I would assume that what he wants is a physics textbook that assumes that you're very fluent in math (not a book about handwaving), thus making the presentation a lot more dense, thus faster...
I'm myself a Ph.D. student in math and I've tried to find such a book myself. It's very booring to read physics books that really do calculations the hard way (e.g. use pages to do something with matrices that directly follows from a theorem concerning linear operators etc.).
Unfortunately, I don't think such books exists...
I think you should get a clue too. This is not a software engineering contest, it's all about solving hard algorithmic problems as fast as possible. I bet that many people can write good maintainable code through lots of experience, but then some people never learn to solve hard computational problems.
As impossible as you might think it is, many people program other stuff than just some enterprise software. In many industrial applications you need to actually do hard computational stuff and it can be extremely hard to actually come up with an algorithm that solves the problem efficiently. You can of course argue that you should be able to find the algorithms in just some book or publication, but many problems aren't initially in such a form that you would immediately recognize it as some known problem. It can actually take very much inginuity to actually figure that the problem your looking at is simply another formulation of a problem you know. While in theory duality is a simple principle, it's not always in practice.
The point with this contest is finding the people that when seeing a problem figures out an efficient algorithm while most people wouldn't even be able to solve the problem in the first place or would come up with some really innefficient algorithm or would spend days finding the solution ultimately stalling a project. Actually being able to find efficient algorithms can help software engineering too by being able to take the simple route in the design. If you have a straight forward easy design, but have to be able to solve a computational problem fast enought to actually be able to implement it, then you have to be able to solve that problem to take that approach. If you're unable to solve it, you might end up doing something a lot more complex. From a software engineering point of view your documentation and code could be the prettiest in the world, but it doesn't help if the underlying algorithmic approach is bad and you've written lots of code to get around it.
Many people go to jail for just accessing systems without permission. This guy actually purposely caused harm... so I really don't see a reason for anyone to complain. Another point that nobody seems to make is that the time the administrators used to fix this was probably not the only time spent. Many managers probably had to spend time working on this, reporting etc.
One problem with this from the OSS perspective seems to be that many important OSS projects e.g. MySQL, Qt etc. are dual licensed. Many of the patents had something to do with storage systems, so they could be helpfull for a database application. Unfortunately they can't use it.
Why go through all this, when you can simply just solve the problem? If you think about it for a few seconds, you come to realize that the solution they want is a simple brute force. Why? Because the digits are completely random and so are the primes too, so there can be no constructive method (with the math we know) to find a faster way to solve it.
If you code a program you'll notice that you'll find a solution almost instantly. The interesting problem is the 2nd one, because it actually requires that you think. But that one is easy too.:)
I don't think this is any different than what most Europeans had to get used to when countries changed their currency to the Euro. To me it took about a year before I could start thinking in Euros. For example if someone told me that xxx cost yyy euros, I had to convert the number to Finnish marks, so that I could tell if it's cheap or not. Now I don't have to do that anymore.
It would be a pain at first, but people do get used to it.
I think the RFID scanner should simply start the alarm if it gets data that it can't read. Garbled data should be a sign of something strange going on. The problem is that this RSA device doesn't remove the signal, it only brakes it.
There's nothing wrong with vi. If someone accidentally removed "lock" from your system and you don't want to logout when you leave, then vi comes to the rescue!
I wonder if they can deliver enough of it for my needs? Wonder when they start to consider my toilet paper consumption as a denial of service attack... as that was clearly not permitted according to the EULA.
If you've come up with a truly asymptotically better algorithm, then why don't you talk with someone at you CS department? I'm sure someone would like to take a quick look, at least it shows that their students actually want to learn and do study/think/research on their own.
Have you analyzed your algorithm yourself? Is it asymptotically equivalent to something like quicksort, but it just has a tighter inner-loop? Does it have a better average running time on random input? Does it work always (or is it flawed/probabilistic)? I think those are the questions you need to have answered.
This seems very promising. The videos from Strang's lectures on linear algebra seemed nice as I've been reading Strang's books my self.
As some people noted, this is in no way revolutionary as practically every university has webpages for each of their courses. Atleast at the university where I'm studying (Helsinki University of Technology) we usually get good lecture notes from the lecturer as a complement to whatever book we're using, but they never get published on the course's webpage. This is unfortunate, as many other people could benefit from these lecture notes if they were put online. There's lots of good quality material in different universities and I'm grateful to the lecturers who put their notes online. When trying to learn material covered on different courses I usually make a little search on the web trying to find lecture notes of the subject.
