If it only tells your location when the code is entered, then it does nothing except invade your privacy at 8:00pm, and is primarily a waste of taxpayer money. It's easy to determine whether or not you're at school by looking at attendance records or having you sign in.
However, in the article they imply that they can find your location at other times:
Parents will be responsible for paying for lost devices. But Miller said that rarely happens. They are tracking devices and typically can be found immediately.
If they can find it when it's lost, then they can find it whenever they want.
No, it wasn't. Reality is that truancy is seldom actually prosecuted.
Attendance requirements are about school funding, not actually caring whether or not the kids are in school. If the kid is there 50% of the time, that's still more money for the school than if they get sent to juvenile detention and are there 0% of the time. The school will still prosecute the most egregious offenders to "set an example" but won't go after the small fry who miss 10-20 days of school in a year.
My point above is that with this program the school found a way to effectively bully the parents into agreeing to a horribly over-the-top program by threatening to send the kids to jail, even if they would seldom actually go that far.
NOTE: I actually think having a mentor call the kids every school day (another part of this program) is a good idea, and probably has more to do with the success of the program than the GPS tracking devices. The GPS tracking is just dumb. Having them sign in and out at the office would accomplish the same thing.
If the District Attorney chooses to prosecute, truant students could be sentenced to juvenile hall and parents could face up to a $2,000 fine, Pardo said.
In other words, if you miss school, we track you or you go to jail.
Apple has decided (probably correctly for them) that the processor is a commodity, not a product differentiator. Apple has figured out that it's better for them to use the same processor chip everyone else does, and differentiate their products with software and industrial design.
The problem with using your own unique high-performance processors is that if you're ahead you might see a small benefit, but if you ever get behind, you lose big. This is almost exactly what happened with PowerPC and the move to Intel. Apple learned that since people don't buy Macs for the processors, it would be silly to invest in building something at best marginally faster/cheaper than the "industry standard" Intel processors.
Mobile phone processors are a little different, because there you're not really building the processor, you're using an ARM design and tacking on a few custom circuits. The costs of doing this are much lower, and you do it for peripheral cost/features, not to be better than your competitors.
Google was indexing the data of others. That's what they do. Bing is also indexing the data of others, but in addition to doing some of their own indexing, they copy some of the indexing google did.
Google sells an index they generate of data that other people sell. Bing also does this, but at least partially populates their index with data obtained from lookups in Google's index. Neither Bing nor Google is copying customer data, but Bing IS indirectly copying Google's index.
Consider: I spend a lot of time and money evaluating plumbers in my town, and create an ad-supported website with "good plumbers" and "bad plumbers" sections, each with links. Google can't tell, based purely on my index, which plumbers are good and which are bad, they just see the plumber links. As a result, all of the plumbers on my site get equal weight in Google. Bing sees that the vast majority of click-through links are to the "good plumbers section" so whenever someone searches for "plumber" on Bing, they get only the good plumber links. Bing has effectively stolen the value from my index.
Feigned outrage at "click fraud" aside, he agreed with Google.
In fact, he said Google was right on the money.
To be clear, they both agree on Microsoft's actions, but not on the meaning of it. Microsoft's use of anonymized click data meant that they observed the results that people were clicking on for these unusual searches. They also both agree that this is just one of many items Microsoft uses to determine relevance. Google even agrees that this isn't so much intentional copying, but an effect of the use of click data.
The Microsoft guy did give a little information on why the results improved so much at one time, which DOES somewhat soften the original cause for the accusation. In other words, our searches got better because we made them better, which is probably true.
Let's be clear: At NO POINT does the Microsoft guy deny the accusations. He confirms them.
Whether or not having your software notice what people are clicking on for your competitor's searches and considering that as an input to your search results constitutes copying their results is left to the reader.
For TCP connections, dropped packets are actually a good thing, and they're part of how the protocol is supposed to work.
The way TCP is supposed to work is that the sender sends out a bunch of packets and waits to see if any are dropped. Then it sends a few more, and a few more. It keeps doing this until it detects packet loss and slows down and resends the dropped packets. The end result of this slowing down is supposed to be that eventually the line is full, but packets aren't sent any faster than the line can handle them.
The problem with large buffers in the middle is that they mess up the analysis of the line capacity. TCP starts sending packets much faster than the capacity of the line. When those packets get stuck in the buffer, it thinks they were dropped and resends them. The first buffered packets then make it to the receiver, and TCP thinks it can speed up again. Problem is, the new packets get stuck behind (or even get dropped due to) the duplicate packets which are now needlessly clogging up the buffer.
There may be ways to alter the TCP congestion analysis algorithms to better handle buffered connections, but the buffers also need to be sensible. If the packet has been in the buffer long enough that the TCP connection is going to think that it was lost, then it needs to be removed from the buffer.
