Assuming that at least half of the places you visit will be fairly modern cities, you should take advantage of the opportunities for internet access while you are in port. Investigate places that offer free or low-cost WiFi service in the ports you'll be visiting. That will at least mitigate the low access levels you are limited to on board ship.
It's not just the ease of patching, it's also the loss of the philosophy of change management. Apart from the non-digital technology used for the change management, compare the level of detail in the change management in Tracy Kidder's The Soul of a New Machine with change management standards used nowadays.
I blame the fact that instead of using expensive Electrical or Computer Engineers, software companies base their development teams around computer science (strong on programming theory, weak on proceses) or Business IT (good business process knowledge, weak technical) people. Just because a person can code doesn't mean they can design or manage. Or vice versa.
I'm thinking when someone reports "suspicious comments" to airline personnel, they should be required to get off the plane so the FBI can interview them as well. And if the suspicions turn out to be unfounded, they get the same treatment as the victims that they falsely reported: If the airline doesn't let the "suspcious" characters back on, the people who reported their suspicions don't get back on either.
If you're going to report stuff like this, put your time and money where your mouth is.
Advice from another user of generators after hurricane strikes:
1) Safety tip #1: You do not want to run the generator into your breaker box unless you (at a minimum) turn off the main breaker to your house. Ideally you should install one of those grid-tie disconnect devices like they use for solar installations. The breaker can be accidentally switched on, while the grid-tie can't. Without the breaker/cutoff, you are running power out to the power company. This risks three serious problems: a) providing free power to your neighbors that overloads the generator, b) the power kills a line worker after being upped to 2400 volts by a power company transformer working in reverse, and c)(not so big a risk in winter...) A lightning strike on the utility wires causes the gas tank of the generator to ignite, burning down your house. 1a) Safety tip #2 and 3: Don't ignore the ground wire for the generator. Essential to prevent sparks when you are refilling the gas. Also, the generator will require some downtime. Most are typically rated for 8 hours of continuous use.
2) The shed is overkill. Get a wheel kit so you can roll it out of the garage when needed and park it outdoors next to your mud room or something. You may want a chain and padlock too. People were having these generators stolen from their yards/driveways down here after the hurricane.
3) Not sure how much power the furnace blower needs, but a 5500 W generator would supply about 40 Amps worth of current overall, so that is probably at the low end of the required capacity. Make sure you read the rating for "continuous" power and not "peak" power.
4) Of common household appliances, coffee-makers draw surprising amounts of current. More than the microwave.
5) The router might not do much good after a while. Phone company and cable company power their systems from batteries that usually last 24-48 hours. Some sites may have generators or their techs may swap fresh batteries in.
6) You can _probably_ get by with a 5500 W generator for $700-$900, but you may want to look at a 7000 W to 10 kW generator in the $1000-$2000 range, especially if your furnace blower is rated >20 Amps.
7) Try to find one with an electric starter. Doing the old "lawnmower pull" on a frozen engine sounds like an exercise in futility.
Well, not necessarily. Evil fighting evil for worldly power is the normal state of affairs.
One reason for separation of powers in the US constitution and the adversarial construction of our courts is to provide a mechanism for greedy scumbags to rat each other out. The founders recognized that in every corrupt scheme there comes a point where some of the criminals either get scared of getting caught or decide that it's more profitable to turn their ex-partners in.
Under that logic, if your original downloaded copy is to a flash drive (or CD), wouldn't right of first sale apply to the copy on the flash drive? That is just a non-volatile version of the actual bit stream that the seller sent to you. It's not a separate copy, it's the tangible storage of the original sold copy.
I suppose the record companies would like to argue that the original copy is just the original bitstream saved into the computer's RAM or in the swap file, but that logic is implementation dependent. It is physically possible to write a software program that takes the bit-stream from the seller and directly copies it to a flash drive with no more than a small memory buffer that would qualify as a fair use excerpt of the copyrighted work.
Non-technical users should not have to do research to distinguish whether their particular download software is coded to allow them first sale rights or not, when from a practical usability standpoint, the only copy of the work that they received is the copy on their flash drive. Temporary copies in RAM or swap are inaccessible by ordinary means (and use of "extra-ordinary" means would be evidence of willful copyright violation in any case). As long as they don't copy the copy that they purchased legally, the right of first sale should apply.
