Jump to content
xisto Community

qwijibow

Members
  • Content Count

    1,359
  • Joined

  • Last visited

Everything posted by qwijibow

  1. Yet anouther Visual Basic Script, and low level threat according to synmatic http://forums.xisto.com/no_longer_exists/ It removes anti-virus enteries in the registry, attempts to spread via email and MSN messenger and will attempt a DDOS attack on http://www.cmegroup.com/
  2. Hi, sorry i didnt see your post before.You may want to read up on UNIX partitoning documentation.Linux distro's tend not to use partitons to their full potential, usually just swap, root and boot, and sometimes a /home partition for advanced users.but partitions can be used for so much more, and even increace security.For example, the /tmp and /var directorys are writable to all users, yet in linux they are usually both put on the same partiton as root.in Unix, it is common to put such directorys onto a seperate partiton, and mount them with the mount options "noexec nodev"(also mount /home with the options noexec and nodev)this will stop the system from allowing any programs or scripts to be executed on the writable partitons.this will also prevent any users from creating dev nodes (like /dev/hda1) on the writable partitons.Many privilage esculation exploits involve a hacker gaining limited access to a system, then executing a specially written program, however with this partiton scheme, un-privilaged users cannot execute any programs unless the root user has installed them to the root partiton.Partitoning in this manner is not a huge security boost, but does add an extra obsticle to anyone attenmting to get into your system.In BSD, a typical partiton scheme may involde../ (root) partiton should be as small as possable, and only include programs vital to a system recovery.. for example the mount program, text editors, etc etc/home/usr/tmp/etcfind the partiton documentation at freeBSD.org
  3. i have just read the slashdot story https://yro.slashdot.org/story/06/01/30/0432257/microsoft-tricks-hacker-into-jail and the *hackers* side of the story http://www.illmob.org/ Im no expert on US law, but doesnt this whole think stink of entrapment ??? In one part of this story, someone contacts this hacker, and asks for MS source code (for money) the hacker replys saying he no-longer has it, but ts already everywhere on the internet on p2p irc etc etc. the guy then replys saying that he will pay the hacker to download it for him, and mail it. he does... and then is later arrested. This hacker was convicted NT because he downloaded the source code (he was'nt the person who leaked it) This hacker was convicted because the police contacted hi, and begged him to comit the crime of seeling it. he would never have sold it if the investicators didnt hastle him into doing it right ??? There is no doubt that he did commit a crime, but serving 2 years for agreeing to take $60 for what was already in wide circulation on the internet.... isnt this all a little harsh ? Lets put this into perspective... Drunk drivers get a fine, and are sentenced to a few AA meeting.
  4. probably means TWM. for the hard core *nix hackers In the oldern days, for gaming i used to squease everything out of my ssytem by running the games as window managers. its quite funny, an usually works quite nicely. shutdown all x cliebnts and servers. echo "ut2003" > ./.xinitrc startx sets unreal tournament as the window manager application. its not a window manager of-cource. but it them means that the game is the only X client running. for when every frame per second , and every kb memory is crucial
  5. Next they will outlaw pen and paper diary's. Forcing everyone to keep an on-line diary at http://forums.xisto.com/no_longer_exists/ that the government can read. people not keeping logs of all their most intimite thoughts and feelings will be deemeda thought criminal, and dissapear late at night.
  6. I think alot of people have comented without reading the article. the article is intented to show that the technology "singularity" will beat all current OS in certain area's. (some ideas sound really good) but in the porcess, they also show linux being able to do some things much ore efficiantly, for example, start new porcess threads. but hey... we never know what linux 3.0 will hold The actual content of the study compareing current windows with current linux is a little trivial... its like compareing apples and pears and finding some apples are usually more round, and pears are usually less shiny. abut the reason i posted this link, was that it seemed to contradict the GET THE FACTS campain, which shows windows beeing twice as efficiant as linux in web serving.
  7. use openSSH.openSSH is like a telnet login, except that it is encrypted.you can log-in to your computer from anywhere, and d whatever you like.when i am at university, i often ssh into my home computer to check on its progress in whatever it is doing, or issue new commands, give it new things to download.SSH isnt just a way to log-into your sever.. it is THE way.
  8. Fortunatly, Linux is just a kernel.There will always be some distro's aimed at computer illiterates.There will always be distro's aimed at the power users / software hackers.And there will always be distro's aimed at toasters, for users who's needs for toast, is equal to there needs for servers.