NOTE: This is similar to the congestion/bandwidth problems that come up with high-bandwidth TCP-over-TCP connections like SSH port forwarding.
I remember running a BBS at 2400 and connecting to it with (no joke) the 300 baud modem from my friend's TI 99/4a. THAT was slow. You could read significantly FASTER than the text would show up on the screen. I was surprised the connection even worked at that speed, but it did.
This is the biggest problem with this definitely unclear area of the law.
There's no difference under the law between a large reseller buying in a foreign country and importing into the US, and a private citizen doing the same thing.
The problem is, for a large reseller this decision is at least somewhat reasonable. For a private citizen, this is ridiculous. It requires a private citizen to receive permission FROM THE COPYRIGHT HOLDER to import any item subject to copyright into the US. Thus, this effectively makes it illegal for a private citizen to import anything manufactured and purchased in a foreign country into the US, which I don't believe was the intention. This is even true if the foreign country has a first sale doctrine similar to that of the US.
Unfortunately, there isn't an obvious legal alternative, which is the really scary part. Who determines "Lawful copy?" Does a US court have to interpret the laws of a foreign country to determine if the copy was made lawfully there? Alternatively, does the court have to apply US law to copies made in a foreign country? What if the copy was lawfully made in the foreign country, but the copy would not have been legal in the US? What if it would have been legal in the US but was illegal in the foreign country?
One possible way to handle this is through a series of treaties and implementing laws that would effectively cause first sale in a foreign country to create a "first sale" in the US, with some reasonable way of ensuring that the proper courts determine legality of manufacture. However, I wouldn't want the same people writing ACTA to be writing that treaty.
Another possible solution would be to restrict first sale of foreign goods to only apply to copes sold by the copyright holder or made with the copyright holder's permission and sold. This would specifically exclude "lawfully made" copies that were made without the copyright holder's permission, but would cover most of the cases that we're really worried about here. Also, I think the excluded copies are what congress was really trying to restrict when they created the import restriction anyway.
While I don't like it, US law says that importation constitutes distribution, which is an exclusive right of the copyright holder which can only be exercised with their permission.
First sale doctrine says that the copyright holder gives up their distribution rights to a particular copy when they "lawfully make" or sell that copy.
What the Ninth circuit decided was that the importation restriction would be meaningless if the first sale doctrine trumps the importation restriction when the sale occurred in a foreign country, and the good was also manufactured in a foreign country. They actually relied almost completely on existing Ninth circuit precedent to make this determination.
Swiss law can't control importation into the US, only exportation from Switzerland, so it would have to be US law being violated here.
I HATE that subversion doesn't have proper tagging and (to a lesser extent) branching. Our IT team wrote gobs of scripts to make SVN behave somewhat like CVS does for tagging and branching. From what I've read, they're not the only ones.
Sometimes what I want is a tag. Ideally, a tag is a memorable string, such as "gold"
A tag isn't a long, easy to mis-type string such as "file:///foo/bar/path/to/svn/repository/some/more/stuff/tags/gold"
For most users, I want to be able to tell them: update to gold, make your changes there, then check them in to the shared repository, to be built into the next gold model.
With CVS, this is: cvs update -r gold; work; cvs update -A; cvs ci -m "gosh that was easy."
With SVN, this is: svn switch file:///arcane/long/path/to/svn/tags/gold ; work ; svn switch file:///arcane/long/path/to/svn/trunk ; svn ci -m "Why are SVN tags so long?"
With CVS, if I check in with an outdated tag on the trunk, it'll fail and tell me. With SVN, it will blindly alter the contents of the tag.
Honestly, I understand why the SVN developers wanted to separate policy from function, but in the case of tagging and branching syntax, that was a mistake. They should have created a bunch of syntactic sugar for doing things like tagging and branching using a "standard" server-side layout where most of the ugly path names are hidden from the user and tags behave like named versions (i.e. tags) and branches behave like branches instead of independent directories, all of which can use short, easy-to-remember names. If the administrator chooses to use some other non-standard repository layout, that's their problem, and the syntactic sugar will just fail.
My suspicion is that, in practice, this is no better than unsigned Diffie-Hellman, because it is subject to exactly the same man-in-the-middle.
The attacker jams the "official" wi-fi. They then set up their own "unofficial" wi-fi and advertise for connections. The laptops are just looking for whatever access point is available, so they silently connect to the "unofficial" one. The attacker then unjams the official one and funnels traffic through it, thus successfully intercepting all traffic.
*IF* you could convince the OS people to get involved, they could send a warning every time the user connects to an unsigned wi-fi connection, but I suspect the average user would just blindly click it. Even the savvy user would have to try and remember "Does joe's coffee shop have an unsigned connection all the time, or was that suzie's coffee palace?"
