Granted, there's some text formatting, but does every character really needs a separate tag around it?
I believe the aim is to precisely mark up the semantic meaning of each of the symbols. For example, the ac from your formula is described as being two separate identifiers that are (implicitly) multiplied together (<mi>a</mi><mi>c</mi>), rather than a single multi-character identifier (<mi>ac</mi>), which addresses what is otherwise one of the major problems with more complex formulæ; ambiguity over what exactly the variables are.
But did it need to be so verbose? Definitely not. Some sensible defaults would have made life ever so much easier for the document author, and that would have hastened adoption. Ways to take the MathML and turn it into an executable entity would also be useful, but they might exist and I might be just ignorant.
As usual, you've got to find someone who develops for Windows and is sufficiently interested to work on the bug. As it is a rendering problem, working on another platform and cross-compiling won't work, and the Windows API is sufficiently different to make it much easier to be a specialist rather than a cross-platform guy. I'd guess that if someone were willing to commit some money (some sort of targeted bug bounty) to pay for the fix, it would get done sooner.
It's not magic. (Or rather it is, but we're all the magicians.)
To do the grunt work on the projects the scientist they're working for wants to do.
Replicating others' work is usually the introduction to that. After all, if you can have your minion show that a rival falsified their results, that'll be one less group competing with you for the money in the next call for grant proposals.
Which is exactly the problem. Nobody's doing follow-up papers. Follow-up papers aren't sexy, don't get published in top-line journals and don't get a lot of cites. In short, they don't advance your career. So scientitsts don't do them.
Follow-up papers are the usual things that a doctoral student starts out their career writing. Sometimes even masters students (depending on the discipline and the difficulty of conducting the experiments). The grad-school grunts don't have a lot of expectation of being heavily cited, so the risk to them is much lower. Once they've duplicated someone else's results, they can start thinking about what they'd do differently...
We have something like that in the US. It's called Boston.
I live in the UK, and I've visited Boston a few times. You've got no idea what you're talking about. While Boston's got the same sort of layout as most UK (and European) cities due to having the road layout designed in pre-motor-vehicle times (and not being "rationalized") most of its roads are not too narrow. Most of the UK uses roads (especially away from major routes) where the road would be entirely adequate without any parking and provided nothing is wider than an ordinary car. But now add plenty of parking, trucks and buses. And this isn't just part of one city. It's virtually all of them, and out in the 'burbs and the countryside too.
No, seriously, I want to see how well this car performs in a city where the posted 40mph speed limit oin the Staten Island Expressway is ignored by the vast majority of cops and motorists, the normal speed is about 70mph or so, and people will rear end you out of spite if you go too slow for them.
Don't worry. The car can be fitted with a rear-facing camera so that the person who speeds into the rear end can be automatically reported to both the police and their insurance company. What's more, there won't be any recourse on the lines of "oh, that shouldn't have been reported" since there will be video evidence of (at least) gross negligence. The most perfect "fuck you, asshole" ever.
Who knows, it might even start to improve driving habits round NYC...
[The police] spend more resources on crimes that are more damaging than simply having a game account stolen (which sucks, but is hardly life-altering) or crimes they can make money off of (speeding, asset forfeiture, etc.).
To be fair, speeding is a major factor (though clearly not the only one) in increasing the frequency and severity of car crashes, which in turn can definitely have life-altering consequences. (You can't get a much more "life-altering consequence" than being killed, and even with modern car safety features, the best way to not die from a car crash is to not be in a car crash at all.) Speed has a two-fold impact on crash frequency and severity: more kinetic energy, and less time for drivers to react so as to avoid the smash or reduce its consequences.
But dried babies don't make good child soldiers when they get older. The guns for which will come hot off the 3d press:)
Just 3D print the babies pre-dried.
Re:It would be safer if cyclists followed traffic
on
How Safe Is Cycling?
·
· Score: 1
Cyclists rarely hurt anyone, and car drivers kill cyclists every day.
It doesn't give you the right to be a jerk. No matter how much others are law-breaking jerks, you don't get a free pass. Nor do they. Being a jerk is wrong, and especially being a jerk in the public sphere (e.g., on a road or sidewalk) is wrong. When someone else on the road is behaving like a jerk, you must not be a jerk back, because you can bet that it will cause problems for others who aren't being jerks as well, and that escalates the danger levels hugely. And all because you can't keep your temper in check.
