The problem is that the guard on the exit *didn't* notice. And the passenger who noticed wasn't allowed to go through the door, and let's face it, probably didn't relish taking the (admittedly small) chance that the person in question really was dangerous, so they found a security officer on their side of the exit to report the breach to. Of course, by then it was too late; they had video footage of the person who went through the door, but they couldn't find him. The problem wasn't that the door was unwatched, it's that every once in a while a guard is going to miss someone. They're human, they can't stare at a door for 8 or more hours straight.
Do those figures include the period after Dec. 25th? Cause another explanation could be that people are pussies and didn't fly because "the terrorists are detonating their underpants on planes". People are pathetic and easily scared; the number who opt out on personal dignity grounds is probably far less than those who are literally giving in to terror.
Just for the record, I'm only counting production uses of the code. I do code in my spare time for fun, and my experience is a bit more extensive than that spent in school and work.
That said, I will admit to having *far* less experience in C#/Java compared to C/C++ and Perl. I will admit to some prejudices in favor of C# due to the support it provides for C and C++ style idioms (without the horror that is C++ compiler errors to boot). It was also quite handy for some old number theory projects, since I could implement custom numeric types with relative ease. But having worked with both C# and Java in production environments and studying language features and performance profiles on my own time, my conclusion remains that C# is a much more developer friendly language, with arguable performance benefits (most of them aren't really noticeable in most situations, though I've coded in some environments where Java's autoboxing introduced noticeable perf problems). I'm willing to accept the somewhat reduced portability; I've never coded for anything but Windows (3 years) and Linux (6 years), so for my needs, and those of quite a lot of other people, the reduced portability doesn't really enter in to the equation.
That said, I've never personally been bitten by Microsoft on a legal issue before. If Mono is ever shut down due to legal issues, I'll probably approach Microsoft the same way I approach Sony; with a cross and holy water in hand. But until that day, I'm going to be practical and when I have a choice, choose a language that makes my job easier. The irony is that right now I'm in an all Java shop, so I don't have that choice.
My understanding is that this is partially a function of the size of the battery and the design of the charger. Traditional NiMH batteries (that is, laptop sized and smaller) weren't that good either; according to Wikipedia, even today their cycle durability (number of complete discharge/charge cycles before substantial degradation) is 500-1000, and they shed a lot of power just sitting there (20-30% per month). Li-Ions have a cycle durability of 1000-1200, and their discharge rate is only 5-10%. Li-Polymer batteries have a similar cycle durability to the Li-Ions, and around a 5% discharge rate. The reason NiMH work so well in cars is mostly economy of scale and better charging design that avoids stressing the battery; it's possible those techniques would also work for Lithium types.
In addition, the Li-Ions and especially the Li-Polys have a *much* greater efficiency. NiMH only pulls around 66%, compared to 80-90% for Li-Ion and 99% for Li-Poly. And the energy density of the NiMH is much worse; roughly a half to a third that of the Li-Ion and Li-Poly respectively.
Granted, a Lithium based battery seems to have a lifespan that is roughly half that of an NiMH (despite the better numbers of cycle durability, which is a bit confusing). Even if that is the case though, using Lithium based batteries would mean:
Substantially more efficient charge/discharge (particularly good for hybrids; a plug in vehicle can take the hit by drawing more power from the wall, while the hybrid needs to charge efficiently on the go from limited sources or it isn't doing as much good)
Either substantially lower weight, or substantially greater range for the same weight
Better charge retention on infrequently used vehicles
So, for those benefits, might it make sense to accept that you'd need to have the battery replaced roughly twice as often? If you gain an extra five miles per gallon in a hybrid (because the car is lighter and you are wasting less energy to the charge/discharge cycle), or you have a substantially greater range in an electric vehicle, that might be a worthy trade-off. Probably not a good financial trade-off just yet (replacement batteries are still fairly expensive), but economies of scale kick in over time, and in the meantime, you have a car that doesn't need refueling as often. People put too low a price on their time and energy; I consider trade-offs like this to be worthwhile in many cases (obviously, you'd need more exact numbers than I'm giving to make a real decision, but it's not as cut and dry as you think).
