Personally I can think of few things more horrific than reviewing the code of a mathematician... I speak from experience.
Few people are as skilled at writing obfuscated, convoluted code with meaningless identifiers and believe it's so "obvious" it doesn't need to be commented much.
Of course there are branches of CS where maths is absolutely required. However, his experience (and mine), is that there are so many parts of CS where you don't need much maths that maths is hardly a fundamental requirement for someone to do useful work in computer science. As such, it's worth questioning why so many CS programs are heavy on required maths courses.
Any competent programmer will know how to do the above, but many will do it because they've either learned the basic truth tables and know how to apply them, or because they've learned the most common transformations by rote. They are "using maths" the same way a builder is "using physics" when putting in an RSJ to prop up a wall, or the way we're "using physics" when we depend on objects not suddenly floating into outer space.
Whether that is a sensible way to look at things or not really depends on your viewpoint. I'd argue it's pointless.
That you can explain something using maths doesn't mean that everyone thinks about maths or "use maths" in any conscious way when they do that something.
I could do the transformation in your example before I'd ever heard of boolean algebra, and learned to spot it without having to think much more about it after having thought through it step by step a few times. My guess would be I figured it out at 7-8 years old based on what I remember of the complexity of my programming back then. I'd argue that I was/am not "using" DeMorgan's law, but just learned a pattern by rote that I understood due to language, not maths.
If you still insist on calling it maths, then fine. But then the logical extension is to conclude that people complaining there is "too much maths" in CS are highly unlikely to be complaining about basic stuff like that, which people can/will figure out without any background in maths as/when neeeded.
I "forget" about it on a regular basis. Almost none of the software engineering work I've done over the last 12 years, or the hobby programming I did for another 15 before that have required much maths beyond basic boolean algebra, some understanding of computational complexity, and assorted other stuff that's mostly been at most at high school level.
I started programming at 5 - boolean algebra was the first maths I learned, because it flowed naturally from learning programming, though it took a few years before I knew it had a name. But really, boolean algebra is just logic with symbols.
You mention programmers/software engineers and computer scientists spearately, and you're right to. The two have about as much in common as a builder and an architect - they'll share some vocabulary and some understanding of methods, but what they need to do their jobs are vastly different.
I enjoy reading CS research papers, and I have an interest in some subsets of CS - particularly compiler design - but I don't particularly enjoy maths, and tend to avoid maths heavy papers simply because my interest in CS is a hobby and maths heavy papers take more effort (and in compiler design you need very little maths apart from some very basic graph theory anyway - when people write maths heavy papers on compiler design, then to me it tends to be a sign they don't understand what they are writing about well enough to explain it plainly - so far I've seen very few exceptions to that).
But ultimately CS isn't my career - software engineering IS. The two are different fields, and it's time people actually realize that... More importantly, it's time more schools realize that, and start offering differentiated computer science and software engineering degrees.
Someone with an MSc or even PhD in Computer Science can easily be useless as software engineers. You wouldn't expect an architect to be able to step right into the job of a builder, after all, and you'd be skeptical about the choices of someone who picked an education as an architect if they wanted to become a builder. I've had to deal with my share of highly educated "software engineers", and frankly none of the best software engineers who have worked for me have had anything above a BSc in CS, and many of them had no degree or unrelated degrees that gave them a good appreciation of the specific domain they developer software for, whereas very few of the people I've hired with MSc's and PhD's in CS have done particularly well (there are the odd exception) - it's marked enough that I've gotten to the point that a MSc or PhD in CS is a warning sign that cause me to probe actual engineering skills a lot more thoroughly, as well as asking some pointed questions about what drove them to pursue their degrees and why they subsequently went into software engineering.
But even in CS, the extent of maths you need depends massively on what your focus is. As I mentioned, compiler design rarely need to use much maths (some people do, but not because it's necessary - people like different tools), and a lot of other areas use only some small subset or other of maths.