However, it takes more than a simple laptop with a standard card to *directionally* jam a wi-fi signal. The guy with the cantenna and a laptop in the coffee shop looks at least a little suspicious, particularly if a bunch of other customers who were already connected start noticing that their wifi reception just went to pot.
Either way, this is about raising the cost of attack. Once it starts taking additional HARDWARE to tap a wi-fi signal, you've already significantly reduced the number of attackers.
And I absolutely agree, some kind of public-key infrastructure would be best, but would in practice likely be subject to effectively the same MitM attack, since even if the standard laptop "prefers" the signed connection, it will probably accept the unsigned one if it couldn't find the signed one (because the attacker is jamming it, see above.)
In other words, the designers of WPA2 screwed up by not using something like Diffie-Hellman to negotiate a private connection before the initial password even changed hands?
I realize this would be subject to man-in-the-middle, but that would seem to be detectable as you would get two different responses when you tried to do the initial negotiation, after which the OS should report "something's screwy with this network" and refuse to connect.
The "free velocity" description is the best explanation I've seen of how this works, and also explains why the treadmill test is a good test.
The idea is that, at windspeed in a traditional sail-type vehicle, there is zero force on the sail and the ground is moving at a constant speed below you.
In a downwind-faster-than-the-wind car, we use some of the "energy" in that moving ground to spin a propeller. So long as the resultant force exerted on the air by the propeller is greater than the frictional force from the ground used to drive the propeller, the system accelerates to faster than the wind.
Determining if this is theoretically possible becomes simply: Is it possible for a theoretical propeller to exert more force on the air than is used to drive the gear system of that propeller?
Based on these real-world tests, the answer to that question appears to be yes.
I DON'T block ads. Generally they aren't too annoying, and keeping yet another application up-to-date on my computer (ad blocker) just doesn't seem worth the effort.
That said, if I have to start deciphering CAPTCHAs to get past the ads that most definitely would cross the line to where getting ad-blocking software would be worth it.
I'd almost be willing to pay money to the ad-blocking company (not the news site) to get rid of CAPTCHA ads.
I have the technical ability to install ad-blocking software. I choose not to. Make ads sufficiently annoying and I definitely WILL install such software.
Why do advertisers not get that this is a case where if you fight the ad-blockers in such a way that you sufficiently upset the ad-viewers, the ad-viewers will just turn into ad-blockers? You won't win the arms race with the ad-blockers, history has consistently shown that. The only way for the advertisers to win is to keep the ad-viewing customers that they have.
My first thought was that I'd like to hosts/firewall these people out of existence. If I thought it would work, I'd ask my ISP to do the same, as this website represents a significant financial risk to their users, as much, if not more so, than the typical phishing site.
The explanation would be "This website purports to be an actual news site. It is, in actuality, a malicious website intended to fool users into agreeing to pay them significant amounts of money."
I don't think you correctly read the parent poster. He said charge the cost of the benefit (putting out the fire) plus 10%.
This would be like your disability insurance company saying: "Okay, you want us to pay you 50k/yr, that's fine. Pay us 55k/yr and we'd be happy to pay you 50k/yr." The reason this doesn't work for something like disability insurance is that the benefit is in the form of pure cash.
Health care already works this way. You can pay $500/month for health insurance, or you can pay out-of-pocket for every procedure. There are increased billing costs, etc. from paying outright, as well as risks that the customer won't pay, but overall it mostly works.
Now, I doubt 10%, is a large enough overhead, it's probably more like double the cost of fighting the fire, but either way it theoretically works. The biggest potential problem here is that if the cost of fighting the fire is really in the $15,000 range then double that is likely to be more than the rural customer's land is worth, in which case the right thing for them to do fiscally is to let it burn. However, they won't be able to make that kind of rational decision while watching their house burn to the ground. If the cost of fighting the fire is more in the $3000-5000 range, then paying to have the fire fought would be the right thing to do.
Let's run some numbers: about 2 million fires a year in US. about 100 million households in US.
This means a fire occurs about 1 in every 50 household-years.
The fire department changes $75/household-year for rural coverage. So, the cost of a fire should be around $3750/fire. If they charged $7500/fire that should about offset their costs, including billing costs, unless the city is significantly subsidizing rural firefighting costs even with the $75/year fee. $7500/fire is low enough that many individuals should consider paying even on a per-fire basis.
What I don't understand from all this NAT stuff is that NAT (for the home user) is mainly about 1 thing: COST. I can spend $50 on a router and save $20 for a switch and $5 per month to get a second IP. The router pays for itself in 6 months, and that's just to connect 2 computers. I have 4 connected to my router right now.