(IANAL) Don't necessarily have to show a full value of damages for copyright violation. Damages for copyright violation of registered works are calculated by statutory formula rather than actual damages. The $220k verdict that was recently overturned by the judge is an example of a statutory damage award.
The overrule was not because the damages were deemed unfair for a genuine case of copyright violation, but because the judge decided that "making available" was not a valid definition of a copyright violation, making his earlier jury instructions invalid. If an actual violation had occurred, the statutory damages would apply for each violation even if the actual damages caused were less.
OTOH, just linking to other sites that actually host an "unauthorized" MP3 isn't distribution or even "making available" by any definition.
The basic math is that in order to prevent hash collisions you need a one to one mapping from the input domain to the hash range. In other words, a unique hash value for every input value. Technically, it isn't a true one-to-one mapping, because there could be extra unused values in the hash range that aren't mapped from the input values, but unused values in the hash space translate into wasted memory storing unneeded bits. A one-to-one mapping is the most efficient possible mapping.
In the case of an arbitrary/random data sequence of finite length, the number of possible values of the input space is 2^n where n is the number of bits in the input.
In order to support this number of variations of the input, the length of the hash must be the same number of bits (or greater). The hash needs to establish a one-to-one mapping of potential inputs to potential outputs.
As an example, if your input data consists of 8 bits with no restrictions on the value of any given bit, you have 2^8 possible bit combinations (the full ASCII table, for example) that must be mapped into the hash. In order to successfully map that many different permutations of input, you need an 8-bit hash. If you only have a 6 bit hash, you are trying to map 2^8 (256) possible input values values into an output space that contains only 2^6 (64) values, so either some values in the domain are not mapped into the range (invalid hash/hash failure) or else there are values in the output range that are reached from multiple inputs (hash collisions).
Now a purely arbitrary or random input sequence is not the only possible condition for hashed data. If an input data sequence can be compressed losslessly due to correlation between input values or because some sequence combinations are prohibited, the hash can be shorter than the input string (but not shorter than the best possible compressed version of the string). Again, this boils down to the basic math of one-to-one mappings and permutations of n bits of data. If data is losslessly compressible, that means a one to one mapping can be created from the uncompressed data to the compressed version. One to one mappings are transitive, so you can create a perfect hash of the compressed data without data loss (or hash collision).
The real problem with the new formula is that it punishes the people driving small efficient cars vs. the gas-guzzlers. So if I'm driving an SUV that gets 12 MPG, I pay only 16 cents per gallon using the per-mile formula (minus any out of state mileage) vs. 25 cents (or whatever) under the regular per-gallon tax. If I'm driving an efficient car (that incidentally does less damage to the highways being maintained) that gets 30 MPG, I'm paying 40 cent a gallon in taxes under the mileage formula.
Since the actual formula for highway maintenance costs involve the distance travelled per vehicle, the weight of the vehicle, the speed of the vehicle and the designed load capacity of the highway, a per mile/per vehicle ton tax is a fair tax, if it is universal. Allowing a gas-guzzler loophole for vehicles that don't conform to the new system inverts the reward mechanism from the desired goal of cutting fuel consumption.
Or put the tax on the tires: of course 1.3 cents a mile for a set of 30,000 mile tires would add about $400 to the price of a set of tires. A bit hard to swallow that all at once.
GPS is way too intrusive and not necessarily all that accurate. Soon they'd start using it to catch speeders.
Probably the best way to do this is a combo of vehicle registration, gas, tire, engine oil and electric taxes, plus any other alterna-fuels that catch on. They can achieve the stated result of proper funding for Oregon highways without forcing tech on people that would enable so much more surveillance.
Whether or not the abuser is known to the child or is a stranger doesn't change the fact that the child is being abused. The emotional reaction of the free speech libertarians is just as extreme as the reaction of the "think of the children" crowd.
The problem is that the faster, more powerful communication allowed by the internet has inflamed both sides of the issue. The "save the children" crowd sees that images and depictions of children being abused are being transmitted to a widespread audience and are concerned both for the safety of the original victims and for the safety of children who come into contact with consumers of this stuff. They erroneously believe that anyone who views this crap is a sexual predator in training if not in actuality. The "free speech" crowd sees the prohibition against written or drawn depictions of such abuse as unconstitutional since it restricts free speech without directly protecting any children. The cost is too high for something that doesn't provide any social benefit.