Wow. Imagine, an open source project cloning the functionality of a commercial product that doesn't support the latest features of the commercial product. Given what I know of MS release cycles, it took them at least a year and half to write the 3.5 release, and they frequently were writing wrappers around functionality that either already existed, or was handed to them by the OS team. It's been two years since 3.5 released, and the Mono team has got a lot of the functionality exposed, without the advantage of OS support for many of the features. Give them a break, Rome wasn't built in a day either you know.
Right, because Microsoft is making a profit off licensing the.NET framework. Wait, you mean they don't charge a cent for it? And C# is a better language than Java, with the Mono project providing cross-platform compatibility, so Windows users have an easier time migrating to Linux if they so choose? Clearly I should listen to random/.er and forswear all use of anything that "supports" "Microsoft products" in any way, including the OpenOffice; after all, it lets people read and write Office documents, and by doing so, indirectly enables the Microsoft hegemony.
P.S. Yes, C# being better than Java is personal opinion. I've used both, Java for two years in school and one and a half years in the workforce, C# for a little under a year in school and half a year in the workforce (plus a few years of various other languages, mostly C/C++ and, yes, Perl). For developers, the lack of rigid ideological adherence to OO dogma is quite helpful; delegates for callbacks and "pass-by-reference" for arguments instead of inane wrapper classes for both (yes, pedantic types, I know it's all pass by reference, but you know what I mean), not needing to think about auto-boxing as much (since.NET collections of primitives really are primitives, not boxed primitives), operator overloading and structs to enable the creation of relatively efficient and easy to use numeric types, etc. I think both languages have merit, and I think both languages are improved by the competition (e.g. without C#, I'm not sure Java would ever have introduced generics, since it violated the spirit of OO). But I'm not going to reject C# just because MS made it.
That's because a government pension subtracts from social security income. And that is because government employees under the pension system were exempt from social security taxes at the time (the government pension system was supposed to replace it for them). She's effectively combined her social security benefits combined with her workplace pension; the source is the government in any event. Your post is disingenuous to say the least.
"...a trademark application for the name was not filed [by the plaintiff] until May - when rumours about Microsoft's new product had already spread widely across the internet."
"Microsoft, meanwhile, filed its own trademark applications for the name in March - for a variety of uses, including search engine software, interface software, advertising, telecoms and for 'providing a website and website links to geographic information, map images and trip routing'."
Says it all really. This company didn't even bother trying to establish trademark rights until two months after Microsoft, after news of the new engine had leaked. This screams trademark troll.
You haven't addressed the secondary issue; that the melting in the Himalayas is only doubled by the soot, not caused by it. And the scrubbers would have little to no effect on glacier melt in the rest of the world. And that "destroying the world economy" is a politically motivated, short sighted conclusion. Most of the reasonable forecasts show it "dragging" the economy down by about 1-3% of the "GWP" (Gross World Product). The economic doomsday types like to discount the possibility that the cost of oil will increase much beyond the rate of inflation, as if the entire world can start living like Americans (or even Western Europeans) without drastically increasing the price of oil.
I believe they're supposed to act as a buffer. They accumulate water during the wet season, and release it during the dry, in roughly equal amounts. If they melt faster than they accrete, then you get more water during the dry season for a while (while the glaciers are close to their original size), then it starts to taper off as the increased melting is offset by the lesser amount of ice. Eventually, the glaciers are reduced to virtually nil, and you get little or nothing after that.
Yes, but in a summary, leaving off the "hundreds of" is fine when you're trying to express the concept of "a lot" in rough terms. I'm fairly sure (non-retarded) people can figure out that there are more than two million people in India.
Feeding a lame troll, but the source of soot is the same source as the CO2. So we're still solving the same problem. And they've already noted that the melting in the Himalayas is abnormally fast, but that doesn't change the fact that all the glaciers are melting, if "only" half as fast as the Himalayas.