I have no reason to believe that the numbers would be significantly different on IPv6. My ISP would probably still charge to get additional IPs on my network, which makes NAT a major cost savings.
I realize from an architecture perspective that NAT is ugly. However, it represents a real cost savings on real networks. The security discussion is just a red herring for cost, which is the real reason for home NAT deployments, which are probably the most common.
This seems to be the design pictured (at least in the second photo) but I'm not sure I buy it. Assuming they could all focus based on the lens settings, this would still only mean, say, 16 independently controllable areas and a standard all-over fill. That's going to look like crap unless your subject happens to completely fill one of those 16 areas.
I'm guessing this only starts to get interesting in the 256-area range. However, I'm not sure that can be done cost-effectively.
If they've come up with a way to have a dynamically controlled gobo/lens that redirects a single flash but still fits within a compact camera, that's pretty darn impressive, but doesn't seem to be disclosed in this patent. Without it, this is a bit like "Method for generating infinite power: Start with a perpetual motion machine..."
Seriously, the interesting part of this idea isn't in the user interface, it's in how you build that redirector.
Agnosticism says that an omniscient god is inherently unknowable. If a god existed whose only interaction with the universe was to create the universe simply by creating the LAWS of the universe, that's (probably provably) impossible to observe from within the system of the universe.
A gigantic invisible space monkey spinning the earth makes specific predictions that can be tested. Specifically, it predicts that the earth will be spinning at a rate other than that which would occur based only on the laws of physics that are observed to hold true in other situations.
Similarly, faeries, alien abductions, homeopathy, and astrology all make specific predictions that can be tested or at least which should be observed statistically. No one can agree on what exactly a ghost is, but many of the ideas also result in specific, testable hypotheses. Reincarnation mostly results in untestable hypotheses, so this falls more in the unknowable area.
For instance, I don't believe fairies exist, but in the same way I don't believe I'll be dining at a 5-star restaurant tonight. However, if you brought me a fairy in a jar, it wouldn't particularly change my belief systems, it would just be an interesting new animal. Until then, I would simply say that it's unlikely that LOTS of fairies exist, particularly in Europe, where there is a limited amount of undeveloped land.
As a proven true example, acupuncture does significantly better than placebo for treating pain in clinical trials. That's a surprising result, but one a scientific mind can accept and then attempt to further explain.
Agnostics believe that until the existence of an omnipotent god results in specific, testable hypotheses, then it is inherently unknowable.
The entire IPv4 address space is around 4 Billion addresses. At 150,000 per day, they will cover the entire IPv4 address space in around 27,000 days, which is only 73 years.
By contrast, the population of Europe is a relatively modest 731 million. At that rate, they could have the entire population of Europe in about 13 years.
150,000 IPs per day is RIDICULOUS, and a real burden on the ISPs. Something like $5 per IP is extremely reasonable and would reduce this to a sensible rate.
I have enough trouble getting my kids to put their lunchbox in the kitchen rather than leaving it in their backpack overnight. I seriously doubt that kids can be counted on to recharge the iPad daily. Thus, having textbooks with a battery IS a major issue. Only if you told me that it would hold up to over a week's worth of use would I believe that it is "good enough." That's about 5 hours a day for 6 days, so 30 hours of use (display on, mostly showing static text) per charge, to ensure that most students "only" don't have their textbook once a week.
That's a fascinating analysis, and I think it makes a lot of sense.
This could be a way of Intel screwing the PC manufacturers by selling directly to the consumer. To the consumer it may look like an upgrade costs the same as the more expensive processor, but Intel makes more profit while the PC maker makes less. If an upgraded part costs less overall than an initially full-spec part then this is almost certainly the case.
It's actually a little more complicated. You see what speeds the chips come out at, and then look at how you can set multiple price points to make the most money by selling the slowest chips at a discount and selling the fastest chips at a premium while throwing away the very slowest chips.
Sometimes, that means you have to sell chips that would work at a higher sort point in a lower bin.
The idea here is to sell something like this:
2-core: $200 4-core down-binned to 2-core: $220 (Market as 2-core upgradeable to 4-core) 4-core: $300
Charge $100 to upgrade the upgradeable parts.
For the consumer, this means: 1.) If you need a 4-core now, buy one. 2.) If a 2-core is all you will ever need, buy that. 3.) If you might need a 4-core sometime in the future, but you might buy a new PC instead, buy the upgradeable chip.
Now, my guess is that the average consumer is going to buy either the fastest chip (which they often don't need) or the cheapest chip (which is usually good enough), so I don't think anyone would buy the upgradeable chip. I also think they'd be right, in that if you don't need the 4-core now, you are probably better off just buying a new PC when you do.
