kSquared

Members
  • Content count

    4378
  • Joined

  • Last visited

Community Reputation

1356 Excellent

About kSquared

  • Rank
    Contributor
  1. Great Idea

    Surely you jest.
  2. Asking Girl to Marry Me Over YouTube? y/n?

    I declare this the greatest idea to be introduced on GDNet since the Giant Space Pipe.
  3. This message is from the future

    Quote:Original post by slayemin Unlike matter, there is not a "smallest size" for time. There is not a window of time which spans between one instant to another. Sort of. Although time appears to be a "river", it might be more accurately described as a conveyor belt moving crates one at a time. See Planck length and Planck time: Quote:This is the ‘quantum of time’, the smallest measurement of time that has any meaning, and is equal to 10-43 seconds. No smaller division of time has any meaning. Within the framework of the laws of physics as we understand them today, we can say only that the universe came into existence when it already had an age of 10-43 seconds.
  4. Picking out classes and triple majoring...

    Quote:Is it over my head to attempt such a thing? I expect it to be hard, but will it drain every last bit of energy out of me and make my college life just plain miserable? Before you jump right in, test the waters. There's no reason to overload yourself right from the start. I was an economics major initially with a minor in computer science, then switched over to double-majoring, then triple-majored with computer science, economics, and applied mathematics, and I didn't find it incredibly strenuous, although some semesters were rougher than others. If salary is any measure, then it worked out well; mine is comfortably in the top decile or so for my location and experience. Make sure you try to spread similar courses out so that they don't overlap too much in one semester. You will be much more likely to burn out if you take 6 CS courses in one semester and 6 econ courses in another, instead of taking 3/3 for two semesters. Find out beforehand from other students what the professors and the course workloads are like. Essentially, this is just like everything else in life: you get out of it what you put into it. Plan ahead and you'll probably be fine.
  5. LOLCode

    For lack of a better word, this is utterly brillant. Someone needs to fire up lex/yacc/bison/etc. and start writing the parser. k2thxbai,
  6. Question about the Halting Problem

    Quote:Original post by ToohrVyk Then, x is the minimal element of a if and only if H(P) returns true (where H is a program solving the halting problem). We have effectively reduced our "is minimal element" problem to the halting problem: if a solution to the halting problem exists, then a solution to the minimal element problem exists. I agree with that, I just don't agree with the other relationship: that merely because you had a P that was easy to H-solve, that your P is an H, which is what the other post seemed to be saying. Quote:In general, in computability theory, reduction is a complex relationship, which depends on the way in which the reduction is done (time and space complexity, for instance) and is generally not symmetrical. Well, there are two types of reduction: Turing (which is symmetrical) and many-one (which isn't). The former is more powerful, because it says that a solution to either is a solution to both. But the latter is more useful in the pragmatic sense, because we don't generally get lucky enough to find Turing reductions (in fact, finding any TR for most "hard" problems amounts to determining that P is NP).
  7. A define is just a preprocessing macro and has no bearing on the runtime structure of your program, at least not from the compiler's perspective (because the compiler never gets to see it). You may place it almost anywhere that you could otherwise place a preprocessing directive. A typedef can go either outside a namespace (which makes it global and causes it to apply everywhere), or inside (which causes it to be scoped to that namespace). Typedefs are safer than defines because the compiler can check the type at compile time, whereas a define is simply a literal substitution ("replace all instances of this word with this one"). hope that helps,
  8. I'm duly impressed that both jpetrie and Washu managed to restrain themselves from using the word "performant" in this thread.
  9. Question about the Halting Problem