Neither side is providing an accurate view of the actual costs and benefits. To argue that all consumers of child porn will be aroused to become active predators ignores both the role sex fantasies play in people's lives and the rational fear that potential predators would have of getting caught. To argue that "artistic" depictions of imaginary children in sexual situations does nothing to advance the behavior of potential sexual predators flies in the face of the evidence provided by the entire history of human art, politics and religion (not to mention the advertising industry). Media influence human emotion and behavior for both good and evil. To take a relevant example: while the "do it for the children" movement is not entirely a media construction, it's breadth and deep emotions are continually fanned by things ranging from the Nancy Grace show on CNN to Fox News to religious sermons. Kiddie porn's message that abusing children is a turn-on and is OK is equally compelling to those who are susceptible.
If the state had the ability to weed out the susceptible people and prevent them from viewing kiddie porn, then the risk of harm would be mitigated and anyone else who was interested in this stuff could be safely left to view it. But the testing, evaluation and monitoring that would be required to implement such a system would be even more intrusive to society in general than the restrictions on free speech are today. The boundary line for drawn porn may be drawn too stringently against free speech, but the line does need to include some scope of cartoons or other drawings within its boundaries.
Except that if all the workers in the "west" are unemployed because their jobs have been taken by the impoverished masses of Asia, who is going to buy these wonderful products? Now that Americans can't charge things to their credit cards anymore, there is too much manufacturing capacity everywhere, including China.
We're in the "Midas Plague" years and we don't even know it.
Wasn't Cygnus the company that originally developed the Cygwin tools before Redhat bought them out? Or is this a different Cygnus? Perhaps they'll be facing a trademark lawsuit from Redhat while they're on the attack against MS and so on...
You call the arms race a huge mistake, I think you'll find that it's an unavoidable natural law.
At the first layer, it appears that the reason we've fallen into this trap is that the whole electronics industry is build on an "arms race" model called Moore's Law. In computers, you have to buy new hardware to run the newest software, then developers come up with new software (Vista) that exceeds the current HW capabilities. You can find similar examples in media recording formats, TVs and so on. So perhaps the arms race is just the result of the Silicon Valley business model?
You brought up the biological analogy of the parasite, and that leads us to the 2nd and 3rd layers. Apart from being based on Moore's law in the electronics industry, the "arms race" mentality is based on the scientific method itself. As hypotheses are proven or disproven by experiment, the experimental results raise new questions that need to be answered by new hypotheses. Every answer spawns a new question, just as every countermeasure against malware spawns a new form of attack (or reversion to older forms such as MBR infection).
In the third layer, the arms race reflects simply the natural law roughly analogous to the theory of evolution. The internet and its connected computers are effectively a simulation of an ecosystem. The organisms (programs) in this ecosystem interact with each other, transmitting binary information the way organisms transmit DNA or consume food. The evolution of these programs is directed by the authors of the programs, either directly by new program releases or indirectly via AI technologies. The problem is that as long as you have programs interacting on the net, you create the possibility (or opportunity) for malicious programmers to attempt to tamper with programs used by innocents.
Solutions? 1) A tamper-proof defense is impossible, unless the "general purpose" computer is limited to run the programs that come with it, and nothing new can be installed at all. We've already decided that the benefits of networking and flexible software outweigh this approach. Otherwise, you've just got a 1G or 2G locked cellphone. 2) A tamper-proof, infallible trust network. In case you haven't noticed, the infallible part is the hard part. Especially since the trust networks for different people will be different. In fact, different people will use different criteria for establishing their trust networks. And sometimes the criteria are wrong: just when you think you've got it figured out, a former NASDAQ chairman turns out to have been running a Ponzi scheme.
So unless Moses (Linus?) comes down from the mountain with a divinely inspired security system (with no code defects in the reference implementation), we're stuck with a trial and error arms race that reflects the combat between the best technologies the white hats and black hats have to offer.
Not to mention the fact that there IS a legal problem here. OP mentions the fact that most pro spammers reside outside US jurisdiction. Jurisdiction is by definition a legal/diplomatic problem.