If you really want to get back at them, phone the number plates of the most jerkish drivers into the cops and say that they were driving erratically. You might also speculate, but not conclude, that they might be under the influence of intoxicants. After all, drunk drivers are a hazard to lots of people...
The disassembler he used is not. So it is (at least theoretically) possible to see if there is a back door.
You can also try disassembling the code by hand. (It's hard work, but you can do it.) Then you've only got to worry about whether the file has some trickery done to it so it looks like one binary sequence when opened normally, but uses something else when executed. Which I suppose is possible, but it's getting really hard to make such things reliable; it's easier to just put a rootkit on the OS and lie to users that way...
This is an awful bill from the inventor perspective. With the "loser pays" rule, trying to enforce a patent, which costs about $1 million and up, becomes even more expensive. Now, suing a big company means you may have to pay for their lawyers. Patent cases are won by patent holders about 40%to 50% of the time, so you have to risk bankruptcy to enforce a patent.
You should realise that it is currently just as viciously horribly hard to enforce a patent right now against a large firm. The costs involved in bringing a patent suit to court are huge, so much so that it's probably the case that any large firm can poach any patent held by a small firm and get away with it, as the small guy is likely to go bankrupt before the first hearing. It's that which is part of the problem.
The other major part is the way that the obviousness bar is too low. It should be hard to get a patent in the first place, so that it is something that is truly worth defending. By making it genuinely push the state of the art hard, the costs of getting a license become more justifiable as well. (There possibly ought to be a mechanism for mandatory licensing in some cases, perhaps after half the protection period. But that's a more speculative thing.)
Nobody promised that being an inventor was going to be an easy way to earn a living.
Funny, one could easily say the same thing about drivers of cars. I see law breaking multiple times on my commute to work by drivers in cars.
There are always assholes about. Doesn't mean you've got the right to be one or that other people shouldn't get cross with you if you are one. That even applies if you're in control of a vehicle. Any kind of vehicle.
For those who forget, Minneapolis is located at the same lat/long as Moscow, and it regularly gets snowfall of several feet come high winter.
If Moscow is at the same lat/long as Minneapolis, then that Moscow must be in Minnesota. The one on Russia is more than ten degrees of latitude further north, nearly as far north as Fort McMurray, Alberta. (I'd guess that the climate would still be similar though; I don't want to be in any of them in the winter...)
FWIW, Minneapolis is at about the latitude as Turin in Italy. The winter's pretty miserable there too if I remember right, but for different reasons.
Given the speed and travelled distance difference between cars and bicycles maybe per-hour accidents would be a better metric.
Accidents-per-person-journey is probably best (or its reciprocal if you like larger numbers), as that most closely matches the likelihood of someone having an accident, given that the profile of time spent in different vehicles is different. It even handles how to compare with various kinds of mass transit schemes. You can then think in terms of "how likely am I to get seriously injured when going from home to work if I travel by bicycle?" which is actually a useful question (and comparable to "... if I travel by car?" by even not too statistically-sophisticated people).
It may be difficult to tell, but I would ALWAYS choose a platform that had capable independent fans over one backed by an enormous corporation. Single entities abandon things seemingly on whims (OK, well actually, expectations of profit, sometimes by folks who can't predict there'll be wind accompanying a hurricane). But if there's a viable community of folks who aren't just fans, but are capable of providing some kind of momentum and support, then the platform will probably survive until something unequivocally better comes along (at which point you would probably want to switch anyhow).
Another way to look at it is to think in terms of looking at how many small businesses are involved in the community associated with a platform. Big organisations can move between platforms at the whim of a manager, but small businesses tend to be highly committed to keeping what they're using going; the cost of moving to another platform can be prohibitive for them. What's more, they'll usually be able to tell you about tricks you'd have never thought about: small businessmen are fairly inclined to share in areas outside their main business as they're often hard-pressed to take on more work.
Sharing things back with them encourages them even more; they tend to quite quickly get the idea of a gift economy. Or at least that's true in my experience. It's larger companies that you shouldn't trust so much, as they tend to want to grow into new areas and care less about getting anything in return other than money-by-year-end (or quarter-end). Such an approach is fundamentally scummier.
Version 1.0 was called Henry Spencer and was developed in Canada.
That explains much. Except that I think that Henry was developed initially in collaboration with H P Lovecraft. (I've read Henry's code; it scares me.)
Why would the pilot program use the more expensive, harder to implement option of meters if they don't intend it in the real program? Unless there is a really good reason that the pilot needed meters instead of odometers, I can't believe the claim that the meters are "just for the pilot".
