.. unless you penalize bad choices and reward good choices. Many non-determinstic search methods in use rely rather heavily on randomizing when choosing the "next step". Simulated annealing, tabu search and genetic algorithms comes quickly to mind. So rand() all by itself is a poor optimizer, but once it finds something useful, why not stick to it try rand() our way to a better solution?
This kind of thinking reminds me of this old 50's or 60's horror flick where they hooked up all the computers of the world and the computers "magically" became a sentient being which subsequently tried to take over the world.
So, how did you like the rest of Terminator 3? I think 60 teraflops is enough for everyone!
If it's useless to you doesn't mean it's useless for everyone. I have prsonally found it good to take those "you had to be there" -shots, like when partying out with friends or just plainly finding yourself in a place you couldn't predict in the morning.
Sure, a real (analog/digital) camera takes much better pictures, but not everyone has space to carry them around to every place they go. + it's a bigger chance they get stolen.
There have been many jokes about friends (like a picture which says they're gay) which I've gotten to film and then send via MMS directly to my friends. Oh the simple joys =).
This was important to game writers, because some combinations would not work. If "P" is "move left", and "O" is "shoot", then moving left and shooting would not be possible.
1. Hold "O". Computer sees that "O" was pressed.
2. Hit "P". Computer sees "P" was pressed. It assumes that the "O" must be released.
Actually if I remember correctly, the keyboard controller sends signals such as "key down" or "key up", and these are sent serially. There is actually nothing preventing from the keyboard from sending all of these, thus keeping track of every key on board (and if I recall correctly, some very very old keyboards could actually do this).
The problem is that many keyboard manufacturers like to save money. They're in a business, and money is all they listen to. So the current keyboard models tend to contains a key matrix. Different wires going horizontally and vertically through the keys. When you press a button, one row and one column line up, and we know which key it is. When pressing many keys at the same time, it doesn't really matter, as long as they're from a different row+column combination. Or else they will be missed out.
This is (in my theory) what causes the blackout when pressing several keys at the say time. Not all combinations are forbidden, just some. And they all depend on the keyboard model and manufacturer.
The keyboard repeats and handling of ctrl/shift/whatnot are done in software on the PC side (ie. you do it in the bios handler or user/operating system software).
I think that it's actually more insecure to pass the session ID in the URL rather than to store it in a cookie. This is because many people like to send links to a site to friends and such. If there's a session ID attached in the URL, many people will not understand this. And someone with fast reflexes may abuse the current session.
Legistrations such as this one simple promote wrong technological solutions to wrong problems (the problem is not with cookies as a technology, because it's a voluntary action on behalf of the CLIENT).
That's why I think that at the start, perr reviewed closed source projects tend to be better than open source projects. But once the project grows older, so does the focus. Within a company there's always incentive to improve on the existing and create more features. Sure enough, these all are per reviewed, but which small group could actually notice all minor design flaws? These minor details may not be critical at the moment, but may come up later. 6-8 months later with a tight deadline.
With open source, there's always people scanning through every bit of code. And if there's a misplaced loop or a bad construcct somewhere, odds are that someone reports it. Yes, this takes time as no-one is forced to do all this. But at least the popular programs will almost inevitable end up in this state.
Actually I have found it helpful to play while working. I play those usual office games that everyone knows, such as minesweeper and bejeweled. They usually let me move my focus out of the problem at hand, thus usually solving it much faster than actually staring at a screen full of code.
Usually when I play those games, my mind tends to go blank and wander around. When someone comes to talk to me, I usually get distracted and must start all over again. Reading slashdot is only a waste of time as it usually stimulates the mind to think about something else.
But for developing and learning towards an end goal of a concrete subject - say to learn Spanish
The first idea that springs to my mind is to create a "live a day in a foreign city" -game. At the start of the day, you have a checklist of things to do in a city. You then pick places, are given with dialogs (all in the language that is to be learned). If you can't understand something, you can hit a "hint" button and get more and more revealed in a language you understand (like english).
And after all the chores are done (something like 2-4 chores/day) you go to a bar to meet the chicks/guys. Depending on how well you actually managed, the easier it is for you to actually get a date. If you dare to create the game with some balls, you could put some boobs in there as well. Nothing like a good pair boobs to get to the mood. "Damn I really wanna buy the carrot from this old hag so I can get laid tonight":).
By sitting in a lab with a few kids and doing something like blocking their favorite website, 9 times out of 10 the kids that go and find away around whatever you did are the ones that relate best to other people and kids.