  9. Im trying to make my pages W3C compliant, however my pages always fail the first test http://validator.w3.org/ The FAQ says that character encoding is passed to the browser through the HTTP headers, and then gives instructions on how to configure appache and other servers to give the correct character encoding. I have run the main page of Xisto.com though the validator, and although Xisto is not W3C compliant, it passes the character encoding test. so it seems the server is configured fine... any idea's whats going wrong ??? EDIT: sorry, solved. it seems apache here is configured to only send char encoding if there is a character encoding meta tag. the wc3 tutorial assumed that the server was confiigured to auto-detect the char encoding. <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> SOLVED.
  10. Your internet connection is possably 3Megabit MAXIMUM. Even a very old 10Mbit home network would easily handle 3Mbit internet (with a 7 Mbits to spare) but i dont think you can even buy 10Mbit cards anymore. and as for instant messageing..... How fast are you planning on Typing !!! You dont need a Gigabit LAN. A Gigabit lan will offer zero improvements in gaming experiance. It will not improve internet performance. It will be a waste of money. Unless ofcourse you are planning on gaming, whilst simultaniously transfereing files the size of DVD's across the network. Use a 100Mbit network, and wherever possable, uses switches instead of hubs.
  11. Anouther reason i dont like windows.I have never understood why region encoding is even legal.I cant imagine beeing raided by the FBI whilst on holiday in america for using a toothbrush i purchased at home in England !!!On my last holiday, i took a video camera. (qhite cheap, records to AVI files)I then used ffmpeg and to convert the avi to an mpeg2 with ac3 audio (a DVD player compatable file)and used dvdauthor to create the video DVD disk onto a writable DVD disk.The DVD plays fine on Windows98, Linux, and Older DVD players (set top boxes)But on new dvd players, and windowsXP the disk will not play !!!Because i used free and open programs to make the DVD, the dvd is NOT css encrypted, and does not have a region lock.Because of this, WindowsXP wasnt quite sure if my DVD was a legitimate DVD (it was, as i recorded the video myself from my holiday)So it decided to lock me out, and refuse to play the media.This Technology Slightly in-coviniences the Film Pirates.And Completely ruins everything for legitimate customers.Do i have to work in Hollywood to make a home-video dvd that will play on my friends computers ????
  12. Sorry to be a wet blanket..... BUT. take the game "Counter Strike: Source" this game uses 10kbps of bandwidth. So the lag factor here is due to latency, NOT bandwidth. in gigabit network, provided that the bandwidth useage is below 100% the latency is no greater than a 10Mbit network. As far as gaming is concerned, the 1gig network is a waste of money. Unless you are planning streaming several DVD video's simultaniusly, anything faster that 100Mbit is wasted. If you go ahead with this, just for arguemnt sake, try replacing the 1Gig router with a 100Mbit router.. if there is any differance to gaming, i will eat y hat, with a large helping of tobasco source
  13. I wouldnt be surprised if a-lot of people were voting for Monster Raving Looney party ! What happens, when aliens make first contact, and we have to explain out time system !!! 24 hours in a day !!! 60 minutes in an hour.. How will the aliens ever be able to take our mathematicians seriously !!! and the beautiful alien women will never come anywhere neer us ! note: this entire post is a joke, im really not insane. honestly
  14. sorry to keep a dieing thread alive, but Xisto is running on a *nix OS.and therefore, chmod will effect which html files and cgi scripts that the server (Apache) will serve / execute.If you have shell access, you can use chmod to restrict page viewing, however not in the way which you are trying.
  15. I ran the test... wasnt too optimistic...then at the end... SMACK.. Gentoo !!!The Distro chooser was spot on for me !i do love my Gentoo !
  16. No. Chmod is enforced by the kernel, allowing or dis-allowing access depending on the user that executed the user-space application requesting access. when you access a web server, you are only a client, and never actually logged on at kernel level. short answer... chmod is not the correct method for restricting access / granting access to differant users browsing the server.
  17. Wow... so half-life 2 should work with the free Wine (ive been paying for cedega) with these patches..Thanks, maybe ill give it a try.