I hardly took any maths at university, and it's rare for me to come across CS papers even outside of compiler/programming language design that I'd have any problems following due to the maths content. What maths content there tends to be is most often limited enough for context alone to be sufficient to get most of it. When I do run into problems, I can usually easily find papers that have no problems expressing the same information without much maths, which signals that it's very much a communications issue rather than something inherent to the problem. The cases where the maths is so integral to the message that it actually makes much difference apart from reducing the potential audience is very limited.
Unnecessary use of maths in CS papers is one of my pet peeves. I'm not advocating "dumbing down" research, but scientists that use "big words" when there is no reas
Often you don't need the control, because you can set things up to just reschedule a task if the hardware it ran on failed. That's where clusters are worthwhile. For batch jobs where the management nodes in the cluster can deem a specific work package as either failed or completed, it simply isn't a big problem if machines fails regularly. You set things up to let you quickly install a fresh image and put in a spare, and use part of the savings over buying a mainframe to hire someone to go around replacing machines.
Where mainframes are worthwhile is where there is lots of IO between CPU's and failure detection needs to be fine grained, e.g. typically transaction oriented systems.
Clusters and mainframes are really largely suitable for completely different workloads. Where PC technology is trying to eat into mainframe territory is with large servers. There I'm much more with you - if you start buying some ridiculously souped up x86 based server where you end up trying to recreate mainframe level redundancy by slotting in dual backplanes w/chipkill support for the RAM, multiple power supplies, dual disk controllers and network cards etc., why not just buy a mainframe? Especially when mainframes these days support LPARs with Linux etc.
One of the things about the "Beowulf cluster" of commodity boxes is that they are cheap, giving some aspects of high-end computing power without the cost. This is for those garage-based start-ups that need some serious power but if a hard-disk drops out or a LAN connection goes dead it's not a huge deal.
Your mainframe setup is for large scale businesses, universities, nuclear research, all that fun kind of stuff. If your job is riding on it, get a mainframe.
That's a bogus distinction. Beowulf clusters and similar is for ANY kind of application where the stability of an individual node does not cause problems for the overall utility of the system. That INCLUDES things like nuclear research, universities and needs of lots of very serious large scale businesses. It's a design trade-off, not an indicator of how "serious" your use is. Essentially you work around the stability of individual nodes by software because the aggregate capacity available is more important than whether a specific unit of work needs to be scheduled more than once due to a hardware failure.
Mainframes are used where it does not pay to try to work around hardware weaknesses, and where IO bandwidth tends to be more important than CPU power - typically applications where integrity is critical and the complexity in ensuring that integrity on a cluster is simply too high.
Put options are the "safe" way of doing this, but realizing a return is hard, since the buying the put options will be more expensive the higher the value and the longer the option is valid, requiring a larger price drop to make money of it. A put option essentially gives you the right but not obligation to sell shares at a certain price at a certain time interval in the future, regardless of the market price at that point. The potential loss for put options is the cost of buying the shares now plus the cost of the option.
If you "know" that the share will lose value, the better (but riskier) alterative is to sell short. It requires you to find a broker that is able to lend you shares in the security in question, which you then sell. When (or if) the shares drop in price, you buy back a sufficient number to cover the amount you borrowed. The problem with shorting, particularly if the shares aren't highly liquid, is that the potential loss is unlimited (you lose the equivalent of any gain in value from you short the shares until you are able to buy them back). Experienced investors will therefore sometimes use call options as a protection. A call option give you the right to buy shares at a certain price at a certain time interval in the future - in other words the reverse of a put. The downside is of course that this protection will eat up a lot of your potential return. Because of the high risk, short selling is highly regulated.
Your better bet, literally, is to find a bookmaker that will take a bet on it, assuming you can find someone who'l give you good odds, and it's legal where you are. UK bookmakers tend to take bets on almost anything they believe they can reasonably calculate the risk of, or where they can pit their customers against eachother and only pocket the spread.
Title 17, circular 92, chapter, 506 of the U.S. Code says this:
"Criminal offenses
(a) Criminal Infringement. - Any person who infringes a copyright willfully either -
(1) for purposes of commercial advantage or private financial gain, or
(2) by the reproduction or distribution, including by electronic means, during any 180-day period, of 1 or more copies or phonorecords of 1 or more copyrighted works, which have a total retail value of more than $1,000,
shall be punished as provided under section 2319 of title 18, United States Code. For purposes of this subsection, evidence of reproduction or distribution of a copyrighted work, by itself, shall not be sufficient to establish willful infringement."