Well duh, of course, because they are the heroes that bring you the pr0n. Ah, it always drives the tech forward..
That's because your conscious mind has motivations, derires, pressure (ie. from time) and preferred ways of doing something. And usually when you find one solution, that does not work perfectly, it's hard to completely let go if it.
Whenever I encounter a hard design problem, I usually enter a "no thoughts" mode by doing something simple, suchs as playing minesweeper. As playing minesweeper takes no thought resources, I usually can find that I've been playing for 30 minutes straight and do not even remember if I thought about anything.
Curiously I usually end up with a better solution this way than actually deciding to solve something in the wake. And as simple solutions are usually more beatiful and seem easy, some on my past employers have had a hard time understanding that when I play, I actually work pretty hard and should not be interrupted.
Actually I was able to land a job at the gaming industry with the help of the knowledge I was able to build from the PS2 Linux Kit. While not imminently useful to you right now, maybe, but it might lead to a 'next cool game' some time later.
And I think this is exactly one of the points Sony intended when the kit was released. It really takes some 6 months to fully handle all the details of the graphics system. I don't think this is the case with the XBox however. All the graphics are in a single chip with easy-to-use APIs.
And furthermore, if the modded XBoxes will be running something MS would have only dreamt of, then they will be mad because they just lost profits.
But if a 15 year old does graffiti on a mall wall, you can assure yourself pretty quickly that nothing else happens. All he/she did, was paint the walls. He did not get keys inside, he did not steal anything and he did not plant anything inside.
Now when a 15 year old 'cracks' a site and writes his name all over it, how can you tell what actually happened? It's easy to cover tracks, so you never know if he left backdoors or DDoS clients inside. Nor can you be sure that he did not take something of value (like customer records, software, research).
After *every* break-in, you should clean the system all the way. Yes, it's easy to say that that's why backups are for. Yeah, backups can be used for this purpose, but they usually always set back things for a day at least (if they are frequently taken). And still the downtime costs money.
So all in all, "cyber crime" is a tougher issue, because it's less concrete.
I guess there's only so much shampoo over here, and we didn't have enough for his hair or something
Well, looking at the pictures and dates on his server (rome.ro), on Jan 2002 he went from this to this. And I guess one year does not make you grow hair that much:).
Another big problem is that since Linux is a virtual memory environment, your DMA chains have to be pre-processed by the kernel in order to translate all the virtual memory addresses to physical ones, which basically means your code will never be as fast as it would be on a TOOL, or a PS2 without Linux.
As a side note, I managed to get some 5-10% speedup from just converting a fairly DMA active graphics application from PS2 Linux to the real thing. My application had quite large DMA lists and when "benchmarking" I prebuilt them, just to see how much effect the kernel really has. Looking from the sources, it's kinda sad; the libs first breaks down the packets to a custom kernel format, then the kernel builds them up again.
But I hear there's coming a patch to the PS2 that will update the DMA speed near to the real thing (as I understand it, it overrides the kernel).
As I don't have a PS2 with a modchip, I can't verify this, but I have heard from many instances that if your PS2 can read CD-R:s, so does the Linux. Of course this means that you already must have a PS2 with the necessary mods and thus could use the homebrewn divx player. Advantages with the Linux version are clear. Some people (like me) keep all their divx movies in a separate network server. Plus it's easier to develop for Linux (own patches, frontends, GUIs, filters etc.)
Also I'm sure on the status of the USB CD-ROM support, but I guess it may be possible to plug in an external CD drive which you can use freely.
Design patterns describe some common code design problems, give a solution to it and analyzes the problem and the solution. So in short when you are facing a structural design problem (like "how should all my GUI widgets interact in the GUI toolkit I am writing"), you might wanna check these out.
Usually programmers "use these" without even knowing them (after all, they are common solutions to common problems).
It *could* be possible, now that the xbox version is official, that the 4 player multiplayer game comes from 4 player splitscreen on the XBox (now idea if it has the balls to run it, though).
Just an idea (although a vague one).
Or perhaps it is the best mathematician the world has ever seen. Although no-one really understands this of course, because they keep it a secret. "lift left leg to 15 degrees, move pelvis 3.5 degrees right.."
Slashdot Mirror
rand() is a poor optimiser.
.. unless you penalize bad choices and reward good choices. Many non-determinstic search methods in use rely rather heavily on randomizing when choosing the "next step". Simulated annealing, tabu search and genetic algorithms comes quickly to mind. So rand() all by itself is a poor optimizer, but once it finds something useful, why not stick to it try rand() our way to a better solution?