I hope many more universities make it a policy to publish their material. I think this would also reduce the amounts of rushed or otherwise bad lecture notes, as I don't think anyone will put handwritten scanned lecture notes online, so everyone would have to typeset them on a computer. This is something you have to struggle with in the introductory math courses here at HUT...
Totally agreed. I'm not very keen of doing the logic at the application side as that means lots of I/O overhead when moving more data than needed from the DB server.
Another thing that I'd like to see soon is foreign keys with integrity checking. It's always nice to do some db corruption prevention on the database side. For all larger databases, this is very important. I know that applications that corrupt its database are broken and should be fixed, but if someone hits on a bug that hasn't been caught and breaks the database you definately want the database to prevent it. The other option, restoring it from backups, is not what you want to do.
I'd recommend some caution when you do that. If you've ever had any problems arguing with your girlfriend, then thing about arguing with 2 or 3 of them...
Huh, no firewall that does this? Have you ever tried IPFilter with any of the BSDs or pf for OpenBSD? What you want to do is run the firewall as a bridge, then you just filter on the incoming or outgoing interfaces. (With ipf, you can only filter on incoming data, but you'll find everything you need in each firewall's documentation...)
And to all those people complaining about this being illegal. I can't think of any company thinking of buying a Cisco, but would then switch to a clone because it's possible. And to all those people telling you how to use a ipf/pf/iptables/ipchains/ipwhatever system for testing, it just ain't gonna teach you much about a PIX.
I've been playing around with a few PIX boxes and if I had never used a Cisco product before (i.e. configured routers/switches, router access-lists, generally knowing IOS) I would have been quite clueless with a PIX. So you can teach yourself firewalling concepts with open source firewalls and the theory behind different protocols (which is mandatory knowledge for anyone configuring firewalls...). Also commercial firewalls really don't have any features not present in open source firewalls (except that configuring stuff on open source firewalls generally is more time consuming when some features aren't integrated, like VPNs...). Anyways if someone would throw a PIX in front of you and tell you to configure a firewall, I bet you wouldn't get the job done (in a reasonable time).
Actually, it's not anti-DRM hardware, it's hardware *without* DRM. Thus there's no circumvention device of any kind, there's just nothing to circumvent.
However, this will be illegal if hardware DRM support becomes a mandatory part of PCs...
You don't call the code in a function pointer. The function pointer is just a variable and it only contains the address of the code you want to run. The machine instruction only gets the address of the function pointer, loads the value of the address from memory and jumps to the memory location given by the function pointer's value. The only thing you'll need to do is to read the function pointer's value (no executable memory needed).
Of course you probably want to modify the function pointer's value, so it has to be in writable memory, thus the kernel has really no way to protect it. So it doesn't have to be on the heap, but that doesn't help you much.
Actually KPNQwest told their employees that they should contact the appropriate authorities to get their salaries from what's left of the company. So basically they've kicked out all their employees without having money to pay their salaries. To me this sounds like they don't have much money...
Well the best multiplayer games ID has ever created were QuakeWorld and Doom. Now somehow it seems like every time a game is designed for a multiplayer experience it gets pretty lame, because the atmosphere in the game just isn't right.
Quake III for example was pretty damn boring. Trying to balance weapons etc. and analyze in advance how the game will be played is just stupid. The nice thing with single-player games is that weapons are usually anything, but balanced also levels aren't symmetric or "balanced" in any way and that is a good thing.
The trend has been to balance the gaming environment and design the levels for better teamplay. This is just stupid as the whole idea with teamplay is to get control and kick the other teams ass. For example look at dm3 in quakeworld. It's extremely unbalanced, but still probably the best teamplay map ever created. Also dm4 is probably the best duel map ever created and both are extremely unbalanced. Now many people say that if a map is unbalanced the whole outcome of a game is set by the first spawn. Then how come certain people always win on dm4, the so called "dm4 specialists"? If it were just a matter of luck, then according to classical probability theory they would lose at least sometimes. In short the whole argument of multiplayer "design" is flawed.
The atmosphere in DOOM III seems to be very cool and I'm quite sure that it's going to be the best multiplayer experience Id has created since QuakeWorld, even though it wasn't designed multiplayer:)
Slashdot Mirror
I would assume that what he wants is a physics textbook that assumes that you're very fluent in math (not a book about handwaving), thus making the presentation a lot more dense, thus faster... I'm myself a Ph.D. student in math and I've tried to find such a book myself. It's very booring to read physics books that really do calculations the hard way (e.g. use pages to do something with matrices that directly follows from a theorem concerning linear operators etc.). Unfortunately, I don't think such books exists...