(The remaining subpart of this paragraph covers criminal offenses in copyright law other than copyright infringement)
So copyright infringement in the US is a criminal offense if its willful AND is either done for profit or if the retail value of all the works are above $1,000. The reason the RIAA sues in civil court is because they a) know that the damages they claim are punitive enough, and the burden of evidence in a civil claim are lower (balance of probabilities vs. reasonable doubt), b) know that they'd have a lot harder time convincing a DA that their evidence would stand up in a criminal trial or be worth the resources to pursue. It's simly a far more cost-effective way of achieving their objectives.
I think I'd gone for a metallic voice screaming "Help me! I'm being stolen! I'm scared he'll hurt me.... Stop the thief!" or something like that rather than the siren...
I like the Norwegian system. In all Norwegian trials apart from cases before the Supreme Court, there is a mix of professional, legally trained, judges and persons drawn from the jury pool (there's no basis for lawyers to reject jury members, as far as I know). Depending on the type of trial, the people drawn from the jury pool either serves as jury members (in serious criminal cases where the sentence can be more than 6 years, there is a jury of 10 to decide guilt) or judges.
When they serve as judges, it will typically be two persons from the jury pool and one professional judge, or four from the jury pool and three professional judges. In those cases, they have the same responsibilities as the professional judges, and they all deliberate together, and all have one vote. The professional judges ensure that the deliberations are focused on deciding the matter according to the law. In some circumstances the professional judges can set aside a decision by the judges drawn from the jury pool - typically that will lead to a retrial, and will require a unanimous decision by the professional judges to happen.
Actually it's already clear that the case WILL go to trial. Reiser was charged a while ago, and asserted his right to a speedy trial, which apparently surprised the prosecutors (it's apparently not common in murder cases).
I agree with you that unless there's something huge that's just not known it seems like there would be very likely a jury would think there's reasonable doubt, though.
RTFA. It's not Columbia that's complaining. It's music store owners. And besides, He isn't giving away anything - you can bet the Mail on Sunday has paid fairly well for the chance.
" ABSTRACT. This paper argues that at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a "posthuman" stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor-simulations is false, unless we are currently living in a simulation. A number of other consequences of this result are also discussed."
Next to noone uses the MM/DD/YY. Most countries (including most European countries) use DD/MM/YY, and even YYYY/MM/DD is more common than the "American" format.
Try going to a better quality US restaurant next time. Americans are as diverse as everyone else, and the "obnoxious US tourist" stereotype has developed because it's the loud, brash, annoying ones we notice. I've been to the US about ten times over the last 18 months, and eaten out several days a week while there, and frankly I've yet to come across anyone being "loud at mealtimes". But then I don't eat at the kind of crappy places where uncivilized annoying people are likely to gather.
Not that I don't have lots of things against the US government, but I know enough Americans to not blame all of them for that.
People are free to choose what to believe, but that's no excuse for teaching superstition in science classes.
There's a near infinite supply of "alternative" theories from crackpots. You can't teach them all, or even a small fraction. What makes creationism worth mentioning?
Unless someone has presented a testable hypothesis there's simply no reason or excuse for presenting it in science classes other than as part of a discussion on how to spot why the theories are not scientific.
Funny. You do realize that they can't push through the new treaty without the agreement of the member states government, don't you?
Furthermore, that one of the real points of contention is that the UK is trying it's best to prevent the treaty from making a charter of fundamental rights for EU's citizens legally binding.
So for once, rather than complaining about the EU in general and Germany in particular, those of us living in the UK should instead be complaining about how our government at every turn tries to prevent from being bound to give it's citizens any form of protection against it's government.
Dragging out Somalia is just plain intellectually dishonest. Somalia haven't had a functioning central government for years. To try to imply that any of their problems are even remotely related to whether or not they check IDs is just disgusting.
There's also a huge difference between ID being checked and someone being arrested and left in jail because some moron police officer refuse to accept what all the checks he did came up clean.
