that banks have three strategies for lowering phishing losses:
1) Deny claims for losses
2) Implement security
3) Screen-out phish-prone customers (e.g., preferentially market to young, tech-savy consumers with high credit scores)
Of these three, the first is a PR nightmare and may become illegal. The second is expensive and may inconvenience customers. The third is interesting but gets into nasty ethical issues if the bank tries too hard to avoid people it thinks are bad phishing risks.
Do you mean an HP 9825, one of the one-line-dot-matrix-display jobbies with the rom drawers under the keyboard? I ask because I think there was also an HP-25 but it was a handheld calculator, and if I'm wrong I'm curious to know more.
No, it was an HP-25 handheld calculator. I learned the basics of programming on it, then switched to BASIC on a CDC 6400/6600 a few months later, then went to APL on a DEC-10 a few months after that. I would have killed for my own HP-9800 series machine back in the day.
That's another interesting difference between then and now. Back then a very wide range of companies made a very wide range of processors. People really argued about the merits of Intel 8080s, Zilog Z80s, MOS 6502, Motorola 6800. HP was renowned for making its own processors for its own workstations. The CPUs, like the software of that day, were so simple (a few thousand transistors total) that you really could understand the entire chip and logic board. Today, x86 dominates and a few embedded CPU/SOC families handle the device world.
Those were some good times (not that I'd ever turn the clock back because I love GHz and GB too much).
I too have lamented the changes in IT. When I first learned to program (1977 on an HP-25), the technical environment was very different. Back then, everyone who wanted to use a computer HAD to know how to program. The scarcity of software meant that everyone wrote their own code or, at least, typed in code from a magazine. Very limited software sharing schemes and the inability to quickly find software meant it was easier to write your own than to find someone else's software. Simple languages, simple hardware, simple interfaces, and simple APIs ruled. When the entire OS plus application suite resided in a few k of RAM, it was easy to both work with the system or create your own. It took very little effort for a novice programmer to produce world-class code because the bar was so low and the functionality so primitive that anyone could make something interesting. In the old days, everyone grew their own code.
Today it's all different. The OS has become a beast that not even a team of programmers can fully comprehend. IDEs, OOP, and layered architectures try to hide the complexity, but its still there. Moreover, almost any bit of code or application that one might want has a multiple incarnations ready for buying/downloading from commercial/shareware/OSS sources. It's now very easy to find the application you want and much harder to write something that is better than anythign else. In the new days, few grow their own code.
Perhaps its like the change from a subsistence-agrarian world to a world of craftsmen (or industry) where programming is like farming. In the past, everyone grew their own code. Today, no one grows their own food and farming is a very minor part of the global economy. Farmers may lament that most children in the city don't know how to milk a cow or thresh wheat, but perhaps those skills aren't needed in most people. Just as one farmer can now feed some 40 people, one programmer servers the programming needs of a growing number of users. Consider that Microsoft as 60,000 employees whose code runs on at least 600 million operating PCs -- more than 10,000 non-programming users per programmer.
As with farming, we now live in a world where few need to grow their own code. As far as schools are concerned we may be entering a world in which fewer than 1 child per class will ever need to know how to program. That makes me sad at some level, I truly enjoyed learning to program, but it may be an inevitable part of the maturation process for IT and the internet.
I agree with you on this - it's too easy to get fixated on rare cases and miss the bigger picture. The challenge is that there are two types of "important" use cases. First, are the ones which most people take -- the 99.9% use cases. Second are the ones that lead to serious consequences -- e.g., falling off the Segway or losing a customer.
The core challenge is to create robust software that performs well most of the time (the 99% common case) and does not perform badly in the rare (1%) cases. Thinking about consequence-weighted use cases (or misuse cases) helps create robust software. I'd never insist that all software provide the perfect response to all cases because that forces the programmer to spend too much time on the myriad of obscure cases. I only recommend that programs detect deviations from "normal" cases and avoid failures. The first step, probably not done in enough situations, is to realize that users can and do violate the assumptions of the normal use cases.