Slashdot Mirror
However, in the article they imply that they can find your location at other times:
Parents will be responsible for paying for lost devices. But Miller said that rarely happens. They are tracking devices and typically can be found immediately.
If they can find it when it's lost, then they can find it whenever they want.
No, it wasn't. Reality is that truancy is seldom actually prosecuted.
Attendance requirements are about school funding, not actually caring whether or not the kids are in school. If the kid is there 50% of the time, that's still more money for the school than if they get sent to juvenile detention and are there 0% of the time. The school will still prosecute the most egregious offenders to "set an example" but won't go after the small fry who miss 10-20 days of school in a year.
My point above is that with this program the school found a way to effectively bully the parents into agreeing to a horribly over-the-top program by threatening to send the kids to jail, even if they would seldom actually go that far.
NOTE: I actually think having a mentor call the kids every school day (another part of this program) is a good idea, and probably has more to do with the success of the program than the GPS tracking devices. The GPS tracking is just dumb. Having them sign in and out at the office would accomplish the same thing.
If the District Attorney chooses to prosecute, truant students could be sentenced to juvenile hall and parents could face up to a $2,000 fine, Pardo said.
In other words, if you miss school, we track you or you go to jail.
Been there, done that, DON'T want to do it again.
It was called: PowerPC.
Apple has decided (probably correctly for them) that the processor is a commodity, not a product differentiator. Apple has figured out that it's better for them to use the same processor chip everyone else does, and differentiate their products with software and industrial design.
The problem with using your own unique high-performance processors is that if you're ahead you might see a small benefit, but if you ever get behind, you lose big. This is almost exactly what happened with PowerPC and the move to Intel. Apple learned that since people don't buy Macs for the processors, it would be silly to invest in building something at best marginally faster/cheaper than the "industry standard" Intel processors.
Mobile phone processors are a little different, because there you're not really building the processor, you're using an ARM design and tacking on a few custom circuits. The costs of doing this are much lower, and you do it for peripheral cost/features, not to be better than your competitors.
I agree with you, but you got the metaphor wrong.
Google was indexing the data of others. That's what they do. Bing is also indexing the data of others, but in addition to doing some of their own indexing, they copy some of the indexing google did.
Google sells an index they generate of data that other people sell. Bing also does this, but at least partially populates their index with data obtained from lookups in Google's index. Neither Bing nor Google is copying customer data, but Bing IS indirectly copying Google's index.
Consider: I spend a lot of time and money evaluating plumbers in my town, and create an ad-supported website with "good plumbers" and "bad plumbers" sections, each with links. Google can't tell, based purely on my index, which plumbers are good and which are bad, they just see the plumber links. As a result, all of the plumbers on my site get equal weight in Google. Bing sees that the vast majority of click-through links are to the "good plumbers section" so whenever someone searches for "plumber" on Bing, they get only the good plumber links. Bing has effectively stolen the value from my index.
Feigned outrage at "click fraud" aside, he agreed with Google.
In fact, he said Google was right on the money.
To be clear, they both agree on Microsoft's actions, but not on the meaning of it. Microsoft's use of anonymized click data meant that they observed the results that people were clicking on for these unusual searches. They also both agree that this is just one of many items Microsoft uses to determine relevance. Google even agrees that this isn't so much intentional copying, but an effect of the use of click data.
The Microsoft guy did give a little information on why the results improved so much at one time, which DOES somewhat soften the original cause for the accusation. In other words, our searches got better because we made them better, which is probably true.
Let's be clear: At NO POINT does the Microsoft guy deny the accusations. He confirms them.
Whether or not having your software notice what people are clicking on for your competitor's searches and considering that as an input to your search results constitutes copying their results is left to the reader.
For TCP connections, dropped packets are actually a good thing, and they're part of how the protocol is supposed to work.
The way TCP is supposed to work is that the sender sends out a bunch of packets and waits to see if any are dropped. Then it sends a few more, and a few more. It keeps doing this until it detects packet loss and slows down and resends the dropped packets. The end result of this slowing down is supposed to be that eventually the line is full, but packets aren't sent any faster than the line can handle them.
The problem with large buffers in the middle is that they mess up the analysis of the line capacity. TCP starts sending packets much faster than the capacity of the line. When those packets get stuck in the buffer, it thinks they were dropped and resends them. The first buffered packets then make it to the receiver, and TCP thinks it can speed up again. Problem is, the new packets get stuck behind (or even get dropped due to) the duplicate packets which are now needlessly clogging up the buffer.
There may be ways to alter the TCP congestion analysis algorithms to better handle buffered connections, but the buffers also need to be sensible. If the packet has been in the buffer long enough that the TCP connection is going to think that it was lost, then it needs to be removed from the buffer.