To their competitors, yes (though in recent years they've bought them out instead of suing them in most cases). But do you have any reason to believe they are going to violate their customers' privacy (at least, any more so than Google)?
There are different ways of triggering a buffer overflow than a simple unbounded copy. While most of them end up as a buffer overflow, they are caused by much more subtle problems, e.g. integer overflow, which can lead to the bounds you set up being incorrect in the first place.
Beyond that, the vast majority of the buffer overflows that have occurred recently aren't in string processing code; they're network facing operations that operate on raw byte buffers. And they've always had bounds checking, but problems like integer overflow occasionally crop up; automated checks can't catch them with 100% accuracy. It's up to security review teams to do fuzzing and code reviews looking for that sort of problem.
And by banned, he doesn't just mean there is a policy against them. They run regular code scans for "illegal" functions and then send out high priority bug reports to the code owners if any are found. I had to fix a couple of them when I worked there (and they weren't for strcpy, it was more subtly problematic functions). In my entire time at MS, I never saw one instance of strcpy. Usually the code used StringCchLength and StringCchCopy, which are not only safe if called with the appropriate buffer size, but function well with both ASCII and UTF-16 strings environments (though in practice, all our code was compiled with unicode support).
I'm don't think this would trip double jeopardy even in the States. If a verdict is based upon misinterpretation of the law, and the appeals chain decides this is the case, then a retrial is feasible. The only way you hit double jeopardy is if the Supreme Court says you do (unless you decide not to appeal the circuit court ruling, though I can't imagine why you wouldn't), and it has to be due to a point of law that went awry in the original trial. Prosecutors can't just re-file charges nilly-willy. In the States, I know re-trials are permissible due to jury contamination and the like, this is just a different reason for a re-trial (and in both cases, it's the judges, not the prosecutors who make the decision that a re-trial is warranted).
Riiiight... I'm sure people *love* blowing millions (or billions) of dollars to hand out free electronics to all and sundry. Tell me, what else is different in your world?
That CD isn't actually necessary for most systems (my machine at home is on Verizon DSL, but I've got no Verizon software installed by choice, same went for the Comcast service I used to have, I just configured it on a throwaway OS so the software wouldn't clutter my machine). And like I mentioned, isn't required for global network QoS anyway. Yes, having the CD enable local QoS is a good thing in any event, but that's for the benefit of that user, it doesn't affect the ISP or its other customers in any way.
Why waste the power? A personal use DNS server is a waste; if your ISPs DNS is slow there are always alternatives (I used Verizon's DNS for years when living in an area where Comcast DNS performance was slow). I know DIY is fun, has geek cread and all that, but your local machine will cache frequently accessed sites anyway, and the benefit gained on uncached sites will be seen so infrequently that you're not benefiting.
Are you advocating a system where the ISP has mandate power over the OS configuration?
A short version of my longer answer below: They already have this power. They don't need to control your OS, because they control your modem, and your modem will obey QoS from their network, even if your OS won't obey it locally. All you'll do is overload the CPU on both OS and modem dropping and resending packets.
The thing is, a larger network has already set a cap for you, the speed you are paying for. Your modem supports it, it doesn't matter how your computer is configured. Your connection and their hardware could support a much higher speed most of the time, but they limit it to what they are capable of providing reliably and inexpensively.
With your roommate, you could simulate a similar scenario by configuring your router to limit the bandwidth of each machine to, say, (connection speed / # of machine) * 1.2. Note that each user is capable of using a little more than his fair share, but he can't completely dominate the network. Some capacity is wasted most of the time (since most people won't use their share), and the users who really want bandwidth get a little extra. If everyone tries to max it out at once, then the top download speed drops by a bit (that final 0.2), but everyone gets a decent experience because the router was enforcing the policy.
Basically what I'm saying is that your situation was caused by poor local network management. The ISPs are perfectly capable of enforcing good global network management on their users (local management doesn't matter, attempts to hog bandwidth by local sources are filtered by your modem, which they do control), so your problem is inapplicable.