    Quote:Original post by Monder Actually in order to prove something uncomputable you'd need to reduce the halting problem to it (as that means if you have a solution to that problem you'd have a solution to the halting problem). Actually, you're saying the same thing either way. "X reduces to Y" means "X is Y"; the precise nature of "is" here depends on the problem you're solving. Whether we say "X reduces to Y" or "Y reduces to X" just depends on which expression is perceived to be more complicated. Because it's generally easier to reduce things to simpler forms, proofs generally show how a more complicated solution is like a previously solved, simpler form. For example, consider the statement "(m^2 - n^2)/(m - n) == m + n". We can prove this either by: (1) Try to gradually make changes to "m + n" until you arrive at "(m^2 - n^2)/(m - n)". (2) Try to gradually make changes to "(m^2 - n^2)/(m - n)" until you arrive at "m + n". If you're successful either way, you've shown that X is Y. Quote:Original post by Monder For a simple example of something computable that could be reduced to the halting problem take the decision problem of deciding whether something is the smallest element of the list. [...] Thus we have reduced the problem to the halting problem, though the original problem is clearly computable. The halting problem doesn't ask whether a specific program will terminate. It asks whether it's possible, in general, to decide if any given program will terminate. In other words, the halting problem asks whether it's possible to write a function H(p) that takes a program p as input, and that returns "true" if the program terminates and "false" if it doesn't. Your program is simply one example of a possible p (any program is, of course), but it's not an H. An H that only works for one particular p is also not an H, so either way, this example does not reduce to the halting problem. Here's another example: imagine that, instead of H, I ask you to write S(n), the successor function. This simply takes any integer n as input and adds one to it. Would you be satisfied that this was really the successor function if it only worked with an input of 271? Your program is trivial enough that you can show that it will, in fact, always halt. You can actually do even better, since you can say how the expected running time of the program is (number of list elements multiplied by the time cost to make a comparison). As long as the list is finite and there is a finite upper bound on how long it takes to determine if a particular element is the smallest thus far, it will terminate.
  10. Quote:Original post by Decept This is all very strange. I had another problem on my computer lately which prevented me from deleting exe files. I thought that it might be caused by some malware, so I have tried a bunch of antimalware programs but none of them found anything. I also thought that it might be connected to the connection problem. Finally I uninstalled my antivirus/firewall program (Panda), so that I could try Norton Internet Security instead. But as soon as I uninstalled Panda, the remove exe problem went away. So I then installed Norton and ran my testprograms and watched Norton Firewall, nothing. Can it be that Panda was the cause of everything, that the programs never really tried to connect? Very strange if that is so. The general mindshare appears to be that Panda doesn't play nice with networking in general, so your suspicion may not be entirely off the mark: -- Panda firewall problems -- Installing XP SP2 broke my Panda firewall et cetera.
  11. hey women

    If there are, I doubt they'll be impressed by your scintillating wit. Have you considered looking here instead?
  12. I'm unstickying this, since it's been up for a while. Hope the class is going well :)
  13. Quote:Original post by BarsMonster This process of serializing an object is also called deflating or marshalling an object. Be careful with this definition. Marshalling generally means moving an object across a boundary of some kind, which is not quite the same thing. Although marshalling necessarily includes a serialization of some kind, it's not serialization in and of itself.
  14. Are you Sirius?

    I am Sirius, and don't call me Shirley.
  15. B&? Donate some blood.

    Quote:Original post by Ravuya Perhaps GDNet should try this out. Turning banned idiots into life-saving bags of fluid human components would make every moderator and staff member into a hero. It's an interesting idea, but there would have to be some kind of alternative for people who couldn't donate, if you wanted to be equitable about it (e.g., donate to the Red Cross an amount equal to the cost of maintaining one unit of blood for a year). For instance, what if you're a hemophiliac? And in the United States (and some other countries), if you're a guy and you've had anal sexual contact with another guy, even once and even if nobody had HIV and you used condoms, you're ineligible to donate blood. If you've been overseas for more than a certain period of time, you're also ineligible. In most cases, being ineligible once makes you ineligible forever. All those qualifications exclude not insignificant portions of the population from donating to begin with. I've volunteered at several blood drives, and you'd be surprised how frequently we have to turn people away because they don't meet the standards.