I don't see the poster you replied to complain about being asked to identify himself, but about being arrested for breaking no laws and then have to worry about that arrest being used against him later.
I have about 400 legally bought DVD's. If they prevent me from playing them from my new file server rather than from the original disk, I will stop buying more and instead download ripped movies, as they'd have taken away an essential feature.
I doubt I'm alone - people who buy large volumes of legal DVD's are the ones who'll be the most affected by this. Actual pirates will easily work around it.
For starters, New York isn't all Manhattan. Secondly, cost isn't everything. Transport costs are artificially depressed, particularly in the US where fuel taxes are so low, by not taking into account the cost of the environmental effects. Yes, it would likely still be a trade-off: Do you zone and/or subsidize farming in urban areas to make it viable economically, because of the potential to reduce emissions, or do you accept the emissions or deal with them in other ways. But that's what politics is for: Making trade-offs between competing interests.
How is that a "better" idea? How do you propose to gather and distribute the produce from thousands of roofs? How do you propose to have farmers tend to "land" that is spread out like that? Move machinery and nutrients? Recycle?
Thanks for playing. Meanwhile in the real world a large part of the web content normal users want to access are in formats you consider useless and "plain shit" because people who actually need to get things done and turn a profit doing it have found them to work well and provide significant benefits.
Slashdot Mirror
Few people are as skilled at writing obfuscated, convoluted code with meaningless identifiers and believe it's so "obvious" it doesn't need to be commented much.
Of course it doesn't apply to all of them...
Of course there are branches of CS where maths is absolutely required. However, his experience (and mine), is that there are so many parts of CS where you don't need much maths that maths is hardly a fundamental requirement for someone to do useful work in computer science. As such, it's worth questioning why so many CS programs are heavy on required maths courses.
Whether that is a sensible way to look at things or not really depends on your viewpoint. I'd argue it's pointless.
That you can explain something using maths doesn't mean that everyone thinks about maths or "use maths" in any conscious way when they do that something.
I could do the transformation in your example before I'd ever heard of boolean algebra, and learned to spot it without having to think much more about it after having thought through it step by step a few times. My guess would be I figured it out at 7-8 years old based on what I remember of the complexity of my programming back then. I'd argue that I was/am not "using" DeMorgan's law, but just learned a pattern by rote that I understood due to language, not maths.
If you still insist on calling it maths, then fine. But then the logical extension is to conclude that people complaining there is "too much maths" in CS are highly unlikely to be complaining about basic stuff like that, which people can/will figure out without any background in maths as/when neeeded.
I started programming at 5 - boolean algebra was the first maths I learned, because it flowed naturally from learning programming, though it took a few years before I knew it had a name. But really, boolean algebra is just logic with symbols.
You mention programmers/software engineers and computer scientists spearately, and you're right to. The two have about as much in common as a builder and an architect - they'll share some vocabulary and some understanding of methods, but what they need to do their jobs are vastly different.
I enjoy reading CS research papers, and I have an interest in some subsets of CS - particularly compiler design - but I don't particularly enjoy maths, and tend to avoid maths heavy papers simply because my interest in CS is a hobby and maths heavy papers take more effort (and in compiler design you need very little maths apart from some very basic graph theory anyway - when people write maths heavy papers on compiler design, then to me it tends to be a sign they don't understand what they are writing about well enough to explain it plainly - so far I've seen very few exceptions to that).
But ultimately CS isn't my career - software engineering IS. The two are different fields, and it's time people actually realize that... More importantly, it's time more schools realize that, and start offering differentiated computer science and software engineering degrees.
Someone with an MSc or even PhD in Computer Science can easily be useless as software engineers. You wouldn't expect an architect to be able to step right into the job of a builder, after all, and you'd be skeptical about the choices of someone who picked an education as an architect if they wanted to become a builder. I've had to deal with my share of highly educated "software engineers", and frankly none of the best software engineers who have worked for me have had anything above a BSc in CS, and many of them had no degree or unrelated degrees that gave them a good appreciation of the specific domain they developer software for, whereas very few of the people I've hired with MSc's and PhD's in CS have done particularly well (there are the odd exception) - it's marked enough that I've gotten to the point that a MSc or PhD in CS is a warning sign that cause me to probe actual engineering skills a lot more thoroughly, as well as asking some pointed questions about what drove them to pursue their degrees and why they subsequently went into software engineering.