Usually the proof itself contains more information than the actual statement of the theorem, so you usually don't get much by just assuming stuff.
I think you should get a clue too. This is not a software engineering contest, it's all about solving hard algorithmic problems as fast as possible. I bet that many people can write good maintainable code through lots of experience, but then some people never learn to solve hard computational problems.
As impossible as you might think it is, many people program other stuff than just some enterprise software. In many industrial applications you need to actually do hard computational stuff and it can be extremely hard to actually come up with an algorithm that solves the problem efficiently. You can of course argue that you should be able to find the algorithms in just some book or publication, but many problems aren't initially in such a form that you would immediately recognize it as some known problem. It can actually take very much inginuity to actually figure that the problem your looking at is simply another formulation of a problem you know. While in theory duality is a simple principle, it's not always in practice.
The point with this contest is finding the people that when seeing a problem figures out an efficient algorithm while most people wouldn't even be able to solve the problem in the first place or would come up with some really innefficient algorithm or would spend days finding the solution ultimately stalling a project. Actually being able to find efficient algorithms can help software engineering too by being able to take the simple route in the design. If you have a straight forward easy design, but have to be able to solve a computational problem fast enought to actually be able to implement it, then you have to be able to solve that problem to take that approach. If you're unable to solve it, you might end up doing something a lot more complex. From a software engineering point of view your documentation and code could be the prettiest in the world, but it doesn't help if the underlying algorithmic approach is bad and you've written lots of code to get around it.
Many people go to jail for just accessing systems without permission. This guy actually purposely caused harm... so I really don't see a reason for anyone to complain. Another point that nobody seems to make is that the time the administrators used to fix this was probably not the only time spent. Many managers probably had to spend time working on this, reporting etc.
One problem with this from the OSS perspective seems to be that many important OSS projects e.g. MySQL, Qt etc. are dual licensed. Many of the patents had something to do with storage systems, so they could be helpfull for a database application. Unfortunately they can't use it.
Why go through all this, when you can simply just solve the problem? If you think about it for a few seconds, you come to realize that the solution they want is a simple brute force. Why? Because the digits are completely random and so are the primes too, so there can be no constructive method (with the math we know) to find a faster way to solve it.
:)
If you code a program you'll notice that you'll find a solution almost instantly. The interesting problem is the 2nd one, because it actually requires that you think. But that one is easy too.
I don't think this is any different than what most Europeans had to get used to when countries changed their currency to the Euro. To me it took about a year before I could start thinking in Euros. For example if someone told me that xxx cost yyy euros, I had to convert the number to Finnish marks, so that I could tell if it's cheap or not. Now I don't have to do that anymore.
It would be a pain at first, but people do get used to it.
I think the RFID scanner should simply start the alarm if it gets data that it can't read. Garbled data should be a sign of something strange going on. The problem is that this RSA device doesn't remove the signal, it only brakes it.
After reading through the press-release, this seems to be just another way for a pathetic company to try get some publicity.
Use crappy headphones or speakers and you're guaranteed to get lots of that precious extra noice...
There's nothing wrong with vi. If someone accidentally removed "lock" from your system and you don't want to logout when you leave, then vi comes to the rescue!
I wonder if they can deliver enough of it for my needs? Wonder when they start to consider my toilet paper consumption as a denial of service attack... as that was clearly not permitted according to the EULA.
If you've come up with a truly asymptotically better algorithm, then why don't you talk with someone at you CS department? I'm sure someone would like to take a quick look, at least it shows that their students actually want to learn and do study/think/research on their own.
Have you analyzed your algorithm yourself? Is it asymptotically equivalent to something like quicksort, but it just has a tighter inner-loop? Does it have a better average running time on random input? Does it work always (or is it flawed/probabilistic)? I think those are the questions you need to have answered.
This seems very promising. The videos from Strang's lectures on linear algebra seemed nice as I've been reading Strang's books my self.
As some people noted, this is in no way revolutionary as practically every university has webpages for each of their courses. Atleast at the university where I'm studying (Helsinki University of Technology) we usually get good lecture notes from the lecturer as a complement to whatever book we're using, but they never get published on the course's webpage. This is unfortunate, as many other people could benefit from these lecture notes if they were put online. There's lots of good quality material in different universities and I'm grateful to the lecturers who put their notes online. When trying to learn material covered on different courses I usually make a little search on the web trying to find lecture notes of the subject.