This doesn't surprise me. It's just another example of how easily programmers fall into the trap of common "use cases" when writing software. Too often, programs are written on the assumption on a simple linear chain of events driven by a use case. In the Segway example, it would appear that the people who wrote the control logic for the scooter assumed that people would get on the machine from a full stop and get off the machine at a full stop. Remounting the machine during the stopping process violated this assumption and exposed a fault in the control logic. I see this type of problem all the time on e-commerce sites (even Amazon.com has the problem) when the buyer attempts to unroll part of a transaction to change something or check an alternative path in the buy/ship/bill/confirm cycle.
The point: always assume the user might do anything at any point in time.
I would hope that this wiki would have tools to let a teacher "lock-in" a version of the text for use during the term. It would be frustrating to discover that midway through the term that the wiki changed material in some chapter -- adding material the teacher had not plan to cover or removing material that they had. I could even see some teachers preferring to retain a particular version of the text for a couple of years if they had invested heavily in teaching plans that were specific to that version.
The ultimate tool would let teachers mix and match chapters -- picking different versions from different years to suit their tastes.
The point is that once a wiki transitions from casual/random access (e.g. wikipedia) to one of methodical use, then the user needs more say in versions or some way to retain their favored version.
IANAL, but it would seem that the defrauded parties could always try to take the person to civil court. For better or worse, you can sue over anything.
Of course, if you can argue that losses in the game have a real-world monetary counterpart, then any gains in that world would seem to have a real world monetary counterpart. That gets the IRS involved with issues of income tax on gains in the game.
If game gains are taxable, then perhaps they can get the Ponzi scheme operator on tax evasion -- it worked with Al Capone.
Trap jaw ants do live in the wild in the southern U.S -- I've studied them in Austin Texas. They're not easy to find as the colonies are very small and the individuals tend to be quite reclusive. They are largish ants (about about 1 cm in length), dark in color, and tend to be fairly slow moving when foraging in leaf litter and under rocks. They walk around with their jaws cocked open and one or two pairs of trigger hairs in the mouth fire the jaw. As the article states a snap of the jaw impales the prey and then the ant stings it. If they fire the jaw on a solid object, the ant goes flying. Either way the jaw emits a loud 'snap' when triggered. Despite the sharp hair-trigger jaw and sting, these ants tend to fall into the "fierce in their nest, but timid in the wild" range of ant behavior.
As amazing as the trap jaw design is, these ants are not unique. The trap jaw concept evolved at least twice in ants. Two collections of ant species on widely separated arms of the ant family tree use a trap jaw mechanism for capturing prey. They share the same jaw design, but have very different head shapes. Ants of genus Odontomachus (the ones in the video) have an odd-shaped lumpy cylindrical head. Those on the other side of the ant family tree (genus Daceton and Strumigenys) have a distinct heart-shaped head. Species of both types occur in the U.S. The Strumigenys that I've seen in the U.S. are very small (about 2 mm) and thrive on similarly tiny creatures found in rotting logs, leaf litter, etc.
First, there are patterns. You are right that those patterns have a limited capacity to absorb trading and that anyone who finds a pattern would do well to keep it a secret. The EMH (Efficient Markets Hypothesis) is best lampooned by the following old joke. Two economists are walking down the sidewalk when one of them spots a $100 bill in the grass. The first economist starts to pick up the $100 when the second economist tells him, "Don't bother, if it were a real $100 bill, someone would have already taken it." Moreover, EMH makes predictions about the statistical distribution of price movements and the volume of trading that are empirically false.
Second, the price dynamics are not entirely caused by exogenous factors. Investors, speculators, the media, and government officials do watch the prices. People make buy and sell judgments without any fundamental basis such a stock being "expensive" just because the stock is $300/share (never mind understanding the relationships between price per share and capitalization). Humans also have instinctual beliefs in patterns such as trends or momentum that are self-fulfilling. If enough traders believe in trends or momentum, they will trade in a way that creates trends.
The profitable patterns do exist (and so do a large number of profitless pseudopatterns). People with a very sound understanding of both market psychology and statistics can and do succeed.