Slashdot Mirror
The problem is that the guard on the exit *didn't* notice. And the passenger who noticed wasn't allowed to go through the door, and let's face it, probably didn't relish taking the (admittedly small) chance that the person in question really was dangerous, so they found a security officer on their side of the exit to report the breach to. Of course, by then it was too late; they had video footage of the person who went through the door, but they couldn't find him. The problem wasn't that the door was unwatched, it's that every once in a while a guard is going to miss someone. They're human, they can't stare at a door for 8 or more hours straight.
Do those figures include the period after Dec. 25th? Cause another explanation could be that people are pussies and didn't fly because "the terrorists are detonating their underpants on planes". People are pathetic and easily scared; the number who opt out on personal dignity grounds is probably far less than those who are literally giving in to terror.
Just for the record, I'm only counting production uses of the code. I do code in my spare time for fun, and my experience is a bit more extensive than that spent in school and work.
That said, I will admit to having *far* less experience in C#/Java compared to C/C++ and Perl. I will admit to some prejudices in favor of C# due to the support it provides for C and C++ style idioms (without the horror that is C++ compiler errors to boot). It was also quite handy for some old number theory projects, since I could implement custom numeric types with relative ease. But having worked with both C# and Java in production environments and studying language features and performance profiles on my own time, my conclusion remains that C# is a much more developer friendly language, with arguable performance benefits (most of them aren't really noticeable in most situations, though I've coded in some environments where Java's autoboxing introduced noticeable perf problems). I'm willing to accept the somewhat reduced portability; I've never coded for anything but Windows (3 years) and Linux (6 years), so for my needs, and those of quite a lot of other people, the reduced portability doesn't really enter in to the equation.
That said, I've never personally been bitten by Microsoft on a legal issue before. If Mono is ever shut down due to legal issues, I'll probably approach Microsoft the same way I approach Sony; with a cross and holy water in hand. But until that day, I'm going to be practical and when I have a choice, choose a language that makes my job easier. The irony is that right now I'm in an all Java shop, so I don't have that choice.
My understanding is that this is partially a function of the size of the battery and the design of the charger. Traditional NiMH batteries (that is, laptop sized and smaller) weren't that good either; according to Wikipedia, even today their cycle durability (number of complete discharge/charge cycles before substantial degradation) is 500-1000, and they shed a lot of power just sitting there (20-30% per month). Li-Ions have a cycle durability of 1000-1200, and their discharge rate is only 5-10%. Li-Polymer batteries have a similar cycle durability to the Li-Ions, and around a 5% discharge rate. The reason NiMH work so well in cars is mostly economy of scale and better charging design that avoids stressing the battery; it's possible those techniques would also work for Lithium types.
In addition, the Li-Ions and especially the Li-Polys have a *much* greater efficiency. NiMH only pulls around 66%, compared to 80-90% for Li-Ion and 99% for Li-Poly. And the energy density of the NiMH is much worse; roughly a half to a third that of the Li-Ion and Li-Poly respectively.
Granted, a Lithium based battery seems to have a lifespan that is roughly half that of an NiMH (despite the better numbers of cycle durability, which is a bit confusing). Even if that is the case though, using Lithium based batteries would mean:
So, for those benefits, might it make sense to accept that you'd need to have the battery replaced roughly twice as often? If you gain an extra five miles per gallon in a hybrid (because the car is lighter and you are wasting less energy to the charge/discharge cycle), or you have a substantially greater range in an electric vehicle, that might be a worthy trade-off. Probably not a good financial trade-off just yet (replacement batteries are still fairly expensive), but economies of scale kick in over time, and in the meantime, you have a car that doesn't need refueling as often. People put too low a price on their time and energy; I consider trade-offs like this to be worthwhile in many cases (obviously, you'd need more exact numbers than I'm giving to make a real decision, but it's not as cut and dry as you think).
Wow. Imagine, an open source project cloning the functionality of a commercial product that doesn't support the latest features of the commercial product. Given what I know of MS release cycles, it took them at least a year and half to write the 3.5 release, and they frequently were writing wrappers around functionality that either already existed, or was handed to them by the OS team. It's been two years since 3.5 released, and the Mono team has got a lot of the functionality exposed, without the advantage of OS support for many of the features. Give them a break, Rome wasn't built in a day either you know.