In some ways it's similar to the "War on Terror": you have a bunch of criminals preying on peaceful neighbors and ducking across the border when things get too hot for them. Unfortunately, we can't send a bunch of Predator drones to launch missile strikes against cyber-criminals in Russia and China with the same impunity we have in Pakistan.
But don't worry, if we wait long enough, the US will be completely bankrupt and the Chinese will be much richer targets for identity theft. Then the shoe will be on the other foot. Of course that assumes the Chinese are stupid enough to fall for that invention called the "credit card".
Some apps (BlueCoat, for example) can disable screen captures (I think what it really does is control the clipboard API access) either globally or based on the source application name, but as you say, this doesn't help in the case of screen capture using an external digicam. So unless you have screen cap disabling software AND confiscate everyone's cellphones on entry into secured areas, you still have a fair risk that a determined spy can grab the data.
The problem is: How can you prevent users with job responsibilities that require them to have access to the data for client A from sharing that (directly or indirectly) with client B. There really isn't a good way to do this, since in the worst case, the user can manually copy the material onto paper or take a picture with their cellphone.
Your best approach is a group of mitigation procedures that make it difficult for information to be intercepted between you and client A, and at least provide audit trail capability for users accessing confidential information.
The bad news is that you probably have no way to win client A's trust back. They've already made the accusation, and since you didn't have any pre-existing mechanism in place to monitor and prevent, you can't investigate their claims effectively. Also, if it turns out that employees of your company shared this information as a short-cut for supporting client B, you're really screwed in terms of legal responsibility and employee ethics. You'd have to fire both the source and the recipient in the data share, just for starters.
For the future: keep confidential documents in an encrypted content-management repository with user access and rights controls that can support segregation of groups, projects and so on. Have all your clients encrypt their data with your company's public key so that there is no MITM risk for items they are sending to you over the net (or Fedex for that matter). Institute a training program that emphasizes the segregation of projects for different clients (especially competitors) unless you are developing a project that is explicitly designed and marketed as a shared or commercial offering. And institute a security policy for your employees and contractors that identifies penalties including termination of employment, civil and criminal liability if data confidentiality policies are violated. You should probably also have a project "non-compete" clause where one person cannot work on projects for competing customers within 6 months of each other (or whatever timeframe is reasonable).
You may also want to look at the physical security of your facilities. If your people are leaving confidential documents in unlocked cabinets or leaving their PCs logged in, anyone with access to the office area (visitors, delivery people, cleaning service) could have taken the information.
Actually, it's a static electric field that is net neutral outside the device, so there is no magnetic field even if you move it around. And whether this charge is dangerous when punctured depends on the voltage differential between the plates, the current carrying capacity of each plate and how many plates you actually touch with the metal implement. Put a power transistor on each plate so that reverse current flow is only allowed when the charger is plugged in, and you can prevent people from getting hit by anything more than charge from the directly affected plates. In fact they need the transistors anyway or a failure in a single plate could cause the whole (expensive) device to fail.
Especially since it isn't just m vs. M. k=kilo, K=Kelvin, h=hour and H=Henry, W=Watt and w=wombat. So are we talking about 52.220 Kelvin*Wombat*Henries here or what?
It's the dark season up there now. Nothing better for them to do, I guess. In fact, now that I think about it, this must be really old news. Mosquitos in Finland all died months ago from the cold.
If the IP address is exposed on the internet, hiding the office that owns the device does nothing to protect it. Hackers already know the device is online. There are enough ongoing attacks on every computer on the internet that it is under constant threat already.
If the IP address is not exposed directly on the internet, then identifying the office that uses the computer tells hackers nothing about the network topology. It's not like they're asking for the subnet, DHCP server and gateway.
Uhh, you can't "throw hardware" at a hardware design. In the HW manufacturing case, you WANT to spend money on the upfront design to reduce the parts cost.
If your design forced to use 1% resistors instead of 2%, you'd better have been building a medical device or something else with tightly regulated specifications. Otherwise, when your boss says to use 2% or 5%, tell him to loosen the specs. Otherwise you're just over-engineering.