I can only hope that this scheme includes distributed storage of the pointers to all the fragments, too. Distributed data is only as reliable as the metadata that record where the data fragments are located. If the user of the system loses their only copy of the map to their fragments, the data is lost. If, on the other hand, each fragment also includes encrypted pointers to a few other fragments, then decrypting any fragment lets one bootstrap recovery of the entire network of fragments (a good thing if you want reliability, maybe less desirable for those seeking security).
I would submit that the long tail, in any form, creates false hopes for content creators. Consider the economics of being in a band with 4 members plus a couple of multi-talented support crew (e.g. manager, equipment engineer, lyricist, sound engineer, etc.). Such a group needs to clear $120,000/year after expenses (equipment, vehicles, gas, marketing, etc.) just to stay above the poverty line (20k/person before taxes). That's suggest a gross of $150,000 to $200,000.
Getting that from iTunes means getting 150,000 to 200,000 downloads per year. If the group creates one album of new songs each year, and if the band's album is like most, then maybe 3 "good" songs shoulder the burden of feeding the band. The band would need their three good songs to average more than 50,000 downloads/year (that's 1,000 per week). That's a 1,000 downloads per week just to stay afloat. If, by some miracle, the band creates 12 good songs every year and has a deep backlist of 24 more good songs with steady sales, then they still need to average 80 downloads per each and every song.
iTunes has about 3 million songs and perhaps 25 million downloads per week. Thus the average song only gets 8 downloads per week. Under the long tail model, the vast majority of songs have average performance and relatively few have above average performance. That means that the vast majority of songs don't pay enough to keep a band above the poverty line. In fact, under this model, iTunes probably represents only 1% to 10% of the money needed to stay afloat.
My point is that the long tail is great for consumers because it gives them more choice. But the model consigns the majority of the content creators to a below subsistence wages existence as they hope that they can climb out of the deep long tail.
Slashdot Mirror
I wish that ISPs would cleanup the bots, but they won't because it would be too expensive. How would an ISP neuter a zombie without disrupting the idiot customer's PC? And if they sever the net connection of bot-infested machines, then who pays for the customer service costs of telling customers that it's the customer's fault that their PC was knocked offline. I'm thinking that each cranky bot-infested customer will cost the ISP $10-$30 in customer service costs (= long calls to explain why they were knocked off the net, what they need to do about it, etc.) for each and every time the customer gets pwned (some customers will probably get infected several times a year or more. Not to mention, the first ISP that does this will get hammered by crime syndicates that don't want there botnets disrupted. Something tells me that the cost to an ISP for letting bots survive is much much lower than the cost of trying to kill them.
It may feel good to blame the idiot that let their machines get infested, but that doesn't pay the salaries of the customer service munchkins (even at low India call center prices).
1. Post "obscene" material to a person's blog, forum, etc.
2. Grab snapshot of said post as proof
3. Wait until past the filing deadline
4. Extort the site owner with proof they had obscene material
5. Profit (no "...." needed)
Because most site owners won't have reported, they face the choice of paying the blackmailer's "reasonable" fee or "up to $300k" in Federal fines.
As much as I like the idea, they've tacked the wrong problem. It's not the car that needs designing, its the manufacturing systems that need designing. Until they can manufacture 1,000,000 of their cars for under $20,000 ea (if they want middle-class buyers in developed nations), or 10,000,000 for under $10,000 ea (if they want worldwide volume), or 100,000,000 for under $5,000 ea (if they want to pre-empt the environmental nightmare of 1 billion new cars in China & India), they've done nothing to address the problem of transportation's contribution to global environmental problems. Form may follow function, but manufacturing defines what form you can make and sell.
As cool as their renderings and open-source specs are, they do nothing to address the real problem. And before someone claims that this is only a concept and that manufacturing can come later, they need to know that 80%-90% of the cost of something is baked in during the design phase (the figure comes from companies such as Volkswagen and Lucent). If manufacturing is an afterthought, there's no hope of getting the costs down because it's too late. Maybe a few stock-option millionaire geeks will be able to spring for the vehicle, but it will never hit a price point that sells the volume that makes a difference.
I hope they switch the focus of the effort to make a breakthrough in manufacturing systems. That would be really cool!