But even in CS, the extent of maths you need depends massively on what your focus is. As I mentioned, compiler design rarely need to use much maths (some people do, but not because it's necessary - people like different tools), and a lot of other areas use only some small subset or other of maths.
I hardly took any maths at university, and it's rare for me to come across CS papers even outside of compiler/programming language design that I'd have any problems following due to the maths content. What maths content there tends to be is most often limited enough for context alone to be sufficient to get most of it. When I do run into problems, I can usually easily find papers that have no problems expressing the same information without much maths, which signals that it's very much a communications issue rather than something inherent to the problem. The cases where the maths is so integral to the message that it actually makes much difference apart from reducing the potential audience is very limited.
Unnecessary use of maths in CS papers is one of my pet peeves. I'm not advocating "dumbing down" research, but scientists that use "big words" when there is no reas
Do you actually totally fail to understand the point, or are you just deliberately being a pedantic twat?
Where mainframes are worthwhile is where there is lots of IO between CPU's and failure detection needs to be fine grained, e.g. typically transaction oriented systems.
Clusters and mainframes are really largely suitable for completely different workloads. Where PC technology is trying to eat into mainframe territory is with large servers. There I'm much more with you - if you start buying some ridiculously souped up x86 based server where you end up trying to recreate mainframe level redundancy by slotting in dual backplanes w/chipkill support for the RAM, multiple power supplies, dual disk controllers and network cards etc., why not just buy a mainframe? Especially when mainframes these days support LPARs with Linux etc.
One of the things about the "Beowulf cluster" of commodity boxes is that they are cheap, giving some aspects of high-end computing power without the cost. This is for those garage-based start-ups that need some serious power but if a hard-disk drops out or a LAN connection goes dead it's not a huge deal.
Your mainframe setup is for large scale businesses, universities, nuclear research, all that fun kind of stuff. If your job is riding on it, get a mainframe.
That's a bogus distinction. Beowulf clusters and similar is for ANY kind of application where the stability of an individual node does not cause problems for the overall utility of the system. That INCLUDES things like nuclear research, universities and needs of lots of very serious large scale businesses. It's a design trade-off, not an indicator of how "serious" your use is. Essentially you work around the stability of individual nodes by software because the aggregate capacity available is more important than whether a specific unit of work needs to be scheduled more than once due to a hardware failure.
Mainframes are used where it does not pay to try to work around hardware weaknesses, and where IO bandwidth tends to be more important than CPU power - typically applications where integrity is critical and the complexity in ensuring that integrity on a cluster is simply too high.
If you "know" that the share will lose value, the better (but riskier) alterative is to sell short. It requires you to find a broker that is able to lend you shares in the security in question, which you then sell. When (or if) the shares drop in price, you buy back a sufficient number to cover the amount you borrowed. The problem with shorting, particularly if the shares aren't highly liquid, is that the potential loss is unlimited (you lose the equivalent of any gain in value from you short the shares until you are able to buy them back). Experienced investors will therefore sometimes use call options as a protection. A call option give you the right to buy shares at a certain price at a certain time interval in the future - in other words the reverse of a put. The downside is of course that this protection will eat up a lot of your potential return. Because of the high risk, short selling is highly regulated.
Your better bet, literally, is to find a bookmaker that will take a bet on it, assuming you can find someone who'l give you good odds, and it's legal where you are. UK bookmakers tend to take bets on almost anything they believe they can reasonably calculate the risk of, or where they can pit their customers against eachother and only pocket the spread.
Title 17, circular 92, chapter, 506 of the U.S. Code says this: "Criminal offenses
(a) Criminal Infringement. - Any person who infringes a copyright willfully either -
(1) for purposes of commercial advantage or private financial gain, or
(2) by the reproduction or distribution, including by electronic means, during any 180-day period, of 1 or more copies or phonorecords of 1 or more copyrighted works, which have a total retail value of more than $1,000,
shall be punished as provided under section 2319 of title 18, United States Code. For purposes of this subsection, evidence of reproduction or distribution of a copyrighted work, by itself, shall not be sufficient to establish willful infringement."