I hope many more universities make it a policy to publish their material. I think this would also reduce the amounts of rushed or otherwise bad lecture notes, as I don't think anyone will put handwritten scanned lecture notes online, so everyone would have to typeset them on a computer. This is something you have to struggle with in the introductory math courses here at HUT...
Hopefully there's only a countably infinite number of licenses...
Totally agreed. I'm not very keen of doing the logic at the application side as that means lots of I/O overhead when moving more data than needed from the DB server.
Another thing that I'd like to see soon is foreign keys with integrity checking. It's always nice to do some db corruption prevention on the database side. For all larger databases, this is very important. I know that applications that corrupt its database are broken and should be fixed, but if someone hits on a bug that hasn't been caught and breaks the database you definately want the database to prevent it. The other option, restoring it from backups, is not what you want to do.
I'd recommend some caution when you do that. If you've ever had any problems arguing with your girlfriend, then thing about arguing with 2 or 3 of them...
Huh, no firewall that does this? Have you ever tried IPFilter with any of the BSDs or pf for OpenBSD? What you want to do is run the firewall as a bridge, then you just filter on the incoming or outgoing interfaces. (With ipf, you can only filter on incoming data, but you'll find everything you need in each firewall's documentation...)
And to all those people complaining about this being illegal. I can't think of any company thinking of buying a Cisco, but would then switch to a clone because it's possible. And to all those people telling you how to use a ipf/pf/iptables/ipchains/ipwhatever system for testing, it just ain't gonna teach you much about a PIX.
I've been playing around with a few PIX boxes and if I had never used a Cisco product before (i.e. configured routers/switches, router access-lists, generally knowing IOS) I would have been quite clueless with a PIX. So you can teach yourself firewalling concepts with open source firewalls and the theory behind different protocols (which is mandatory knowledge for anyone configuring firewalls...). Also commercial firewalls really don't have any features not present in open source firewalls (except that configuring stuff on open source firewalls generally is more time consuming when some features aren't integrated, like VPNs...). Anyways if someone would throw a PIX in front of you and tell you to configure a firewall, I bet you wouldn't get the job done (in a reasonable time).
Actually, it's not anti-DRM hardware, it's hardware *without* DRM. Thus there's no circumvention device of any kind, there's just nothing to circumvent.
However, this will be illegal if hardware DRM support becomes a mandatory part of PCs...
You don't call the code in a function pointer. The function pointer is just a variable and it only contains the address of the code you want to run. The machine instruction only gets the address of the function pointer, loads the value of the address from memory and jumps to the memory location given by the function pointer's value. The only thing you'll need to do is to read the function pointer's value (no executable memory needed).
Of course you probably want to modify the function pointer's value, so it has to be in writable memory, thus the kernel has really no way to protect it. So it doesn't have to be on the heap, but that doesn't help you much.
It's called sysctl.
Actually KPNQwest told their employees that they should contact the appropriate authorities to get their salaries from what's left of the company. So basically they've kicked out all their employees without having money to pay their salaries. To me this sounds like they don't have much money...
That's why I surf for pr0n using lynx...
Well the best multiplayer games ID has ever created were QuakeWorld and Doom. Now somehow it seems like every time a game is designed for a multiplayer experience it gets pretty lame, because the atmosphere in the game just isn't right.
:)
Quake III for example was pretty damn boring. Trying to balance weapons etc. and analyze in advance how the game will be played is just stupid. The nice thing with single-player games is that weapons are usually anything, but balanced also levels aren't symmetric or "balanced" in any way and that is a good thing.
The trend has been to balance the gaming environment and design the levels for better teamplay. This is just stupid as the whole idea with teamplay is to get control and kick the other teams ass. For example look at dm3 in quakeworld. It's extremely unbalanced, but still probably the best teamplay map ever created. Also dm4 is probably the best duel map ever created and both are extremely unbalanced. Now many people say that if a map is unbalanced the whole outcome of a game is set by the first spawn. Then how come certain people always win on dm4, the so called "dm4 specialists"? If it were just a matter of luck, then according to classical probability theory they would lose at least sometimes. In short the whole argument of multiplayer "design" is flawed.
The atmosphere in DOOM III seems to be very cool and I'm quite sure that it's going to be the best multiplayer experience Id has created since QuakeWorld, even though it wasn't designed multiplayer