I'm not going to argue over whether the total money spent on science is going up or down or whether it should go up or down.
Instead, I'd like to focus on productivity. If there's any field of endeavor that should have benefited from the last 20 years of information technology, it's science. All the new better tools for simulations, data analysis, experimental equipment, and knowledge sharing mean scientists of today are more productive. Just look at the strides made in genetic sequencing. What used to be a 10-year multibillion dollar "Manhattan Project" venture to sequence one animal's DNA is fast becoming something that an ordinary consumer (or cash-starved bio lab) could afford. I look at the research people can do these days with genetic knockouts, gene splicing, and magical laser-MRI-spectrometric instruments and realize how a scientist of today can do so much more science with a given unit of resource.
I wonder if the "OMG science budgets are dropping" is like saying "OMG the budget to buy desktop PCs is dropping." People don't need $7,000 to buy a PC ($10,000 if adjusted for inflation) -- we can now get a desktop machine that's state-of-art for 1/10 the price. I'm not saying that science follows Moore's Law (twice the bang from half the buck every couple of years), but I'd wager that we can now do a lot more science for the same dollars or the same amount of science for less dollars than in the past).
The point is that the inputs to research ($ in the fiscal plan) don't matter, it's the outputs that matter. Increases in productivity mean that we can do more science for less. Furthermore the same tools that let use do more science for less money also mean we can convert scientific knowledge into valuable engineered products.
Any website switching to openID exclusively will lose my business
There's no need to abandon a place just because they use openID. Why not setup multiple IDs with different user names, passwords, and email addresses? (I assume that's possible under OpenID?).
I agree that a single collection of IDs (all-eggs-one-basket) represents a dangerous single point of failure. But just because someone implements a new potentially better basket doesn't mean you have to put all your eggs in that basket or avoid using sites that use that type of basket.
HDHP + HSA is the way to go if you are at all heathy and fiscally prudent. Low deductible insurance is a money loser. With a low deductible, you are all but guaranteed to pay more in premiums each year than you would by saving the money and paying from savings. The tax-deferred/tax-free nature of the HSA makes this even more true.
Also, the HSA regs give you tax advantaged savings based on the money you put into the HSA (not the money you take out of it). Check with your accountant, but I believe that nothing in the IRS regs says you must pay for all healthcare expenses with HSA money. Yes, you can't use HSA money for anything but healthcare (unless you are over 65 or disabled), but that doesn't imply that you can't use non-HSA money for healthcare costs. An HSA is a great way to build more tax-deferred savings if you've hit the limits on other tax-deferred savings programs.
Software such as MS Word exemplifies the one-choice-for-all model of software. The result is bloatware when a single piece of software must support a diversity of users.
We all agree that Word is 90% bloatware, but we can't agree on which 10% of functionality to keep.
Although this research is nice, it does not address the worst vulnerabilities of the current internet. Botnets, ARP poisoning, DNS poisoning, pwned routers seem to be a more dangerous risk than mere unreliable components. Cyberterrorism and criminal exploitation of the internet means subverting the system rather than just breaking pieces of it.
The original internet design carried the naive assumption that all the devices on the net could be trusted -- all the devices assumed the validity of all control data, responses to protocols, etc. In the original model, devices had two primary states -- "unavailable" and "available" where "unavailable" might cover both damaged or overloaded components (a slightly more sophisticated version assesses capacity or latency as gradations between the binary unavailable/available dichotomy). In this one dimensional two-state model, disruption tolerance means routing around "Unavailable" or overloaded components.
Yet the rising threat is from malicious entities that want to subvert the network's functioning, not just disable it. Spam, phishing, click fraud, and extortion depend on twisting a functioning network, not just poking holes in the network -- all the parts remain "available" but their data and responses become deceptive. Thus future fault-tolerant networks will need to distinguish between trustworthy and untrustworthy components. This suggests employing techniques such as cryptographic signatures, polling systems, blacklisting, FOAF, firmware integrity checks, and device-to-device secret questions.
Designing a more robust internet is a laudable task but we need to spend more effort on securing against the true threat of untrustworthy components rather than unavailable components.
