I would say that software is far less limited than hardware. Hardware is limited by technological, physiological and cost restraints. Software is affected by these on a much lesser scale.
With one piece of hardware (the x86 instruction code, for instance) you can create millions upon millions of software applications. One piece of software, however, can usually only be made to run on a small number of pieces of hardware.
The whole principle of hacking is to share and divulge information and ideas. You claim information wants to be free, then proceed to "lock down" your Linux box (this is the modern day hacker's paradox).
According to Steven Levy, the "original hackers" (those working at MIT) had a small drawer where all the paper tapes could be read, borrowed, changed and altered. Anyone could have access to the box.
When the university developed their first time-sharing machine (replete with user names and passwords) they hated it. Not for the fact that they were losing computer power (although some would argue this was important) but because information was hidden from other users.
They accepted the time-sharing box under one condition: users could have passwords, but all user information could be read by any other user. This included admins. So the standard user could change and view the admin's files, and vice versa.
The tape drawer was reborn.
One would argue that the modern-day hacker has really lost touch with what hacking is supposed to mean. It's about sharing, and by "locking down" boxes and trying to break into web servers to expose security, we're moving farther and farther away from that ideal. We should go back to the tape drawer, people.
P.S. Read the book "Hackers" by Steven Levy. You'll be glad you did.
So Microsoft is taking access information for the default bookmarks it puts in its own browser. What's the big deal? You can always change the bookmarks to go to the site directly (right-click any bookmark and choose Properties).
has MS done anything to not deserve a criticism like that
I'd like to have you run Windows 2000 client for a couple of weeks (burn a copy off your friend if you must) and see if you still have the same sentiment. I'm not a huge fan of Microsoft, but Windows 2000 is perhaps the single best piece of software I've seen them write.
Many Microsoft Research studies are purely academics. They do some great work. Only after it gets to their employer (and Bill Gates's teams of programmers get to hammer at it) does it become flawed.
You have to separate the company from the research team.
I don't care what anyone says, I've visited Microsoft Research and they absolutely rock. They are totally disenfranchised with the whole "embrace and extend" tactics of their corporate employer, instead focusing on dedicated research in many awesome fields (think Lucent [Bell labs], Xerox Palo Alto Research Center [Xerox-PARC] and Agilent [HP labs]).
What guarantee is there that the winner will be competent enough to board Mir? Is this a Survivor-like thing, where the winner won't necessarily be voted as "the most competent" but "the best at playing the game"?
Some serious legal precedents could be set up by sending some normal citizen up who passed the test -- but is actually crazy. I'm sure NBC's ratings would skyrocket if the winner accidentally opened an airlock hatch... but at what cost?
Good point. If the monitors are ditched (and the candy-colored covers as well), what's the point in using iMac boards at all? Why not throw together some G4 boards with dual-processing chips(going offtopic -> does LinuxPPC support dual-processing).
It doesn't make a whole lot of sense to purchase $X amount of iMacs if you plan on using only the boards.
While I don't believe the Department of Defense has the strongest security in the world (recent web site hacks are good examples) I wouldn't give them a mere D+. I don't know how much the subcommittee actually looked into the situation, but I would give them at least a C.
It seems like crackers are focusing more on low-level hacks like web site defacement. That said, most of the major hacks (breaking into a classified server not immediately on the web and stealing its contents) seem to be at a low minimum.
I disagree totally. There's a certain pattern and scheme I run based on the players I'm playing and the arena I'm in. There's a lot of strategy involved.
If you're not familiar with UT, yeah, then it seems like you're pointing the mouse at a bunch of pixels and clicking. You're also the player I tend to set up in my sights. Any player who runs around shooting usually get a rude surprise coming around a corner... as their opponent ends up in back of them. I actually think there's a lot more strategy involved with the myriad of maps and players then in chess (played on a 8 by 8 board with set denominations of moves for each player).
I'd highly disagree. All of my practice hours in Unreal Tournament go into earning the big bucks in the big tournaments. Sure, it's not *physically* demanding, it's mentally demanding. Would you say Gary Karposav doesn't deserve the checks he gets from winning chess?
Personally, I'd rather see Gnutella die. It's vastly unfinished, the polish is dreadful, and its cornerstone in the open source community is debatable (I can call a shareware program GNUCrap and people would eat it up).
Napster is an incredibly simple system: one of the reasons it is so revered.
My understanding was that Beam-It software only provided MP3's for CD's users already had. It confirmed the users' ownership by having them place the CD's in and verify the serial number.
Granted, this could be hacked pretty easily. But I'm sure the majority of users wouldn't know how, so I'm at a bit of a loss to see where the willful loss is here.
Sarah Lawrence College, where I go to (quick plug, we got Time Magazine's school of the year award with 4 other schools), has a damn extensive privacy policy. No school administrator or Dean can touch a student's email on the main server, users aren't logged, users have the option to install Netware or not (to use a few shared servers) and won't be logged anyway, all commercial and user web access is kept private, and users don't need to sign in when they use the lab.
This beat my last college, Boston College, hands down. Working in the student computing lab here, the administrators wouldn't even think to use a system like Carnivore.
I've pretty much gone back to Windows since playing around with Linux. It had some nice features (the stability was great, and programming simple programs with gcc was a breeze), but I couldn't stand not running my favorite apps, and the GUI left a lot to be desired.