NOTE: This is similar to the congestion/bandwidth problems that come up with high-bandwidth TCP-over-TCP connections like SSH port forwarding.
No way can I read at 2400 baud.
I remember running a BBS at 2400 and connecting to it with (no joke) the 300 baud modem from my friend's TI 99/4a. THAT was slow. You could read significantly FASTER than the text would show up on the screen. I was surprised the connection even worked at that speed, but it did.
This is the biggest problem with this definitely unclear area of the law.
There's no difference under the law between a large reseller buying in a foreign country and importing into the US, and a private citizen doing the same thing.
The problem is, for a large reseller this decision is at least somewhat reasonable. For a private citizen, this is ridiculous. It requires a private citizen to receive permission FROM THE COPYRIGHT HOLDER to import any item subject to copyright into the US. Thus, this effectively makes it illegal for a private citizen to import anything manufactured and purchased in a foreign country into the US, which I don't believe was the intention. This is even true if the foreign country has a first sale doctrine similar to that of the US.
Unfortunately, there isn't an obvious legal alternative, which is the really scary part. Who determines "Lawful copy?" Does a US court have to interpret the laws of a foreign country to determine if the copy was made lawfully there? Alternatively, does the court have to apply US law to copies made in a foreign country? What if the copy was lawfully made in the foreign country, but the copy would not have been legal in the US? What if it would have been legal in the US but was illegal in the foreign country?
One possible way to handle this is through a series of treaties and implementing laws that would effectively cause first sale in a foreign country to create a "first sale" in the US, with some reasonable way of ensuring that the proper courts determine legality of manufacture. However, I wouldn't want the same people writing ACTA to be writing that treaty.
Another possible solution would be to restrict first sale of foreign goods to only apply to copes sold by the copyright holder or made with the copyright holder's permission and sold. This would specifically exclude "lawfully made" copies that were made without the copyright holder's permission, but would cover most of the cases that we're really worried about here. Also, I think the excluded copies are what congress was really trying to restrict when they created the import restriction anyway.
While I don't like it, US law says that importation constitutes distribution, which is an exclusive right of the copyright holder which can only be exercised with their permission.
First sale doctrine says that the copyright holder gives up their distribution rights to a particular copy when they "lawfully make" or sell that copy.
What the Ninth circuit decided was that the importation restriction would be meaningless if the first sale doctrine trumps the importation restriction when the sale occurred in a foreign country, and the good was also manufactured in a foreign country. They actually relied almost completely on existing Ninth circuit precedent to make this determination.
Swiss law can't control importation into the US, only exportation from Switzerland, so it would have to be US law being violated here.
The two are NOT equivalent.
I HATE that subversion doesn't have proper tagging and (to a lesser extent) branching. Our IT team wrote gobs of scripts to make SVN behave somewhat like CVS does for tagging and branching. From what I've read, they're not the only ones.
Sometimes what I want is a tag. Ideally, a tag is a memorable string, such as "gold"
A tag isn't a long, easy to mis-type string such as "file:///foo/bar/path/to/svn/repository/some/more/stuff/tags/gold"
For most users, I want to be able to tell them: update to gold, make your changes there, then check them in to the shared repository, to be built into the next gold model.
With CVS, this is: cvs update -r gold; work; cvs update -A; cvs ci -m "gosh that was easy."
With SVN, this is: svn switch file:///arcane/long/path/to/svn/tags/gold ; work ; svn switch file:///arcane/long/path/to/svn/trunk ; svn ci -m "Why are SVN tags so long?"
With CVS, if I check in with an outdated tag on the trunk, it'll fail and tell me. With SVN, it will blindly alter the contents of the tag.
Honestly, I understand why the SVN developers wanted to separate policy from function, but in the case of tagging and branching syntax, that was a mistake. They should have created a bunch of syntactic sugar for doing things like tagging and branching using a "standard" server-side layout where most of the ugly path names are hidden from the user and tags behave like named versions (i.e. tags) and branches behave like branches instead of independent directories, all of which can use short, easy-to-remember names. If the administrator chooses to use some other non-standard repository layout, that's their problem, and the syntactic sugar will just fail.
My suspicion is that, in practice, this is no better than unsigned Diffie-Hellman, because it is subject to exactly the same man-in-the-middle.
The attacker jams the "official" wi-fi. They then set up their own "unofficial" wi-fi and advertise for connections. The laptops are just looking for whatever access point is available, so they silently connect to the "unofficial" one. The attacker then unjams the official one and funnels traffic through it, thus successfully intercepting all traffic.
*IF* you could convince the OS people to get involved, they could send a warning every time the user connects to an unsigned wi-fi connection, but I suspect the average user would just blindly click it. Even the savvy user would have to try and remember "Does joe's coffee shop have an unsigned connection all the time, or was that suzie's coffee palace?"