A friend gave me an old iMac G4 because the HD (Quantum Fireball 13 GB) was fried. The HD's motor driver chip had a nice burn mark where the chip had spewed it's magic smoke. I yanked the circuit board of a similar HD (Quantum Fireball 10 GB) -- the circuit boards "look" identical -- and the Frankenstein HD worked. My friend got her data back and I got to keep the iMac.
The point is that electronics problems with HDs (but not mechanical problems) can be fixed by swapping circuit boards.
You might also consider specific subcultures, including corporate cultures, in your study. Look at the book "the HP Way" for an example of a once strong corporate culture than affected computer products. I've talked to HP employees from different countries and they all felt more affinity to HP's culture than to their own geographic culture.
The larger point: don't forget self-selection biases. The people that make computing products are not random samples drawn from their respective cultures. They are people that gravitate to computing because computing resonates with them in some way. As such, the people that make computers may be more like each other than they are like their fellow citizens or birth-culture.
I always bring an extension cord (a 12' cord with 2-prong plug and 3 outlets) to 3rd-space workplaces. Not only does it let me work from a table that's a little further from the outlet, but I can often convince outlet-hogs to let me plug in my cord and let them plug their stuff into my extension cord.
Share the juice!
GeoTrust TrustWatch caught 99%, but had a 32% false positive rate.
I'd be interested to know about these false positives. I'd bet that some legitimate sites use designs that are are hard to distinguish from phishing sites. I would argue this is bad.
Perhaps GeoTrust is right and the false positive sites are wrong.
Is it possible you're the victim of a dictionary attack? These days spammers are sending junk to $RANDOMNAME@knowndomainname.com. I've seen this on both big national ISP domain names and dinky domain names that I own. If your user name a common name and letter, it might be getting hit at random without any need for compromising your account.
Given all the issues with "liquids" and flaming laptop batteries, I doubt this will be allowed on aircraft. (We'll see if methanol fuel cells pass TSA muster). I guess that's just another example of the terrorists winning their goal of keeping people out of the 21st century.
You don't really think the bank is going to create money to pay for the losses, do you? Make no mistake about it--banks, like every other convenient, abstract legal fiction--don't pay for anything. Individuals pay for things.
Exactly true in the short-term, but not true in the long-term because customers can chose which bank to do business with. Banks still compete and the ones that can levy the lowest fees because they have the lowest phishing related losses will get the most business. The interesting issue is that banks have three strategies for lowering phishing losses:
1) Deny claims for losses
2) Implement security
3) Screen-out phish-prone customers (e.g., preferentially market to young, tech-savy consumers with high credit scores)
Of these three, the first is a PR nightmare and may become illegal. The second is expensive and may inconvenience customers. The third is interesting but gets into nasty ethical issues if the bank tries too hard to avoid people it thinks are bad phishing risks.
Do you mean an HP 9825, one of the one-line-dot-matrix-display jobbies with the rom drawers under the keyboard? I ask because I think there was also an HP-25 but it was a handheld calculator, and if I'm wrong I'm curious to know more.
No, it was an HP-25 handheld calculator. I learned the basics of programming on it, then switched to BASIC on a CDC 6400/6600 a few months later, then went to APL on a DEC-10 a few months after that. I would have killed for my own HP-9800 series machine back in the day.
That's another interesting difference between then and now. Back then a very wide range of companies made a very wide range of processors. People really argued about the merits of Intel 8080s, Zilog Z80s, MOS 6502, Motorola 6800. HP was renowned for making its own processors for its own workstations. The CPUs, like the software of that day, were so simple (a few thousand transistors total) that you really could understand the entire chip and logic board. Today, x86 dominates and a few embedded CPU/SOC families handle the device world.
Those were some good times (not that I'd ever turn the clock back because I love GHz and GB too much).
I too have lamented the changes in IT. When I first learned to program (1977 on an HP-25), the technical environment was very different. Back then, everyone who wanted to use a computer HAD to know how to program. The scarcity of software meant that everyone wrote their own code or, at least, typed in code from a magazine. Very limited software sharing schemes and the inability to quickly find software meant it was easier to write your own than to find someone else's software. Simple languages, simple hardware, simple interfaces, and simple APIs ruled. When the entire OS plus application suite resided in a few k of RAM, it was easy to both work with the system or create your own. It took very little effort for a novice programmer to produce world-class code because the bar was so low and the functionality so primitive that anyone could make something interesting. In the old days, everyone grew their own code.