Slashdot Mirror
Assuming that at least half of the places you visit will be fairly modern cities, you should take advantage of the opportunities for internet access while you are in port. Investigate places that offer free or low-cost WiFi service in the ports you'll be visiting. That will at least mitigate the low access levels you are limited to on board ship.
It's not just the ease of patching, it's also the loss of the philosophy of change management. Apart from the non-digital technology used for the change management, compare the level of detail in the change management in Tracy Kidder's The Soul of a New Machine with change management standards used nowadays.
I blame the fact that instead of using expensive Electrical or Computer Engineers, software companies base their development teams around computer science (strong on programming theory, weak on proceses) or Business IT (good business process knowledge, weak technical) people. Just because a person can code doesn't mean they can design or manage. Or vice versa.
I'm thinking when someone reports "suspicious comments" to airline personnel, they should be required to get off the plane so the FBI can interview them as well. And if the suspicions turn out to be unfounded, they get the same treatment as the victims that they falsely reported: If the airline doesn't let the "suspcious" characters back on, the people who reported their suspicions don't get back on either.
If you're going to report stuff like this, put your time and money where your mouth is.
<3 - Definitely a scrotum. Or trucknutz.
A heart symbol? I thought that was the symbol for trucknutz. <3
Advice from another user of generators after hurricane strikes:
1) Safety tip #1: You do not want to run the generator into your breaker box unless you (at a minimum) turn off the main breaker to your house. Ideally you should install one of those grid-tie disconnect devices like they use for solar installations. The breaker can be accidentally switched on, while the grid-tie can't. Without the breaker/cutoff, you are running power out to the power company. This risks three serious problems: a) providing free power to your neighbors that overloads the generator, b) the power kills a line worker after being upped to 2400 volts by a power company transformer working in reverse, and c)(not so big a risk in winter...) A lightning strike on the utility wires causes the gas tank of the generator to ignite, burning down your house.
1a) Safety tip #2 and 3: Don't ignore the ground wire for the generator. Essential to prevent sparks when you are refilling the gas. Also, the generator will require some downtime. Most are typically rated for 8 hours of continuous use.
2) The shed is overkill. Get a wheel kit so you can roll it out of the garage when needed and park it outdoors next to your mud room or something. You may want a chain and padlock too. People were having these generators stolen from their yards/driveways down here after the hurricane.
3) Not sure how much power the furnace blower needs, but a 5500 W generator would supply about 40 Amps worth of current overall, so that is probably at the low end of the required capacity. Make sure you read the rating for "continuous" power and not "peak" power.
4) Of common household appliances, coffee-makers draw surprising amounts of current. More than the microwave.
5) The router might not do much good after a while. Phone company and cable company power their systems from batteries that usually last 24-48 hours. Some sites may have generators or their techs may swap fresh batteries in.
6) You can _probably_ get by with a 5500 W generator for $700-$900, but you may want to look at a 7000 W to 10 kW generator in the $1000-$2000 range, especially if your furnace blower is rated >20 Amps.
7) Try to find one with an electric starter. Doing the old "lawnmower pull" on a frozen engine sounds like an exercise in futility.
Well, not necessarily. Evil fighting evil for worldly power is the normal state of affairs.
One reason for separation of powers in the US constitution and the adversarial construction of our courts is to provide a mechanism for greedy scumbags to rat each other out. The founders recognized that in every corrupt scheme there comes a point where some of the criminals either get scared of getting caught or decide that it's more profitable to turn their ex-partners in.
Under that logic, if your original downloaded copy is to a flash drive (or CD), wouldn't right of first sale apply to the copy on the flash drive? That is just a non-volatile version of the actual bit stream that the seller sent to you. It's not a separate copy, it's the tangible storage of the original sold copy.
I suppose the record companies would like to argue that the original copy is just the original bitstream saved into the computer's RAM or in the swap file, but that logic is implementation dependent. It is physically possible to write a software program that takes the bit-stream from the seller and directly copies it to a flash drive with no more than a small memory buffer that would qualify as a fair use excerpt of the copyrighted work.
Non-technical users should not have to do research to distinguish whether their particular download software is coded to allow them first sale rights or not, when from a practical usability standpoint, the only copy of the work that they received is the copy on their flash drive. Temporary copies in RAM or swap are inaccessible by ordinary means (and use of "extra-ordinary" means would be evidence of willful copyright violation in any case). As long as they don't copy the copy that they purchased legally, the right of first sale should apply.