Right, because Microsoft is making a profit off licensing the .NET framework. Wait, you mean they don't charge a cent for it? And C# is a better language than Java, with the Mono project providing cross-platform compatibility, so Windows users have an easier time migrating to Linux if they so choose? Clearly I should listen to random /.er and forswear all use of anything that "supports" "Microsoft products" in any way, including the OpenOffice; after all, it lets people read and write Office documents, and by doing so, indirectly enables the Microsoft hegemony.
P.S. Yes, C# being better than Java is personal opinion. I've used both, Java for two years in school and one and a half years in the workforce, C# for a little under a year in school and half a year in the workforce (plus a few years of various other languages, mostly C/C++ and, yes, Perl). For developers, the lack of rigid ideological adherence to OO dogma is quite helpful; delegates for callbacks and "pass-by-reference" for arguments instead of inane wrapper classes for both (yes, pedantic types, I know it's all pass by reference, but you know what I mean), not needing to think about auto-boxing as much (since .NET collections of primitives really are primitives, not boxed primitives), operator overloading and structs to enable the creation of relatively efficient and easy to use numeric types, etc. I think both languages have merit, and I think both languages are improved by the competition (e.g. without C#, I'm not sure Java would ever have introduced generics, since it violated the spirit of OO). But I'm not going to reject C# just because MS made it.
That's because a government pension subtracts from social security income. And that is because government employees under the pension system were exempt from social security taxes at the time (the government pension system was supposed to replace it for them). She's effectively combined her social security benefits combined with her workplace pension; the source is the government in any event. Your post is disingenuous to say the least.
Democrat is to Republican as:
C) Pot is to Kettle
A clearly inaccurate analogy. Everyone knows there are no black Republicans.
...
Okay, Steele, but I'm not sure he's Republican so much as he is plain nuts. *dives out window to avoid getting served with libel papers*
"...a trademark application for the name was not filed [by the plaintiff] until May - when rumours about Microsoft's new product had already spread widely across the internet."
"Microsoft, meanwhile, filed its own trademark applications for the name in March - for a variety of uses, including search engine software, interface software, advertising, telecoms and for 'providing a website and website links to geographic information, map images and trip routing'."
Says it all really. This company didn't even bother trying to establish trademark rights until two months after Microsoft, after news of the new engine had leaked. This screams trademark troll.
You haven't addressed the secondary issue; that the melting in the Himalayas is only doubled by the soot, not caused by it. And the scrubbers would have little to no effect on glacier melt in the rest of the world. And that "destroying the world economy" is a politically motivated, short sighted conclusion. Most of the reasonable forecasts show it "dragging" the economy down by about 1-3% of the "GWP" (Gross World Product). The economic doomsday types like to discount the possibility that the cost of oil will increase much beyond the rate of inflation, as if the entire world can start living like Americans (or even Western Europeans) without drastically increasing the price of oil.
I believe they're supposed to act as a buffer. They accumulate water during the wet season, and release it during the dry, in roughly equal amounts. If they melt faster than they accrete, then you get more water during the dry season for a while (while the glaciers are close to their original size), then it starts to taper off as the increased melting is offset by the lesser amount of ice. Eventually, the glaciers are reduced to virtually nil, and you get little or nothing after that.
Yes, but in a summary, leaving off the "hundreds of" is fine when you're trying to express the concept of "a lot" in rough terms. I'm fairly sure (non-retarded) people can figure out that there are more than two million people in India.
Feeding a lame troll, but the source of soot is the same source as the CO2. So we're still solving the same problem. And they've already noted that the melting in the Himalayas is abnormally fast, but that doesn't change the fact that all the glaciers are melting, if "only" half as fast as the Himalayas.
Hundreds of millions is millions you know. Don't be such a pedant.