Today it's all different. The OS has become a beast that not even a team of programmers can fully comprehend. IDEs, OOP, and layered architectures try to hide the complexity, but its still there. Moreover, almost any bit of code or application that one might want has a multiple incarnations ready for buying/downloading from commercial/shareware/OSS sources. It's now very easy to find the application you want and much harder to write something that is better than anythign else. In the new days, few grow their own code.
Perhaps its like the change from a subsistence-agrarian world to a world of craftsmen (or industry) where programming is like farming. In the past, everyone grew their own code. Today, no one grows their own food and farming is a very minor part of the global economy. Farmers may lament that most children in the city don't know how to milk a cow or thresh wheat, but perhaps those skills aren't needed in most people. Just as one farmer can now feed some 40 people, one programmer servers the programming needs of a growing number of users. Consider that Microsoft as 60,000 employees whose code runs on at least 600 million operating PCs -- more than 10,000 non-programming users per programmer.
As with farming, we now live in a world where few need to grow their own code. As far as schools are concerned we may be entering a world in which fewer than 1 child per class will ever need to know how to program. That makes me sad at some level, I truly enjoyed learning to program, but it may be an inevitable part of the maturation process for IT and the internet.
I agree with you on this - it's too easy to get fixated on rare cases and miss the bigger picture. The challenge is that there are two types of "important" use cases. First, are the ones which most people take -- the 99.9% use cases. Second are the ones that lead to serious consequences -- e.g., falling off the Segway or losing a customer.
The core challenge is to create robust software that performs well most of the time (the 99% common case) and does not perform badly in the rare (1%) cases. Thinking about consequence-weighted use cases (or misuse cases) helps create robust software. I'd never insist that all software provide the perfect response to all cases because that forces the programmer to spend too much time on the myriad of obscure cases. I only recommend that programs detect deviations from "normal" cases and avoid failures. The first step, probably not done in enough situations, is to realize that users can and do violate the assumptions of the normal use cases.
This doesn't surprise me. It's just another example of how easily programmers fall into the trap of common "use cases" when writing software. Too often, programs are written on the assumption on a simple linear chain of events driven by a use case. In the Segway example, it would appear that the people who wrote the control logic for the scooter assumed that people would get on the machine from a full stop and get off the machine at a full stop. Remounting the machine during the stopping process violated this assumption and exposed a fault in the control logic. I see this type of problem all the time on e-commerce sites (even Amazon.com has the problem) when the buyer attempts to unroll part of a transaction to change something or check an alternative path in the buy/ship/bill/confirm cycle.
The point: always assume the user might do anything at any point in time.
I would hope that this wiki would have tools to let a teacher "lock-in" a version of the text for use during the term. It would be frustrating to discover that midway through the term that the wiki changed material in some chapter -- adding material the teacher had not plan to cover or removing material that they had. I could even see some teachers preferring to retain a particular version of the text for a couple of years if they had invested heavily in teaching plans that were specific to that version.
The ultimate tool would let teachers mix and match chapters -- picking different versions from different years to suit their tastes.
The point is that once a wiki transitions from casual/random access (e.g. wikipedia) to one of methodical use, then the user needs more say in versions or some way to retain their favored version.
IANAL, but it would seem that the defrauded parties could always try to take the person to civil court. For better or worse, you can sue over anything.
Of course, if you can argue that losses in the game have a real-world monetary counterpart, then any gains in that world would seem to have a real world monetary counterpart. That gets the IRS involved with issues of income tax on gains in the game.
If game gains are taxable, then perhaps they can get the Ponzi scheme operator on tax evasion -- it worked with Al Capone.