One reason might be that they plan to compare the level of deviation between odometers and actual measured distances. There's also the interesting question of what to do about vehicles that regularly leave the state (consider someone who lives and works in Portland, but drives once a week up to Tacoma, WA). Without some baseline data, it becomes hard to figure out how much you're over- or under-charging; one of the key things a pilot program does is help identify that all these awkward cases exist so that everything doesn't fall apart when the full rollout hits reality.
Still seems over-complex to me. A flat charge per year would be simpler to administer, and could be divided into as many categories (e.g., for different weight classes) as required. Don't want to pay so much? Get a lower-rate car. (People who have a hummer they only drive once a year get screwed by this. But that's just wasteful anyway.)
Of course there's a good reason 4 may be no - you may believe you will understand your emotional state beforehand but you will be executing your decisions while *in* that state. Two very different things. Hormones exert influence and you can't gauge for them.
But hormones are a stimulus! A different kind to the electrical stimuli transferred via synapses, to be sure, but nonetheless an input to the function computed by the neuron in question. (More complex is the fact that the history of stimuli alters the function computed and what other nodes in the network are connected to; that's much more significant in trying to understand why brains aren't like normal computers.)
"There is a subtlety here in that computational universality requires that you be able to add new memory to the computer or smart phone when it needs more -- for the moment let's assume that additional memory is at hand." -- This is the escape hatch for the finite state problem. With infinite memory, you can get undecidability from a deterministic system. Without infinite memory, you can't.
It should be noted that just because a system is finite doesn't mean that you can analyse it completely with the resources at your disposal. Completely characterising every possible state that a modern computer (64-bit Intel CPU, 16GB memory, few TB disk) can be in is totally beyond any sort of practicality. But it's finite!
"Finite" doesn't help nearly as much as you might think it does at first glance.
Slashdot Mirror
Granted, there's some text formatting, but does every character really needs a separate tag around it?
I believe the aim is to precisely mark up the semantic meaning of each of the symbols. For example, the ac from your formula is described as being two separate identifiers that are (implicitly) multiplied together (<mi>a</mi><mi>c</mi>), rather than a single multi-character identifier (<mi>ac</mi>), which addresses what is otherwise one of the major problems with more complex formulæ; ambiguity over what exactly the variables are.
But did it need to be so verbose? Definitely not. Some sensible defaults would have made life ever so much easier for the document author, and that would have hastened adoption. Ways to take the MathML and turn it into an executable entity would also be useful, but they might exist and I might be just ignorant.
After six months, zero progress on fixing it.
As usual, you've got to find someone who develops for Windows and is sufficiently interested to work on the bug. As it is a rendering problem, working on another platform and cross-compiling won't work, and the Windows API is sufficiently different to make it much easier to be a specialist rather than a cross-platform guy. I'd guess that if someone were willing to commit some money (some sort of targeted bug bounty) to pay for the fix, it would get done sooner.
It's not magic. (Or rather it is, but we're all the magicians.)
To do the grunt work on the projects the scientist they're working for wants to do.
Replicating others' work is usually the introduction to that. After all, if you can have your minion show that a rival falsified their results, that'll be one less group competing with you for the money in the next call for grant proposals.
Which is exactly the problem. Nobody's doing follow-up papers. Follow-up papers aren't sexy, don't get published in top-line journals and don't get a lot of cites. In short, they don't advance your career. So scientitsts don't do them.
Follow-up papers are the usual things that a doctoral student starts out their career writing. Sometimes even masters students (depending on the discipline and the difficulty of conducting the experiments). The grad-school grunts don't have a lot of expectation of being heavily cited, so the risk to them is much lower. Once they've duplicated someone else's results, they can start thinking about what they'd do differently...
Having lived in East Germany, I can tell you. East Germans didn't pretend they were free.
So you're saying that America is better at propaganda than East Germany was?
We have something like that in the US. It's called Boston.
I live in the UK, and I've visited Boston a few times. You've got no idea what you're talking about. While Boston's got the same sort of layout as most UK (and European) cities due to having the road layout designed in pre-motor-vehicle times (and not being "rationalized") most of its roads are not too narrow. Most of the UK uses roads (especially away from major routes) where the road would be entirely adequate without any parking and provided nothing is wider than an ordinary car. But now add plenty of parking, trucks and buses. And this isn't just part of one city. It's virtually all of them, and out in the 'burbs and the countryside too.
