I suspect that this could be solved in a reasonable amount of time with a heuristic search algorith, such as ACO or a genetic algorithm using the number of matches as a heuristic.
Obviously, brute force, even at a massively parallel execution, is completely out of the question.
Out of interest, how would you access a private method from outside of the class? I'm guessing you could overwrite the vtable or a pointer if you knew the exact entry point of the private method, but how could you determine this at runtime?
You do a little pointer manipulation. In general, point to a public member, increment/decrement the pointer by the distance the desired private member is located in the class, and enjoy. The catch is that you need to know the address of the private member, which may or may not be available. Here is a c++ example that, through a bit of 'cheating', accesses a private member where it shouldn't be doing so via pointer manipulation:
#include <stdio.h> #include <stdlib.h>
int distance;// distance between public and private functions
class Test{
int privateFoo(); public:
int publicFoo();
Test(); }; typedef int (Test::*TestFn)();
int Test::privateFoo(){
printf("Hello from privateFoo!\n"); }
int Test::publicFoo(){
printf("Hello from publicFoo!\n"); }
int main(int argc, char **args){
TestFn functionPointer = (TestFn) (&Test::publicFoo);
Test *instance = new Test();
(instance->*functionPointer)();
char temp[100];
sprintf(temp, "%d", functionPointer);
int publicFooAddress;
sscanf(temp, "%d", &publicFooAddress); /* have to use strings because the compiler won't let you directly manipulate the pointer via math operators*/
int privateFooAddress = publicFooAddress-distance;
I don't like the direction some of the video games are going these days, but this is a bad precedence. When selling games is strictly forbidden for attribute 'A', pretty soon games will be forbidden for attributes 'B' through 'Z' (religion, race, nationality, etc.). Congress should support and enforce the ESRB rating system (more funding, etc) instead of taking direct action themselves - the ESRB is more properly equipped to research ratings and make intelligent, UNBIASED decisions.
There are MANY situations when immutable objects suck, and you've mentioned a good one with regards to speed. Let's say that the variable indent equals 1,000,000. If you use an immutable object, this loop creates 1,000,000 objects and uses at least 500,000,500,000 bytes of ram [n*(n+1)*.5]. With a mutable approach, the lower bound is only 1,000,000 bytes for the entire loop and you'll only create one object instead of one million.
I disagree with Marvin Minsky's statement that "AI has been braindead sine the 1970s". AI, although not quite yet up to par with Hal or Data, has made exciting progress over the past ten years. AI techniques are the main reason that we're learning so much with bioinformatics. Bayesian networks, for example, have been proven to have excellent inference capability and can be used in diagnosis, prediction, and causal determination.
In order to create true AI, though, computer scientists need to work closely with psychologists (or better yet, study psychology in-depth themselves, specifically learning). Learning needs to go far beyond statistics and into more association-based learning that humans exhibit.
In a related story, AOL/Time-Warner [wants to launch] "advanced" IM services without letting others access the IM network.
At what point will the FCC finally force IM interoperability? Imagine if the phone companies could get away with this. Sprint customers wouldn't be able to call MCI customers, for example.
So here's how the game is played: someone notifies the authorities about a child porn site. Several days (weeks) go by before the claim is looked into. Several more days pass by until the site is eventually shut down.
In the mean time, ISP's can act on their own accord by blocking access to known child porn sites. This is a good thing (TM).
Google is already so powerful and so popular that it's already a verb in most people's vocabulary. It is unlikely that Microsoft would be able to overcome this popularity so late in the game, especially since Google is totally platform independent.
DNA is credited to the inception of genetic algorithms. The main idea behind genetic algorithms is the emulatation of natural selection and evolution by means of DNA manipulation. This is accomplished by many DNA manipulation techniques; the two most prominent are crossover, where two different chromosomes swap DNA information, and genetic mutation, where a random [DNA] bit is rotated. If you're interested in genetic algorithms, check out this introduction.
One could construct a two-tape turing machine that simulates the four combinations; if you're interested in mixing computer science with DNA, check out this paper.
is "The Sins", followed promptly by "The Sins Online." Both feature exquisite graphics involving.. well, perhaps it's best to leave that to the imagination.
PAYBACK v1.0 is a new kind of anti-hacker application called an IRS (Intruder Retaliation System) and is based upon "guerrilla" programming protocols and algorithms originally developed for the Chinese Space Program.
The IRS is not only every American's nightmare, it is now Godless Red Thread In Cyberspace
I can forsee the day when software vendors join the likes of RIAA in endless lawsuits because people get around software copyright protection by 'sandboxing' installations on virtual pc.
There haven't been any innovations due to the monopoly that microsoft has. If microsoft had any serious competitor (by serious, I mean a substantial hold in the market share), we would most likely see a lot of innovations.
Suggest to the alumn that you use the money to buy a linux lab. Any funds will go towards equipment and support for the new linux lab. That amount of cash really isn't a lot in the grand scheme of things (think five years down the road.. remember how much changed from 1995 to 2000, for example).
it's the age-old practice of manufacturing a problem while providing the means to get around the problem. i.e., m$ sends out spam and then sells you their software to prevent spam.