Trap jaw ants do live in the wild in the southern U.S -- I've studied them in Austin Texas. They're not easy to find as the colonies are very small and the individuals tend to be quite reclusive. They are largish ants (about about 1 cm in length), dark in color, and tend to be fairly slow moving when foraging in leaf litter and under rocks. They walk around with their jaws cocked open and one or two pairs of trigger hairs in the mouth fire the jaw. As the article states a snap of the jaw impales the prey and then the ant stings it. If they fire the jaw on a solid object, the ant goes flying. Either way the jaw emits a loud 'snap' when triggered. Despite the sharp hair-trigger jaw and sting, these ants tend to fall into the "fierce in their nest, but timid in the wild" range of ant behavior.
As amazing as the trap jaw design is, these ants are not unique. The trap jaw concept evolved at least twice in ants. Two collections of ant species on widely separated arms of the ant family tree use a trap jaw mechanism for capturing prey. They share the same jaw design, but have very different head shapes. Ants of genus Odontomachus (the ones in the video) have an odd-shaped lumpy cylindrical head. Those on the other side of the ant family tree (genus Daceton and Strumigenys) have a distinct heart-shaped head. Species of both types occur in the U.S. The Strumigenys that I've seen in the U.S. are very small (about 2 mm) and thrive on similarly tiny creatures found in rotting logs, leaf litter, etc.
First, there are patterns. You are right that those patterns have a limited capacity to absorb trading and that anyone who finds a pattern would do well to keep it a secret. The EMH (Efficient Markets Hypothesis) is best lampooned by the following old joke. Two economists are walking down the sidewalk when one of them spots a $100 bill in the grass. The first economist starts to pick up the $100 when the second economist tells him, "Don't bother, if it were a real $100 bill, someone would have already taken it." Moreover, EMH makes predictions about the statistical distribution of price movements and the volume of trading that are empirically false.
Second, the price dynamics are not entirely caused by exogenous factors. Investors, speculators, the media, and government officials do watch the prices. People make buy and sell judgments without any fundamental basis such a stock being "expensive" just because the stock is $300/share (never mind understanding the relationships between price per share and capitalization). Humans also have instinctual beliefs in patterns such as trends or momentum that are self-fulfilling. If enough traders believe in trends or momentum, they will trade in a way that creates trends.
The profitable patterns do exist (and so do a large number of profitless pseudopatterns). People with a very sound understanding of both market psychology and statistics can and do succeed.
I can only hope that this scheme includes distributed storage of the pointers to all the fragments, too. Distributed data is only as reliable as the metadata that record where the data fragments are located. If the user of the system loses their only copy of the map to their fragments, the data is lost. If, on the other hand, each fragment also includes encrypted pointers to a few other fragments, then decrypting any fragment lets one bootstrap recovery of the entire network of fragments (a good thing if you want reliability, maybe less desirable for those seeking security).
I would submit that the long tail, in any form, creates false hopes for content creators. Consider the economics of being in a band with 4 members plus a couple of multi-talented support crew (e.g. manager, equipment engineer, lyricist, sound engineer, etc.). Such a group needs to clear $120,000/year after expenses (equipment, vehicles, gas, marketing, etc.) just to stay above the poverty line (20k/person before taxes). That's suggest a gross of $150,000 to $200,000.
Getting that from iTunes means getting 150,000 to 200,000 downloads per year. If the group creates one album of new songs each year, and if the band's album is like most, then maybe 3 "good" songs shoulder the burden of feeding the band. The band would need their three good songs to average more than 50,000 downloads/year (that's 1,000 per week). That's a 1,000 downloads per week just to stay afloat. If, by some miracle, the band creates 12 good songs every year and has a deep backlist of 24 more good songs with steady sales, then they still need to average 80 downloads per each and every song.
iTunes has about 3 million songs and perhaps 25 million downloads per week. Thus the average song only gets 8 downloads per week. Under the long tail model, the vast majority of songs have average performance and relatively few have above average performance. That means that the vast majority of songs don't pay enough to keep a band above the poverty line. In fact, under this model, iTunes probably represents only 1% to 10% of the money needed to stay afloat.
My point is that the long tail is great for consumers because it gives them more choice. But the model consigns the majority of the content creators to a below subsistence wages existence as they hope that they can climb out of the deep long tail.