(The remaining subpart of this paragraph covers criminal offenses in copyright law other than copyright infringement)
So copyright infringement in the US is a criminal offense if its willful AND is either done for profit or if the retail value of all the works are above $1,000. The reason the RIAA sues in civil court is because they a) know that the damages they claim are punitive enough, and the burden of evidence in a civil claim are lower (balance of probabilities vs. reasonable doubt), b) know that they'd have a lot harder time convincing a DA that their evidence would stand up in a criminal trial or be worth the resources to pursue. It's simly a far more cost-effective way of achieving their objectives.
I think I'd gone for a metallic voice screaming "Help me! I'm being stolen! I'm scared he'll hurt me.... Stop the thief!" or something like that rather than the siren...
Why is it you think she can't surface? Nothing would prevent her from surfacing in Russia if she's alive.
When they serve as judges, it will typically be two persons from the jury pool and one professional judge, or four from the jury pool and three professional judges. In those cases, they have the same responsibilities as the professional judges, and they all deliberate together, and all have one vote. The professional judges ensure that the deliberations are focused on deciding the matter according to the law. In some circumstances the professional judges can set aside a decision by the judges drawn from the jury pool - typically that will lead to a retrial, and will require a unanimous decision by the professional judges to happen.
I agree with you that unless there's something huge that's just not known it seems like there would be very likely a jury would think there's reasonable doubt, though.
RTFA. It's not Columbia that's complaining. It's music store owners. And besides, He isn't giving away anything - you can bet the Mail on Sunday has paid fairly well for the chance.
" ABSTRACT. This paper argues that at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a "posthuman" stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor-simulations is false, unless we are currently living in a simulation. A number of other consequences of this result are also discussed."
Next to noone uses the MM/DD/YY. Most countries (including most European countries) use DD/MM/YY, and even YYYY/MM/DD is more common than the "American" format.
Not that I don't have lots of things against the US government, but I know enough Americans to not blame all of them for that.
There's a near infinite supply of "alternative" theories from crackpots. You can't teach them all, or even a small fraction. What makes creationism worth mentioning?
Unless someone has presented a testable hypothesis there's simply no reason or excuse for presenting it in science classes other than as part of a discussion on how to spot why the theories are not scientific.
Furthermore, that one of the real points of contention is that the UK is trying it's best to prevent the treaty from making a charter of fundamental rights for EU's citizens legally binding.
So for once, rather than complaining about the EU in general and Germany in particular, those of us living in the UK should instead be complaining about how our government at every turn tries to prevent from being bound to give it's citizens any form of protection against it's government.
There's also a huge difference between ID being checked and someone being arrested and left in jail because some moron police officer refuse to accept what all the checks he did came up clean.
I don't see the poster you replied to complain about being asked to identify himself, but about being arrested for breaking no laws and then have to worry about that arrest being used against him later.
I doubt I'm alone - people who buy large volumes of legal DVD's are the ones who'll be the most affected by this. Actual pirates will easily work around it.
For starters, New York isn't all Manhattan. Secondly, cost isn't everything. Transport costs are artificially depressed, particularly in the US where fuel taxes are so low, by not taking into account the cost of the environmental effects. Yes, it would likely still be a trade-off: Do you zone and/or subsidize farming in urban areas to make it viable economically, because of the potential to reduce emissions, or do you accept the emissions or deal with them in other ways. But that's what politics is for: Making trade-offs between competing interests.
How is that a "better" idea? How do you propose to gather and distribute the produce from thousands of roofs? How do you propose to have farmers tend to "land" that is spread out like that? Move machinery and nutrients? Recycle?
Thanks for playing. Meanwhile in the real world a large part of the web content normal users want to access are in formats you consider useless and "plain shit" because people who actually need to get things done and turn a profit doing it have found them to work well and provide significant benefits.
RTFA. This is _not_ about sending fake invoices, but about sending spam e-mails with malicious apps masquerading as attached invoices.