To their competitors, yes (though in recent years they've bought them out instead of suing them in most cases). But do you have any reason to believe they are going to violate their customers' privacy (at least, any more so than Google)?
There are different ways of triggering a buffer overflow than a simple unbounded copy. While most of them end up as a buffer overflow, they are caused by much more subtle problems, e.g. integer overflow, which can lead to the bounds you set up being incorrect in the first place.
Beyond that, the vast majority of the buffer overflows that have occurred recently aren't in string processing code; they're network facing operations that operate on raw byte buffers. And they've always had bounds checking, but problems like integer overflow occasionally crop up; automated checks can't catch them with 100% accuracy. It's up to security review teams to do fuzzing and code reviews looking for that sort of problem.
How do you know Google counts as "good hands"?
And by banned, he doesn't just mean there is a policy against them. They run regular code scans for "illegal" functions and then send out high priority bug reports to the code owners if any are found. I had to fix a couple of them when I worked there (and they weren't for strcpy, it was more subtly problematic functions). In my entire time at MS, I never saw one instance of strcpy. Usually the code used StringCchLength and StringCchCopy, which are not only safe if called with the appropriate buffer size, but function well with both ASCII and UTF-16 strings environments (though in practice, all our code was compiled with unicode support).
I'm don't think this would trip double jeopardy even in the States. If a verdict is based upon misinterpretation of the law, and the appeals chain decides this is the case, then a retrial is feasible. The only way you hit double jeopardy is if the Supreme Court says you do (unless you decide not to appeal the circuit court ruling, though I can't imagine why you wouldn't), and it has to be due to a point of law that went awry in the original trial. Prosecutors can't just re-file charges nilly-willy. In the States, I know re-trials are permissible due to jury contamination and the like, this is just a different reason for a re-trial (and in both cases, it's the judges, not the prosecutors who make the decision that a re-trial is warranted).
Riiiight... I'm sure people *love* blowing millions (or billions) of dollars to hand out free electronics to all and sundry. Tell me, what else is different in your world?
That CD isn't actually necessary for most systems (my machine at home is on Verizon DSL, but I've got no Verizon software installed by choice, same went for the Comcast service I used to have, I just configured it on a throwaway OS so the software wouldn't clutter my machine). And like I mentioned, isn't required for global network QoS anyway. Yes, having the CD enable local QoS is a good thing in any event, but that's for the benefit of that user, it doesn't affect the ISP or its other customers in any way.
Why waste the power? A personal use DNS server is a waste; if your ISPs DNS is slow there are always alternatives (I used Verizon's DNS for years when living in an area where Comcast DNS performance was slow). I know DIY is fun, has geek cread and all that, but your local machine will cache frequently accessed sites anyway, and the benefit gained on uncached sites will be seen so infrequently that you're not benefiting.
Correction, make that *above*.
Are you advocating a system where the ISP has mandate power over the OS configuration?
A short version of my longer answer below: They already have this power. They don't need to control your OS, because they control your modem, and your modem will obey QoS from their network, even if your OS won't obey it locally. All you'll do is overload the CPU on both OS and modem dropping and resending packets.
The thing is, a larger network has already set a cap for you, the speed you are paying for. Your modem supports it, it doesn't matter how your computer is configured. Your connection and their hardware could support a much higher speed most of the time, but they limit it to what they are capable of providing reliably and inexpensively.
With your roommate, you could simulate a similar scenario by configuring your router to limit the bandwidth of each machine to, say, (connection speed / # of machine) * 1.2. Note that each user is capable of using a little more than his fair share, but he can't completely dominate the network. Some capacity is wasted most of the time (since most people won't use their share), and the users who really want bandwidth get a little extra. If everyone tries to max it out at once, then the top download speed drops by a bit (that final 0.2), but everyone gets a decent experience because the router was enforcing the policy.
Basically what I'm saying is that your situation was caused by poor local network management. The ISPs are perfectly capable of enforcing good global network management on their users (local management doesn't matter, attempts to hog bandwidth by local sources are filtered by your modem, which they do control), so your problem is inapplicable.