I commute by train; I'm not crazy.
No, seriously, I want to see how well this car performs in a city where the posted 40mph speed limit oin the Staten Island Expressway is ignored by the vast majority of cops and motorists, the normal speed is about 70mph or so, and people will rear end you out of spite if you go too slow for them.
Don't worry. The car can be fitted with a rear-facing camera so that the person who speeds into the rear end can be automatically reported to both the police and their insurance company. What's more, there won't be any recourse on the lines of "oh, that shouldn't have been reported" since there will be video evidence of (at least) gross negligence. The most perfect "fuck you, asshole" ever.
Who knows, it might even start to improve driving habits round NYC...
[The police] spend more resources on crimes that are more damaging than simply having a game account stolen (which sucks, but is hardly life-altering) or crimes they can make money off of (speeding, asset forfeiture, etc.).
To be fair, speeding is a major factor (though clearly not the only one) in increasing the frequency and severity of car crashes, which in turn can definitely have life-altering consequences. (You can't get a much more "life-altering consequence" than being killed, and even with modern car safety features, the best way to not die from a car crash is to not be in a car crash at all.) Speed has a two-fold impact on crash frequency and severity: more kinetic energy, and less time for drivers to react so as to avoid the smash or reduce its consequences.
But dried babies don't make good child soldiers when they get older. The guns for which will come hot off the 3d press :)
Just 3D print the babies pre-dried.
Cyclists rarely hurt anyone, and car drivers kill cyclists every day.
It doesn't give you the right to be a jerk. No matter how much others are law-breaking jerks, you don't get a free pass. Nor do they. Being a jerk is wrong, and especially being a jerk in the public sphere (e.g., on a road or sidewalk) is wrong. When someone else on the road is behaving like a jerk, you must not be a jerk back, because you can bet that it will cause problems for others who aren't being jerks as well, and that escalates the danger levels hugely. And all because you can't keep your temper in check.
If you really want to get back at them, phone the number plates of the most jerkish drivers into the cops and say that they were driving erratically. You might also speculate, but not conclude, that they might be under the influence of intoxicants. After all, drunk drivers are a hazard to lots of people...
The disassembler he used is not. So it is (at least theoretically) possible to see if there is a back door.
You can also try disassembling the code by hand. (It's hard work, but you can do it.) Then you've only got to worry about whether the file has some trickery done to it so it looks like one binary sequence when opened normally, but uses something else when executed. Which I suppose is possible, but it's getting really hard to make such things reliable; it's easier to just put a rootkit on the OS and lie to users that way...
This is an awful bill from the inventor perspective. With the "loser pays" rule, trying to enforce a patent, which costs about $1 million and up, becomes even more expensive. Now, suing a big company means you may have to pay for their lawyers. Patent cases are won by patent holders about 40%to 50% of the time, so you have to risk bankruptcy to enforce a patent.
You should realise that it is currently just as viciously horribly hard to enforce a patent right now against a large firm. The costs involved in bringing a patent suit to court are huge, so much so that it's probably the case that any large firm can poach any patent held by a small firm and get away with it, as the small guy is likely to go bankrupt before the first hearing. It's that which is part of the problem.
The other major part is the way that the obviousness bar is too low. It should be hard to get a patent in the first place, so that it is something that is truly worth defending. By making it genuinely push the state of the art hard, the costs of getting a license become more justifiable as well. (There possibly ought to be a mechanism for mandatory licensing in some cases, perhaps after half the protection period. But that's a more speculative thing.)
Nobody promised that being an inventor was going to be an easy way to earn a living.
(I'll wait patiently for my check now)
Cash it quickly when you get it...
Funny, one could easily say the same thing about drivers of cars. I see law breaking multiple times on my commute to work by drivers in cars.
There are always assholes about. Doesn't mean you've got the right to be one or that other people shouldn't get cross with you if you are one. That even applies if you're in control of a vehicle. Any kind of vehicle.
For those who forget, Minneapolis is located at the same lat/long as Moscow, and it regularly gets snowfall of several feet come high winter.
If Moscow is at the same lat/long as Minneapolis, then that Moscow must be in Minnesota. The one on Russia is more than ten degrees of latitude further north, nearly as far north as Fort McMurray, Alberta. (I'd guess that the climate would still be similar though; I don't want to be in any of them in the winter...)