Yes, that's absolutely true.
However, it takes more than a simple laptop with a standard card to *directionally* jam a wi-fi signal. The guy with the cantenna and a laptop in the coffee shop looks at least a little suspicious, particularly if a bunch of other customers who were already connected start noticing that their wifi reception just went to pot.
Either way, this is about raising the cost of attack. Once it starts taking additional HARDWARE to tap a wi-fi signal, you've already significantly reduced the number of attackers.
And I absolutely agree, some kind of public-key infrastructure would be best, but would in practice likely be subject to effectively the same MitM attack, since even if the standard laptop "prefers" the signed connection, it will probably accept the unsigned one if it couldn't find the signed one (because the attacker is jamming it, see above.)
In other words, the designers of WPA2 screwed up by not using something like Diffie-Hellman to negotiate a private connection before the initial password even changed hands?
I realize this would be subject to man-in-the-middle, but that would seem to be detectable as you would get two different responses when you tried to do the initial negotiation, after which the OS should report "something's screwy with this network" and refuse to connect.
Thank you.
The "free velocity" description is the best explanation I've seen of how this works, and also explains why the treadmill test is a good test.
The idea is that, at windspeed in a traditional sail-type vehicle, there is zero force on the sail and the ground is moving at a constant speed below you.
In a downwind-faster-than-the-wind car, we use some of the "energy" in that moving ground to spin a propeller. So long as the resultant force exerted on the air by the propeller is greater than the frictional force from the ground used to drive the propeller, the system accelerates to faster than the wind.
Determining if this is theoretically possible becomes simply: Is it possible for a theoretical propeller to exert more force on the air than is used to drive the gear system of that propeller?
Based on these real-world tests, the answer to that question appears to be yes.
I DON'T block ads. Generally they aren't too annoying, and keeping yet another application up-to-date on my computer (ad blocker) just doesn't seem worth the effort.
That said, if I have to start deciphering CAPTCHAs to get past the ads that most definitely would cross the line to where getting ad-blocking software would be worth it.
I'd almost be willing to pay money to the ad-blocking company (not the news site) to get rid of CAPTCHA ads.
I have the technical ability to install ad-blocking software. I choose not to. Make ads sufficiently annoying and I definitely WILL install such software.
Why do advertisers not get that this is a case where if you fight the ad-blockers in such a way that you sufficiently upset the ad-viewers, the ad-viewers will just turn into ad-blockers? You won't win the arms race with the ad-blockers, history has consistently shown that. The only way for the advertisers to win is to keep the ad-viewing customers that they have.
Really?
My first thought was that I'd like to hosts/firewall these people out of existence. If I thought it would work, I'd ask my ISP to do the same, as this website represents a significant financial risk to their users, as much, if not more so, than the typical phishing site.
The explanation would be "This website purports to be an actual news site. It is, in actuality, a malicious website intended to fool users into agreeing to pay them significant amounts of money."
I don't think you correctly read the parent poster. He said charge the cost of the benefit (putting out the fire) plus 10%.
This would be like your disability insurance company saying: "Okay, you want us to pay you 50k/yr, that's fine. Pay us 55k/yr and we'd be happy to pay you 50k/yr." The reason this doesn't work for something like disability insurance is that the benefit is in the form of pure cash.
Health care already works this way. You can pay $500/month for health insurance, or you can pay out-of-pocket for every procedure. There are increased billing costs, etc. from paying outright, as well as risks that the customer won't pay, but overall it mostly works.
Now, I doubt 10%, is a large enough overhead, it's probably more like double the cost of fighting the fire, but either way it theoretically works. The biggest potential problem here is that if the cost of fighting the fire is really in the $15,000 range then double that is likely to be more than the rural customer's land is worth, in which case the right thing for them to do fiscally is to let it burn. However, they won't be able to make that kind of rational decision while watching their house burn to the ground. If the cost of fighting the fire is more in the $3000-5000 range, then paying to have the fire fought would be the right thing to do.
Let's run some numbers:
about 2 million fires a year in US.
about 100 million households in US.
This means a fire occurs about 1 in every 50 household-years.
The fire department changes $75/household-year for rural coverage. So, the cost of a fire should be around $3750/fire. If they charged $7500/fire that should about offset their costs, including billing costs, unless the city is significantly subsidizing rural firefighting costs even with the $75/year fee. $7500/fire is low enough that many individuals should consider paying even on a per-fire basis.
What I don't understand from all this NAT stuff is that NAT (for the home user) is mainly about 1 thing: COST. I can spend $50 on a router and save $20 for a switch and $5 per month to get a second IP. The router pays for itself in 6 months, and that's just to connect 2 computers. I have 4 connected to my router right now.