This kind of thinking reminds me of this old 50's or 60's horror flick where they hooked up all the computers of the world and the computers "magically" became a sentient being which subsequently tried to take over the world.
So, how did you like the rest of Terminator 3? I think 60 teraflops is enough for everyone!
If it's useless to you doesn't mean it's useless for everyone. I have prsonally found it good to take those "you had to be there" -shots, like when partying out with friends or just plainly finding yourself in a place you couldn't predict in the morning.
Sure, a real (analog/digital) camera takes much better pictures, but not everyone has space to carry them around to every place they go. + it's a bigger chance they get stolen.
There have been many jokes about friends (like a picture which says they're gay) which I've gotten to film and then send via MMS directly to my friends. Oh the simple joys =).
This was important to game writers, because some combinations would not work. If "P" is "move left", and "O" is "shoot", then moving left and shooting would not be possible. 1. Hold "O". Computer sees that "O" was pressed. 2. Hit "P". Computer sees "P" was pressed. It assumes that the "O" must be released.
Actually if I remember correctly, the keyboard controller sends signals such as "key down" or "key up", and these are sent serially. There is actually nothing preventing from the keyboard from sending all of these, thus keeping track of every key on board (and if I recall correctly, some very very old keyboards could actually do this).
The problem is that many keyboard manufacturers like to save money. They're in a business, and money is all they listen to. So the current keyboard models tend to contains a key matrix. Different wires going horizontally and vertically through the keys. When you press a button, one row and one column line up, and we know which key it is. When pressing many keys at the same time, it doesn't really matter, as long as they're from a different row+column combination. Or else they will be missed out.
This is (in my theory) what causes the blackout when pressing several keys at the say time. Not all combinations are forbidden, just some. And they all depend on the keyboard model and manufacturer.
The keyboard repeats and handling of ctrl/shift/whatnot are done in software on the PC side (ie. you do it in the bios handler or user/operating system software).
Please correct me if I'm wrong.
Well I do, but no way in earth would I do something like this to her (unless she specifically asks for it :).
I think that it's actually more insecure to pass the session ID in the URL rather than to store it in a cookie. This is because many people like to send links to a site to friends and such. If there's a session ID attached in the URL, many people will not understand this. And someone with fast reflexes may abuse the current session.
Legistrations such as this one simple promote wrong technological solutions to wrong problems (the problem is not with cookies as a technology, because it's a voluntary action on behalf of the CLIENT).
No, just something along the lines of this :)
That's why I think that at the start, perr reviewed closed source projects tend to be better than open source projects. But once the project grows older, so does the focus. Within a company there's always incentive to improve on the existing and create more features. Sure enough, these all are per reviewed, but which small group could actually notice all minor design flaws? These minor details may not be critical at the moment, but may come up later. 6-8 months later with a tight deadline.
With open source, there's always people scanning through every bit of code. And if there's a misplaced loop or a bad construcct somewhere, odds are that someone reports it. Yes, this takes time as no-one is forced to do all this. But at least the popular programs will almost inevitable end up in this state.
Actually I have found it helpful to play while working. I play those usual office games that everyone knows, such as minesweeper and bejeweled. They usually let me move my focus out of the problem at hand, thus usually solving it much faster than actually staring at a screen full of code.
Usually when I play those games, my mind tends to go blank and wander around. When someone comes to talk to me, I usually get distracted and must start all over again. Reading slashdot is only a waste of time as it usually stimulates the mind to think about something else.
But for developing and learning towards an end goal of a concrete subject - say to learn Spanish
:).
The first idea that springs to my mind is to create a "live a day in a foreign city" -game. At the start of the day, you have a checklist of things to do in a city. You then pick places, are given with dialogs (all in the language that is to be learned). If you can't understand something, you can hit a "hint" button and get more and more revealed in a language you understand (like english).
And after all the chores are done (something like 2-4 chores/day) you go to a bar to meet the chicks/guys. Depending on how well you actually managed, the easier it is for you to actually get a date. If you dare to create the game with some balls, you could put some boobs in there as well. Nothing like a good pair boobs to get to the mood. "Damn I really wanna buy the carrot from this old hag so I can get laid tonight"
Just an idea =).
By sitting in a lab with a few kids and doing something like blocking their favorite website, 9 times out of 10 the kids that go and find away around whatever you did are the ones that relate best to other people and kids.
Well duh, of course, because they are the heroes that bring you the pr0n. Ah, it always drives the tech forward..