  18. My sisters Guinnee pig was very old, and sick.it possably had a stroke.it couldnt move the back half of its body, it apreared to be parilysed.however, it could eat and drink if the food was placed directly in front of it.she took it to the vet.everyone agreed that the only humane thing to do, was to put it to sleep. (lethal injection, painless)rather than let it continue to live in pain for the few days it had remaining.Ending its life was considered more humane, and ethically he correct thing to do.not giving it a lethal injection was considered to be cruelty.HOWEVER... the guinne pig did not ask for the injection, and the guinnee pig did not give its concent or permission.On the Other hand, a Human, who is fully capable of asking to die, giving concent and permission, is not allowed to do so in most countries.It doesnt make sence.I think peoples emotions, and in-ability to understand someone elses situation cloud there judgements in there areas.What if the authoities tried to solve the urder problem by adding procedures for this type of thing.perhaps a form, that needs to be signed by the person wishing to die, a doctor, and several un-related witnesses , psychologist / nurse / oter doctors / police officer.Or maybe the problem here is religion.religous politicians will not legalise it because their chosen religion conflicts with it.In this case, i would ask the politicians o respect the beliefs of the parties involved, and let the individual concerned decide what his or her religion thinks about it, and choose to follow, or ignore it.
  19. This is intended as an introduction to AI. It covers depth first searches, but not anything too complicated like Neural networks or genetic programming. Exaple code, and the finsihed source code is written in C++, so a basic knoledge of C++ will help. As will a good understanding of recursive functions. A recursive function is a function that calls itself. for example, the following function... int factorial(int n) { if(n==1) { return 1; } else { return n * factorial(n-1); }} example, factorial(5) == 5 * 4 * 3 * 2 * 1 = 120. The game of sudoku is very simple, you can read the rules of the game, along with tips on how to solve the game at http://sudoku.com/ I picked sudoku as an example, because although the game can be difficult to solve, there is a definate finite number of moves untill the game is over, so the AI will know very quickly wether the game solvable or not, and how close the computer is to fins#ding a solution. games like the rubix cube could go on for hours, with no indication of how close the computer is to finding a solution. If you do not know the rules to sudoku, read up on them now. /***************************************************************************************************** STEP 1: Choosing the best type of AI to solve the problem. This type of problem requires a tree method. where all possable moves are considered, all good moves are varried out, untill the game is solved. there are two main differant types, BREADTH FIRST and DEPTH FIRST. This is a Breadth first search: KEY: CIRCLES are game states, variables hodling what game pieces are in what politions, or in sudoku, what numbers aer in which squares. LINES represent a possable move, which would turn the game state from the state at the top of the line, to the state below the line, for example, in chess, quees moves forwards 2 squares, or in sudoku, I place a "3" at coordinate (2,5). NUMBERS within the circles show the order in which the game states were created and examined. This completed tree, shows that in order to complete this puzzle, from the initial game state, the player must make the moves A->B->B in that order. (whre the letters represent a move, e.g. move queen 2 spaces forward) Breadth first search is good in games where we need to find all possable solutions, OR the best solution out of many solutions, OR where the total nuumber of moves to complete the game is un-known. In sudoku, we know the maximum number of moves.. (keep playing untill all squares are not empty), No Solution is any better than any other solution, and we only need to know one solution. So we will be using a Depth first search. This is a depth first search: In this method, we go deep first, going towards the end game as quickly as possable until we get stuck, Then we go backwards untill we find an earlyer state with possable moves not yet explored, and then explore one of them in the same depth first way. The solution to this problem was A->B->D This all looks complicated, but in fact its incredably simple to program using stacks / lists. here is the pseudo code for breadth first search.. // breadth first searchloop { state = Remove a state from the LEFT side of the list. for (all possable moves we can perform to state) { new_state = state.Do_Possable_Move if(new_state == game_over) { finsihed } else { put new_state to the RIGHT side of the list } }} To turn this breadth first search into a depth first search, simple change the line to Think about that pseudo code, and imagine it going through those trees in the diagram.Its really quite simple ! /***************************************************************************************************** STEP 2: Creating the STATE class. The state class needs to remember every details about the current state, and be as memory efficiant as possable, In some types of AI, there will be millions of states created, in sudoku, however, only a few thousand will be used, and only 40ish will be loaded and in memory at one time, so its not too important. we will need... 1)A char array of 81 (to hold a 9x9 board of numbers) i use char instead of int, as char uses less memory as an integer, and we only need to hold 0 to 9. 