(IANAL) Don't necessarily have to show a full value of damages for copyright violation. Damages for copyright violation of registered works are calculated by statutory formula rather than actual damages. The $220k verdict that was recently overturned by the judge is an example of a statutory damage award.
The overrule was not because the damages were deemed unfair for a genuine case of copyright violation, but because the judge decided that "making available" was not a valid definition of a copyright violation, making his earlier jury instructions invalid. If an actual violation had occurred, the statutory damages would apply for each violation even if the actual damages caused were less.
OTOH, just linking to other sites that actually host an "unauthorized" MP3 isn't distribution or even "making available" by any definition.
The basic math is that in order to prevent hash collisions you need a one to one mapping from the input domain to the hash range. In other words, a unique hash value for every input value. Technically, it isn't a true one-to-one mapping, because there could be extra unused values in the hash range that aren't mapped from the input values, but unused values in the hash space translate into wasted memory storing unneeded bits. A one-to-one mapping is the most efficient possible mapping.
In the case of an arbitrary/random data sequence of finite length, the number of possible values of the input space is 2^n where n is the number of bits in the input.
In order to support this number of variations of the input, the length of the hash must be the same number of bits (or greater). The hash needs to establish a one-to-one mapping of potential inputs to potential outputs.
As an example, if your input data consists of 8 bits with no restrictions on the value of any given bit, you have 2^8 possible bit combinations (the full ASCII table, for example) that must be mapped into the hash. In order to successfully map that many different permutations of input, you need an 8-bit hash. If you only have a 6 bit hash, you are trying to map 2^8 (256) possible input values values into an output space that contains only 2^6 (64) values, so either some values in the domain are not mapped into the range (invalid hash/hash failure) or else there are values in the output range that are reached from multiple inputs (hash collisions).
Now a purely arbitrary or random input sequence is not the only possible condition for hashed data. If an input data sequence can be compressed losslessly due to correlation between input values or because some sequence combinations are prohibited, the hash can be shorter than the input string (but not shorter than the best possible compressed version of the string). Again, this boils down to the basic math of one-to-one mappings and permutations of n bits of data. If data is losslessly compressible, that means a one to one mapping can be created from the uncompressed data to the compressed version. One to one mappings are transitive, so you can create a perfect hash of the compressed data without data loss (or hash collision).
The real problem with the new formula is that it punishes the people driving small efficient cars vs. the gas-guzzlers. So if I'm driving an SUV that gets 12 MPG, I pay only 16 cents per gallon using the per-mile formula (minus any out of state mileage) vs. 25 cents (or whatever) under the regular per-gallon tax. If I'm driving an efficient car (that incidentally does less damage to the highways being maintained) that gets 30 MPG, I'm paying 40 cent a gallon in taxes under the mileage formula.
Since the actual formula for highway maintenance costs involve the distance travelled per vehicle, the weight of the vehicle, the speed of the vehicle and the designed load capacity of the highway, a per mile/per vehicle ton tax is a fair tax, if it is universal. Allowing a gas-guzzler loophole for vehicles that don't conform to the new system inverts the reward mechanism from the desired goal of cutting fuel consumption.
Or put the tax on the tires: of course 1.3 cents a mile for a set of 30,000 mile tires would add about $400 to the price of a set of tires. A bit hard to swallow that all at once.
GPS is way too intrusive and not necessarily all that accurate. Soon they'd start using it to catch speeders.
Probably the best way to do this is a combo of vehicle registration, gas, tire, engine oil and electric taxes, plus any other alterna-fuels that catch on. They can achieve the stated result of proper funding for Oregon highways without forcing tech on people that would enable so much more surveillance.
Whether or not the abuser is known to the child or is a stranger doesn't change the fact that the child is being abused. The emotional reaction of the free speech libertarians is just as extreme as the reaction of the "think of the children" crowd.