I have no reason to believe that the numbers would be significantly different on IPv6. My ISP would probably still charge to get additional IPs on my network, which makes NAT a major cost savings.
I realize from an architecture perspective that NAT is ugly. However, it represents a real cost savings on real networks. The security discussion is just a red herring for cost, which is the real reason for home NAT deployments, which are probably the most common.
This seems to be the design pictured (at least in the second photo) but I'm not sure I buy it. Assuming they could all focus based on the lens settings, this would still only mean, say, 16 independently controllable areas and a standard all-over fill. That's going to look like crap unless your subject happens to completely fill one of those 16 areas.
I'm guessing this only starts to get interesting in the 256-area range. However, I'm not sure that can be done cost-effectively.
If they've come up with a way to have a dynamically controlled gobo/lens that redirects a single flash but still fits within a compact camera, that's pretty darn impressive, but doesn't seem to be disclosed in this patent. Without it, this is a bit like "Method for generating infinite power: Start with a perpetual motion machine..."
Seriously, the interesting part of this idea isn't in the user interface, it's in how you build that redirector.
Agnosticism says that an omniscient god is inherently unknowable. If a god existed whose only interaction with the universe was to create the universe simply by creating the LAWS of the universe, that's (probably provably) impossible to observe from within the system of the universe.
:)
A gigantic invisible space monkey spinning the earth makes specific predictions that can be tested. Specifically, it predicts that the earth will be spinning at a rate other than that which would occur based only on the laws of physics that are observed to hold true in other situations.
Similarly, faeries, alien abductions, homeopathy, and astrology all make specific predictions that can be tested or at least which should be observed statistically. No one can agree on what exactly a ghost is, but many of the ideas also result in specific, testable hypotheses. Reincarnation mostly results in untestable hypotheses, so this falls more in the unknowable area.
For instance, I don't believe fairies exist, but in the same way I don't believe I'll be dining at a 5-star restaurant tonight. However, if you brought me a fairy in a jar, it wouldn't particularly change my belief systems, it would just be an interesting new animal. Until then, I would simply say that it's unlikely that LOTS of fairies exist, particularly in Europe, where there is a limited amount of undeveloped land.
As a proven true example, acupuncture does significantly better than placebo for treating pain in clinical trials. That's a surprising result, but one a scientific mind can accept and then attempt to further explain.
Agnostics believe that until the existence of an omnipotent god results in specific, testable hypotheses, then it is inherently unknowable.
Same goes for string theory.
The entire IPv4 address space is around 4 Billion addresses. At 150,000 per day, they will cover the entire IPv4 address space in around 27,000 days, which is only 73 years.
By contrast, the population of Europe is a relatively modest 731 million. At that rate, they could have the entire population of Europe in about 13 years.
150,000 IPs per day is RIDICULOUS, and a real burden on the ISPs. Something like $5 per IP is extremely reasonable and would reduce this to a sensible rate.
I have enough trouble getting my kids to put their lunchbox in the kitchen rather than leaving it in their backpack overnight. I seriously doubt that kids can be counted on to recharge the iPad daily. Thus, having textbooks with a battery IS a major issue. Only if you told me that it would hold up to over a week's worth of use would I believe that it is "good enough." That's about 5 hours a day for 6 days, so 30 hours of use (display on, mostly showing static text) per charge, to ensure that most students "only" don't have their textbook once a week.
That's a fascinating analysis, and I think it makes a lot of sense.
This could be a way of Intel screwing the PC manufacturers by selling directly to the consumer. To the consumer it may look like an upgrade costs the same as the more expensive processor, but Intel makes more profit while the PC maker makes less. If an upgraded part costs less overall than an initially full-spec part then this is almost certainly the case.
It's actually a little more complicated. You see what speeds the chips come out at, and then look at how you can set multiple price points to make the most money by selling the slowest chips at a discount and selling the fastest chips at a premium while throwing away the very slowest chips.
Sometimes, that means you have to sell chips that would work at a higher sort point in a lower bin.
The idea here is to sell something like this:
2-core: $200
4-core down-binned to 2-core: $220 (Market as 2-core upgradeable to 4-core)
4-core: $300
Charge $100 to upgrade the upgradeable parts.
For the consumer, this means:
1.) If you need a 4-core now, buy one.
2.) If a 2-core is all you will ever need, buy that.
3.) If you might need a 4-core sometime in the future, but you might buy a new PC instead, buy the upgradeable chip.
Now, my guess is that the average consumer is going to buy either the fastest chip (which they often don't need) or the cheapest chip (which is usually good enough), so I don't think anyone would buy the upgradeable chip. I also think they'd be right, in that if you don't need the 4-core now, you are probably better off just buying a new PC when you do.