FWIW, Minneapolis is at about the latitude as Turin in Italy. The winter's pretty miserable there too if I remember right, but for different reasons.
Given the speed and travelled distance difference between cars and bicycles maybe per-hour accidents would be a better metric.
Accidents-per-person-journey is probably best (or its reciprocal if you like larger numbers), as that most closely matches the likelihood of someone having an accident, given that the profile of time spent in different vehicles is different. It even handles how to compare with various kinds of mass transit schemes. You can then think in terms of "how likely am I to get seriously injured when going from home to work if I travel by bicycle?" which is actually a useful question (and comparable to "... if I travel by car?" by even not too statistically-sophisticated people).
It may be difficult to tell, but I would ALWAYS choose a platform that had capable independent fans over one backed by an enormous corporation. Single entities abandon things seemingly on whims (OK, well actually, expectations of profit, sometimes by folks who can't predict there'll be wind accompanying a hurricane). But if there's a viable community of folks who aren't just fans, but are capable of providing some kind of momentum and support, then the platform will probably survive until something unequivocally better comes along (at which point you would probably want to switch anyhow).
Another way to look at it is to think in terms of looking at how many small businesses are involved in the community associated with a platform. Big organisations can move between platforms at the whim of a manager, but small businesses tend to be highly committed to keeping what they're using going; the cost of moving to another platform can be prohibitive for them. What's more, they'll usually be able to tell you about tricks you'd have never thought about: small businessmen are fairly inclined to share in areas outside their main business as they're often hard-pressed to take on more work.
Sharing things back with them encourages them even more; they tend to quite quickly get the idea of a gift economy. Or at least that's true in my experience. It's larger companies that you shouldn't trust so much, as they tend to want to grow into new areas and care less about getting anything in return other than money-by-year-end (or quarter-end). Such an approach is fundamentally scummier.
Version 1.0 was called Henry Spencer and was developed in Canada.
That explains much. Except that I think that Henry was developed initially in collaboration with H P Lovecraft. (I've read Henry's code; it scares me.)
Then you have to apply both kinds of cryptography! So whatever you do, always add a layer of ROT13.
I go a bit beyond that and triple-ROT13 the data. Like that, kid sisters that work for the NSA will surely be defeated!
You can't use the budget to take positive action like killing babies
You have a twisted mind.
Why would the pilot program use the more expensive, harder to implement option of meters if they don't intend it in the real program? Unless there is a really good reason that the pilot needed meters instead of odometers, I can't believe the claim that the meters are "just for the pilot".
One reason might be that they plan to compare the level of deviation between odometers and actual measured distances. There's also the interesting question of what to do about vehicles that regularly leave the state (consider someone who lives and works in Portland, but drives once a week up to Tacoma, WA). Without some baseline data, it becomes hard to figure out how much you're over- or under-charging; one of the key things a pilot program does is help identify that all these awkward cases exist so that everything doesn't fall apart when the full rollout hits reality.
Still seems over-complex to me. A flat charge per year would be simpler to administer, and could be divided into as many categories (e.g., for different weight classes) as required. Don't want to pay so much? Get a lower-rate car. (People who have a hummer they only drive once a year get screwed by this. But that's just wasteful anyway.)
Of course there's a good reason 4 may be no - you may believe you will understand your emotional state beforehand but you will be executing your decisions while *in* that state. Two very different things. Hormones exert influence and you can't gauge for them.
But hormones are a stimulus! A different kind to the electrical stimuli transferred via synapses, to be sure, but nonetheless an input to the function computed by the neuron in question. (More complex is the fact that the history of stimuli alters the function computed and what other nodes in the network are connected to; that's much more significant in trying to understand why brains aren't like normal computers.)
"There is a subtlety here in that computational universality requires that you be able to add new memory to the computer or smart phone when it
needs more -- for the moment let's assume that additional memory is at hand." -- This is the escape hatch for the finite state problem. With infinite memory, you can get undecidability from a deterministic system. Without infinite memory, you can't.
It should be noted that just because a system is finite doesn't mean that you can analyse it completely with the resources at your disposal. Completely characterising every possible state that a modern computer (64-bit Intel CPU, 16GB memory, few TB disk) can be in is totally beyond any sort of practicality. But it's finite!
"Finite" doesn't help nearly as much as you might think it does at first glance.
nothing of value was lossed.
But your reputation as an English literate.
You're ascribing value to that?
Not an option on Win8.x tablets, unfortunately
What, all three of them?