The problem is that the faster, more powerful communication allowed by the internet has inflamed both sides of the issue. The "save the children" crowd sees that images and depictions of children being abused are being transmitted to a widespread audience and are concerned both for the safety of the original victims and for the safety of children who come into contact with consumers of this stuff. They erroneously believe that anyone who views this crap is a sexual predator in training if not in actuality. The "free speech" crowd sees the prohibition against written or drawn depictions of such abuse as unconstitutional since it restricts free speech without directly protecting any children. The cost is too high for something that doesn't provide any social benefit.
Neither side is providing an accurate view of the actual costs and benefits. To argue that all consumers of child porn will be aroused to become active predators ignores both the role sex fantasies play in people's lives and the rational fear that potential predators would have of getting caught. To argue that "artistic" depictions of imaginary children in sexual situations does nothing to advance the behavior of potential sexual predators flies in the face of the evidence provided by the entire history of human art, politics and religion (not to mention the advertising industry). Media influence human emotion and behavior for both good and evil. To take a relevant example: while the "do it for the children" movement is not entirely a media construction, it's breadth and deep emotions are continually fanned by things ranging from the Nancy Grace show on CNN to Fox News to religious sermons. Kiddie porn's message that abusing children is a turn-on and is OK is equally compelling to those who are susceptible.
If the state had the ability to weed out the susceptible people and prevent them from viewing kiddie porn, then the risk of harm would be mitigated and anyone else who was interested in this stuff could be safely left to view it. But the testing, evaluation and monitoring that would be required to implement such a system would be even more intrusive to society in general than the restrictions on free speech are today. The boundary line for drawn porn may be drawn too stringently against free speech, but the line does need to include some scope of cartoons or other drawings within its boundaries.
Except that if all the workers in the "west" are unemployed because their jobs have been taken by the impoverished masses of Asia, who is going to buy these wonderful products? Now that Americans can't charge things to their credit cards anymore, there is too much manufacturing capacity everywhere, including China.
We're in the "Midas Plague" years and we don't even know it.
Wasn't Cygnus the company that originally developed the Cygwin tools before Redhat bought them out? Or is this a different Cygnus? Perhaps they'll be facing a trademark lawsuit from Redhat while they're on the attack against MS and so on...
You call the arms race a huge mistake, I think you'll find that it's an unavoidable natural law.
At the first layer, it appears that the reason we've fallen into this trap is that the whole electronics industry is build on an "arms race" model called Moore's Law. In computers, you have to buy new hardware to run the newest software, then developers come up with new software (Vista) that exceeds the current HW capabilities. You can find similar examples in media recording formats, TVs and so on. So perhaps the arms race is just the result of the Silicon Valley business model?
You brought up the biological analogy of the parasite, and that leads us to the 2nd and 3rd layers. Apart from being based on Moore's law in the electronics industry, the "arms race" mentality is based on the scientific method itself. As hypotheses are proven or disproven by experiment, the experimental results raise new questions that need to be answered by new hypotheses. Every answer spawns a new question, just as every countermeasure against malware spawns a new form of attack (or reversion to older forms such as MBR infection).
In the third layer, the arms race reflects simply the natural law roughly analogous to the theory of evolution. The internet and its connected computers are effectively a simulation of an ecosystem. The organisms (programs) in this ecosystem interact with each other, transmitting binary information the way organisms transmit DNA or consume food. The evolution of these programs is directed by the authors of the programs, either directly by new program releases or indirectly via AI technologies. The problem is that as long as you have programs interacting on the net, you create the possibility (or opportunity) for malicious programmers to attempt to tamper with programs used by innocents.
Solutions? 1) A tamper-proof defense is impossible, unless the "general purpose" computer is limited to run the programs that come with it, and nothing new can be installed at all. We've already decided that the benefits of networking and flexible software outweigh this approach. Otherwise, you've just got a 1G or 2G locked cellphone. 2) A tamper-proof, infallible trust network. In case you haven't noticed, the infallible part is the hard part. Especially since the trust networks for different people will be different. In fact, different people will use different criteria for establishing their trust networks. And sometimes the criteria are wrong: just when you think you've got it figured out, a former NASDAQ chairman turns out to have been running a Ponzi scheme.
So unless Moses (Linus?) comes down from the mountain with a divinely inspired security system (with no code defects in the reference implementation), we're stuck with a trial and error arms race that reflects the combat between the best technologies the white hats and black hats have to offer.