Slashdot Mirror
And in other news, the original researchers received a massive donation from Microsoft.
I suspect that this could be solved in a reasonable amount of time with a heuristic search algorith, such as ACO or a genetic algorithm using the number of matches as a heuristic.
Obviously, brute force, even at a massively parallel execution, is completely out of the question.
Something good coming out of Kansas other than a tasteless wizard of oz joke!
You do a little pointer manipulation. In general, point to a public member, increment/decrement the pointer by the distance the desired private member is located in the class, and enjoy. The catch is that you need to know the address of the private member, which may or may not be available. Here is a c++ example that, through a bit of 'cheating', accesses a private member where it shouldn't be doing so via pointer manipulation:
I don't like the direction some of the video games are going these days, but this is a bad precedence. When selling games is strictly forbidden for attribute 'A', pretty soon games will be forbidden for attributes 'B' through 'Z' (religion, race, nationality, etc.). Congress should support and enforce the ESRB rating system (more funding, etc) instead of taking direct action themselves - the ESRB is more properly equipped to research ratings and make intelligent, UNBIASED decisions.
for (int x= 0; x indent; x++) output += "\t";
There are MANY situations when immutable objects suck, and you've mentioned a good one with regards to speed. Let's say that the variable indent equals 1,000,000. If you use an immutable object, this loop creates 1,000,000 objects and uses at least 500,000,500,000 bytes of ram [n*(n+1)*.5]. With a mutable approach, the lower bound is only 1,000,000 bytes for the entire loop and you'll only create one object instead of one million.
I disagree with Marvin Minsky's statement that "AI has been braindead sine the 1970s". AI, although not quite yet up to par with Hal or Data, has made exciting progress over the past ten years. AI techniques are the main reason that we're learning so much with bioinformatics. Bayesian networks, for example, have been proven to have excellent inference capability and can be used in diagnosis, prediction, and causal determination.
In order to create true AI, though, computer scientists need to work closely with psychologists (or better yet, study psychology in-depth themselves, specifically learning). Learning needs to go far beyond statistics and into more association-based learning that humans exhibit.
we're in effect giving an anti-spam company a DDOS with the /. effect. way to go, guys.
(since it's already slashdotted, that is)
Does anyone have a link to an online video of this commercial for those of us who haven't seen it?
This is now every Dentist's business model.
1. Have coder install endless soda fountain.
2. ????
3. PROFIT! [because coders' teeth fall out]
Here's my submission:
Wait for someone to have a great idea, then steal it.
Oh, wait, they already do that.
In a related story, AOL/Time-Warner [wants to launch] "advanced" IM services without letting others access the IM network.
At what point will the FCC finally force IM interoperability? Imagine if the phone companies could get away with this. Sprint customers wouldn't be able to call MCI customers, for example.
The FCC needs to intervene soon!
So here's how the game is played: someone notifies the authorities about a child porn site. Several days (weeks) go by before the claim is looked into. Several more days pass by until the site is eventually shut down.
In the mean time, ISP's can act on their own accord by blocking access to known child porn sites. This is a good thing (TM).
Google is already so powerful and so popular that it's already a verb in most people's vocabulary. It is unlikely that Microsoft would be able to overcome this popularity so late in the game, especially since Google is totally platform independent.
DNA is credited to the inception of genetic algorithms. The main idea behind genetic algorithms is the emulatation of natural selection and evolution by means of DNA manipulation. This is accomplished by many DNA manipulation techniques; the two most prominent are crossover, where two different chromosomes swap DNA information, and genetic mutation, where a random [DNA] bit is rotated. If you're interested in genetic algorithms, check out this introduction.
One could construct a two-tape turing machine that simulates the four combinations; if you're interested in mixing computer science with DNA, check out this paper.
It's official.. any traffic correspondance with slashdot posts made by CmdrTaco will automatically have the "evil bit" set to 1.
is "The Sins", followed promptly by "The Sins Online." Both feature exquisite graphics involving .. well, perhaps it's best to leave that to the imagination.
PAYBACK v1.0 is a new kind of anti-hacker application called an IRS (Intruder Retaliation System) and is based upon "guerrilla" programming protocols and algorithms originally developed for the Chinese Space Program.
The IRS is not only every American's nightmare, it is now Godless Red Thread In Cyberspace
I can forsee the day when software vendors join the likes of RIAA in endless lawsuits because people get around software copyright protection by 'sandboxing' installations on virtual pc.
There haven't been any innovations due to the monopoly that microsoft has. If microsoft had any serious competitor (by serious, I mean a substantial hold in the market share), we would most likely see a lot of innovations.
Competition = innovation.
No competition = not much changes in 10 years.
You pay your law enforcement agencies for their time because you could 'possibly' commit a crime.
Suggest to the alumn that you use the money to buy a linux lab. Any funds will go towards equipment and support for the new linux lab. That amount of cash really isn't a lot in the grand scheme of things (think five years down the road.. remember how much changed from 1995 to 2000, for example).
it's the age-old practice of manufacturing a problem while providing the means to get around the problem. i.e., m$ sends out spam and then sells you their software to prevent spam.