2) Copy constructors to make copys of game states from other states, or other char arrays. 3) a function to perform a move on the current game state. 4) a function to test if a move is legal 5) a function to check if the puzzle is completed 6) a function to display the current game state graphically. (well.. in an ascii art grid ) here is the class prototype... class state {private: char board[81]; // 9x9 board. coords(x,y) = board[(9*y)+x] bool legalMoveX(int,char) const; // (Y,N) legal to put N into horizontal where line y=Y ? bool legalMoveY(int,char) const; // (X,N) legal to put N into verticle where x = X ? bool legalMoveB(int,int,char) const; // (X,Y,N) legal to put N into 3x3 grid starting at X,Y ?public: state(); // Start with a completely empty board state(const state &); // Copy constructor state(const char *); // Start with board copyed from chat array state(const state &, int,int,char); // (S,X,Y,N) Copy state S, then put N into coord X,Y const char *getMemRef() const; // get pointer to baord (for copy constructors use only) bool legalMove(int,int,char) const; // (X,Y,N) legal to put N into coord X,Y ? bool isComplete() const; // is this state the solved puzzle ? bool isEmpty(int,int) const; // (X,Y) board[(Y*9)+X] == 0 ? void print() const; // Output a graphical rpresentation of the board.}; The full source code to state is http://forums.xisto.com/no_longer_exists/ and http://forums.xisto.com/no_longer_exists/ As this is an AI tutorial, and not a C++ tutorial, i will not be talking through the state class code. However, i have fully commented the source code, so have a look at state.cpp source code for explanations of how the state works.. its all very simple really, only a very basic understanding of c++ is needed. /***************************************************************************************************** STEP 3: The AI !!! The last source code file is here (and contains the AI unction expand()) http://forums.xisto.com/no_longer_exists/ The AI is all done in the function expand(const state &s); bool expand(const state &S) { // RECURSIVE FUNCTION ! char x,y,n; // if this state is complete, print it, and return true. // returning true at any point will cause all instances of expand() to exit. if(S.isComplete()) { cout << "Solution is..." << endl; S.print(); return true; } // for each square on the board (x,y) for(y=0; y<9; y++) { for(x=0; x<9; x++) { // for each possable number to put in that square for(n=1; n<10; n++) { // legal to put n into board at coordinate (x,y) ??? if(S.legalMove(int(x),int(y),n)) { // if yes, perform this move on a new state, and expand it (RECURSE !!!) if(expand( state( S, int(x), int(y), n) )) { // if expanding this function returns true, // then a future state of this state has solved the puzzle // end the recursice loop, also return true return true; } } } // if this square is not solved if(S.isEmpty(x,y)) { // then dont attempt to solve the next one, if its un-solvable now, it will always be so. // an error hase been made, end this brand of recursion, but not all recursion return false; } } } return false;} At first, i wrote this code as a loop (no recursion), with a stack holding all the states. And each state had to remember what possable moves had ben attempted, and which had not yet been attempted. then i realised, that i was simply re-coding exactly what the c++ compiler does with the internal stacks when recursive functions are used. Allowing the c++ compiler to handle all the stack code via a recursive function halfed the size of the expand function, made it a little faster, so i decided to go with recusrion instead of a stack and loop. However recursive runctions can be a little confusing if you are new to them. If you are new to recursion, take a look at this tutorial http://www.cprogramming.com/tutorial/lesson16.html The above function, follows the logic... 1) Is this state a finsihed puzzle state.. if yes, exist all recursion and print the sollution, else continue2) are there any possable legal moves that have not yet been expanded ?3) if no, goto 64) if yes, expand one of them as a daughter state, and start again at 1),4) if expand returned true, exit all recursion.5) if expanded returns false, goto 2)6) dead end ! go back ot a parent state, and start again at 1) read the function code comments for a detailed description of what is happening. NOTE: since this tutorial was posted, i have been working on this program. it has a Graphical user interface (written in c++) and runs under winows, linux unix and macOSX. http://forums.xisto.com/no_longer_exists/ Corrections / Surgestions / Comments, post them here, thanks.