Not to mention the fact that there IS a legal problem here. OP mentions the fact that most pro spammers reside outside US jurisdiction. Jurisdiction is by definition a legal/diplomatic problem.
In some ways it's similar to the "War on Terror": you have a bunch of criminals preying on peaceful neighbors and ducking across the border when things get too hot for them. Unfortunately, we can't send a bunch of Predator drones to launch missile strikes against cyber-criminals in Russia and China with the same impunity we have in Pakistan.
But don't worry, if we wait long enough, the US will be completely bankrupt and the Chinese will be much richer targets for identity theft. Then the shoe will be on the other foot. Of course that assumes the Chinese are stupid enough to fall for that invention called the "credit card".
Some apps (BlueCoat, for example) can disable screen captures (I think what it really does is control the clipboard API access) either globally or based on the source application name, but as you say, this doesn't help in the case of screen capture using an external digicam. So unless you have screen cap disabling software AND confiscate everyone's cellphones on entry into secured areas, you still have a fair risk that a determined spy can grab the data.
The problem is: How can you prevent users with job responsibilities that require them to have access to the data for client A from sharing that (directly or indirectly) with client B. There really isn't a good way to do this, since in the worst case, the user can manually copy the material onto paper or take a picture with their cellphone.
Your best approach is a group of mitigation procedures that make it difficult for information to be intercepted between you and client A, and at least provide audit trail capability for users accessing confidential information.
The bad news is that you probably have no way to win client A's trust back. They've already made the accusation, and since you didn't have any pre-existing mechanism in place to monitor and prevent, you can't investigate their claims effectively. Also, if it turns out that employees of your company shared this information as a short-cut for supporting client B, you're really screwed in terms of legal responsibility and employee ethics. You'd have to fire both the source and the recipient in the data share, just for starters.
For the future: keep confidential documents in an encrypted content-management repository with user access and rights controls that can support segregation of groups, projects and so on. Have all your clients encrypt their data with your company's public key so that there is no MITM risk for items they are sending to you over the net (or Fedex for that matter). Institute a training program that emphasizes the segregation of projects for different clients (especially competitors) unless you are developing a project that is explicitly designed and marketed as a shared or commercial offering. And institute a security policy for your employees and contractors that identifies penalties including termination of employment, civil and criminal liability if data confidentiality policies are violated. You should probably also have a project "non-compete" clause where one person cannot work on projects for competing customers within 6 months of each other (or whatever timeframe is reasonable).
You may also want to look at the physical security of your facilities. If your people are leaving confidential documents in unlocked cabinets or leaving their PCs logged in, anyone with access to the office area (visitors, delivery people, cleaning service) could have taken the information.
Actually, it's a static electric field that is net neutral outside the device, so there is no magnetic field even if you move it around. And whether this charge is dangerous when punctured depends on the voltage differential between the plates, the current carrying capacity of each plate and how many plates you actually touch with the metal implement. Put a power transistor on each plate so that reverse current flow is only allowed when the charger is plugged in, and you can prevent people from getting hit by anything more than charge from the directly affected plates. In fact they need the transistors anyway or a failure in a single plate could cause the whole (expensive) device to fail.
Especially since it isn't just m vs. M. k=kilo, K=Kelvin, h=hour and H=Henry, W=Watt and w=wombat. So are we talking about 52.220 Kelvin*Wombat*Henries here or what?
It's the dark season up there now. Nothing better for them to do, I guess. In fact, now that I think about it, this must be really old news. Mosquitos in Finland all died months ago from the cold.
Not anymore...
This security argument is completely bogus.
If the IP address is exposed on the internet, hiding the office that owns the device does nothing to protect it. Hackers already know the device is online. There are enough ongoing attacks on every computer on the internet that it is under constant threat already.
If the IP address is not exposed directly on the internet, then identifying the office that uses the computer tells hackers nothing about the network topology. It's not like they're asking for the subnet, DHCP server and gateway.
Uhh, you can't "throw hardware" at a hardware design. In the HW manufacturing case, you WANT to spend money on the upfront design to reduce the parts cost.
If your design forced to use 1% resistors instead of 2%, you'd better have been building a medical device or something else with tightly regulated specifications. Otherwise, when your boss says to use 2% or 5%, tell him to loosen the specs. Otherwise you're just over-engineering.