That's because your conscious mind has motivations, derires, pressure (ie. from time) and preferred ways of doing something. And usually when you find one solution, that does not work perfectly, it's hard to completely let go if it.
:)
Whenever I encounter a hard design problem, I usually enter a "no thoughts" mode by doing something simple, suchs as playing minesweeper. As playing minesweeper takes no thought resources, I usually can find that I've been playing for 30 minutes straight and do not even remember if I thought about anything.
Curiously I usually end up with a better solution this way than actually deciding to solve something in the wake. And as simple solutions are usually more beatiful and seem easy, some on my past employers have had a hard time understanding that when I play, I actually work pretty hard and should not be interrupted.
Oh well..
Let's not forget about the Text Mode Demo Compo. Certainly some nice uses for ASCII.
"competition in Radio And Concert Industries aCT", hereafter known as the 'RACICT' (spoken like 'racist'). Where do they come up with these names?
Actually I was able to land a job at the gaming industry with the help of the knowledge I was able to build from the PS2 Linux Kit. While not imminently useful to you right now, maybe, but it might lead to a 'next cool game' some time later.
And I think this is exactly one of the points Sony intended when the kit was released. It really takes some 6 months to fully handle all the details of the graphics system. I don't think this is the case with the XBox however. All the graphics are in a single chip with easy-to-use APIs.
And furthermore, if the modded XBoxes will be running something MS would have only dreamt of, then they will be mad because they just lost profits.
But if a 15 year old does graffiti on a mall wall, you can assure yourself pretty quickly that nothing else happens. All he/she did, was paint the walls. He did not get keys inside, he did not steal anything and he did not plant anything inside.
Now when a 15 year old 'cracks' a site and writes his name all over it, how can you tell what actually happened? It's easy to cover tracks, so you never know if he left backdoors or DDoS clients inside. Nor can you be sure that he did not take something of value (like customer records, software, research).
After *every* break-in, you should clean the system all the way. Yes, it's easy to say that that's why backups are for. Yeah, backups can be used for this purpose, but they usually always set back things for a day at least (if they are frequently taken). And still the downtime costs money.
So all in all, "cyber crime" is a tougher issue, because it's less concrete.
Worse or better, depending on who your boss is. I would mind working "under" a she.
I guess there's only so much shampoo over here, and we didn't have enough for his hair or something
:).
Well, looking at the pictures and dates on his server (rome.ro), on Jan 2002 he went from this to this. And I guess one year does not make you grow hair that much
Another big problem is that since Linux is a virtual memory environment, your DMA chains have to be pre-processed by the kernel in order to translate all the virtual memory addresses to physical ones, which basically means your code will never be as fast as it would be on a TOOL, or a PS2 without Linux.
As a side note, I managed to get some 5-10% speedup from just converting a fairly DMA active graphics application from PS2 Linux to the real thing. My application had quite large DMA lists and when "benchmarking" I prebuilt them, just to see how much effect the kernel really has. Looking from the sources, it's kinda sad; the libs first breaks down the packets to a custom kernel format, then the kernel builds them up again.
But I hear there's coming a patch to the PS2 that will update the DMA speed near to the real thing (as I understand it, it overrides the kernel).
As I don't have a PS2 with a modchip, I can't verify this, but I have heard from many instances that if your PS2 can read CD-R:s, so does the Linux. Of course this means that you already must have a PS2 with the necessary mods and thus could use the homebrewn divx player. Advantages with the Linux version are clear. Some people (like me) keep all their divx movies in a separate network server. Plus it's easier to develop for Linux (own patches, frontends, GUIs, filters etc.)
Also I'm sure on the status of the USB CD-ROM support, but I guess it may be possible to plug in an external CD drive which you can use freely.
http://www.darwinawards.com/darwin/darwin1995-04.h tml.
Enough said.
Design patterns describe some common code design problems, give a solution to it and analyzes the problem and the solution. So in short when you are facing a structural design problem (like "how should all my GUI widgets interact in the GUI toolkit I am writing"), you might wanna check these out.
Usually programmers "use these" without even knowing them (after all, they are common solutions to common problems).
yeah, and 'elite' is written like '31173' or '31337' just to hide the fact that we are really talking about lamers ;)
It *could* be possible, now that the xbox version is official, that the 4 player multiplayer game comes from 4 player splitscreen on the XBox (now idea if it has the balls to run it, though). Just an idea (although a vague one).
Or perhaps it is the best mathematician the world has ever seen. Although no-one really understands this of course, because they keep it a secret. "lift left leg to 15 degrees, move pelvis 3.5 degrees right.."