  20. Righty... ooohhhh you cannot mount an audio-cdrom, because there is no filesystem. just aplain wav dumped right ontop o each track. (however you can *Explore* audio cd's and copy from them, but thats a differant question) There are old ways of playing audio cd's, but these include polling the cr-rom drive every few seconds, and is generally quite messy. The best way to do it, is with HAL (Hardware Abstraction layer) make sure the hardware abstraction layer daemon is running, and use a HAL client, that begins playing an audio cd, when HAL detects an audio CD. KDE-3.5 (and possably other versions, im not sure) are such a hal client. in Kde-3.5, when you insert removable media, KDE pops up a message box, asking what you want to do when this Type of media is inserted. for example, with an audio cd, you could select, Play, Rip to MP3, export to Ipod / explore. I would recomend upgrading KDE.
  21. 500$ Us dolalrs for a graphics card !!!Exactly one year ago a built f#the following macine for £550 (900$ish american ???)(ohh, bought the hardware from a ware-house like place, so a lot cheaper than most on-line / commercial hardware vendors)Amd Athlon64 3400+Chainteck Motherboard (400 mhz FSB)1 Gig DDR-400 RamDvd +- Re WriterNvidia GeForce FX graphics (costing £50 at the time)17 inch TFT screen80 gig hard diskMult slot memory card readermy 1 year old graphics card, that only cost $100 still runs games like HalfLife-2 on maximum quality, with a very repectable frame rate (possabl beccause of my 3400+ cpu.. who knows)why spend 5 times more ?I sometimes play half-life death match online.and i hear people boasting about there computer running the game at maximum quality, at 200 frames per second...When the monitor vertical scan beam is only 60 - 80 Hz ( 60 - 80 frames per second)what good is 300 frames per second.dont waste money drawing tripple the number of frames your monitor can handle.. you wont see them.LOL.
  22. http://gumstix.org/ gimstix-connex + net-MMC-stix + audioStix + (optional, breakoutGS + LCD / TFT screen) and you can build your own mp3 player. next time someone looks all smug with there brand new shiny mp3 player... you can whip out your stack of gum sized circuit board thing... and smugly say "Aswell as playing mp3's running a PDA OS, and Apache web server.... my mp3 player is also, in the background recompliling its kernel... can your ipdo to that !??
  23. The hardware compatability issue is a self inflicted wound.There are 2 ways to make hardware compatable with an OS.There is the MS windows Method, and the Linux Method.I will explain..MS windows has a un-changing kernel ABI for device drivers.This allows hardware developers to ship their hardware with a device driver that will probably out-live the hardwares life.For example for many device drivers written for Windows NT/2k still work with WindowsXP Service pack2 (6 years later)The people whomake the hardware, also provide its device driver.The advantage of this, is hardware compatability.However the dis-advantage is stability.Many BSODS, and other compatability issues are caused by faulty device drivers.Device drivers run in kernel mode, and have sufficiant access to brink the whole system down.MS have tried to get around this major issue by poping up message boxes when you install drivers that have not ben "tested by microsoft"but my 4 year old PC-world computer still contains many un-tested drivers (of which many come with the WindowsXP cd !!!)In Linux, the kernel ABI is always open to change.A binary device driver written one day, might not work a month later with the latest kernel.hardware vendors (except for Nvidia.. woo nvidia !) dont have the resources to release a new closed soure binary only driver every time Linux kernel is updated.but thats Okay, Linux doesnt want this, Linux hackers want hardware vendors to release documentation on the hardware so that linux programmers can write there own device drivers, to an equal quality to the rest of linux.or possably to release the drives open source, so thet they can be ported, and any instability's rsolved.Linux chooses hardware incompatability over in-stability.in my opinion, this was the correct choice..I would rather spend an extra 10 minutes checking for compatability before buying hardware, than have to suffer from in-stability.FreeBSD has a reputation for beeing more stable than linux, however FreeBSD-6.0 rashes constantly on my machine due to a buggy reverse engineered nForce-3 network card driver.an excellent OS ruined on my machine, because it listed support for a device, that had a closed source buggy driver.
  24. In the oldern days, there was a certain *Eletism* that went with anyone who even knew the words *GNU/Linux*, Especially for those who could use it.I would prefere it to stay an *Elete* OS.I can just image Linux going mainstream, and millions of people talking abouut a new OS thats just been released, not realising the 2 decaes of work that has gone into it, and 2 decades of people using it before them.
  25. Hardware raid is better. Software RAID is like RAID emulation for people who dont have hardware RAID.
×
×
  • Create New...

Important Information

Terms of Use | Privacy Policy | Guidelines | We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.