I've now started using Windows 2000, and am pretty impressed. It does crash, but it's a well-documented visual bug (playing around with OpenGL with beta Voodoo 3 drivers), and only if I attempt a set group of tasks. It runs games extremely well, in some cases better than their Windows 98 counterparts (e.g. Unreal Tournament). I also can use Visual C++ to quickly create W32 apps, and list them as shareware for hundreds of millions of "normal" computer users to use (instead of just Freshmeat users, which though cool, don't represent the average user).
This was posted by an AC under one of my posts, and I thought it deserved more of a spotlight, so here it is again:
I agree - I am nothing but impressed with W2K Advanced Server. I have extensive experience with Solaris/HP/AIX in a production environment, and we have been playing with W2K recently.
I decided to port some production solaris code to W32 to do some 'real world' tests. These apps take large (12+ GB) files from a mainframe and process them for a datawarehouse. They are C++ programs that do file processing --- file in (read) --- manipulate the data (process) --- and file out (write) (the port was simple - no code changes). They are very processor intensive (not so much disk).
We have an 8 way P800 (for W2K AS) and a 24x E10000 and several K class HP's. I moved the data files to the W2K box and ran the fileproc app 8 times at low priority each working on a different file at the same time. This used 100% of the box. Because it was running at low priority, all other box functions worked beautifully. You could not tell they were running from a system perspective. So far just like HPUX or Solaris from a scheduling perspective. A single file test on NT4 Server has same result - the scheduling piece is not new...
The good news is that the W2K box ran all 8 programs in 2 hr 47 minutes. The same result took 5 hr 56 min on the E10000 (6 mths old - 100% of 8 procs) and 14 hr 45 min on HP UX 11 on a 6 way K. Couldn't test AIX box:( I am predicting the results would be similar to the K
Pretty telling - we are now in the process of moving all our mainframe file manipulation software and reporting to W32. The current plan is to sell the 10000 and buy another 8 way or possibly Datacenter. The money saved on hardware is incredible.
BTW - for fun I tried this on Linux box (RH6.1), but the SMP and filesystem (can't handle file's that big) problems prevented any sort of real tests to work. Did work on a 2 way box, but the results were uninspiring (1 program took 4:23) on a cut 2 gb file. Same code - all optimized for the platform and processor.
And for those that haven't used Winders since 3.1 (most of slashdot) - I can have a terminal from anywhere in the world to do remote admin. All the arguments are gone guys - I seriously believe the days of UNIX are numbered. Please check your own facts - ours are strong enough to phase unix out of our shops in the next 12 months.
You've got to admit, you're not going to get a very honest group of answers here. This place is primarily Linux/FreeBSD based, and most of the users hate/do not use Windows or any of its breathren.
Personally, I like Windows 2000. I think it's a very capable system. But getting an honest answer here is like going to a Microsoft convention and saying "What do you think of Linux?"
Slashdot Mirror
With one piece of hardware (the x86 instruction code, for instance) you can create millions upon millions of software applications. One piece of software, however, can usually only be made to run on a small number of pieces of hardware.
Anyone see a major paradox in all of this?
"Freedom is dead! Long live freedom!" ?
According to Steven Levy, the "original hackers" (those working at MIT) had a small drawer where all the paper tapes could be read, borrowed, changed and altered. Anyone could have access to the box.
When the university developed their first time-sharing machine (replete with user names and passwords) they hated it. Not for the fact that they were losing computer power (although some would argue this was important) but because information was hidden from other users.
They accepted the time-sharing box under one condition: users could have passwords, but all user information could be read by any other user. This included admins. So the standard user could change and view the admin's files, and vice versa.
The tape drawer was reborn.
One would argue that the modern-day hacker has really lost touch with what hacking is supposed to mean. It's about sharing, and by "locking down" boxes and trying to break into web servers to expose security, we're moving farther and farther away from that ideal. We should go back to the tape drawer, people.
P.S. Read the book "Hackers" by Steven Levy. You'll be glad you did.
So Microsoft is taking access information for the default bookmarks it puts in its own browser. What's the big deal? You can always change the bookmarks to go to the site directly (right-click any bookmark and choose Properties).
I'd like to have you run Windows 2000 client for a couple of weeks (burn a copy off your friend if you must) and see if you still have the same sentiment. I'm not a huge fan of Microsoft, but Windows 2000 is perhaps the single best piece of software I've seen them write.
Many Microsoft Research studies are purely academics. They do some great work. Only after it gets to their employer (and Bill Gates's teams of programmers get to hammer at it) does it become flawed.
You have to separate the company from the research team.
I'm really considering working there.
Some serious legal precedents could be set up by sending some normal citizen up who passed the test -- but is actually crazy. I'm sure NBC's ratings would skyrocket if the winner accidentally opened an airlock hatch... but at what cost?
It doesn't make a whole lot of sense to purchase $X amount of iMacs if you plan on using only the boards.
It seems like crackers are focusing more on low-level hacks like web site defacement. That said, most of the major hacks (breaking into a classified server not immediately on the web and stealing its contents) seem to be at a low minimum.
Actually, the code looks like it could use a little work. I'm a little concerned about the use of some arrays here.
If you're not familiar with UT, yeah, then it seems like you're pointing the mouse at a bunch of pixels and clicking. You're also the player I tend to set up in my sights. Any player who runs around shooting usually get a rude surprise coming around a corner... as their opponent ends up in back of them. I actually think there's a lot more strategy involved with the myriad of maps and players then in chess (played on a 8 by 8 board with set denominations of moves for each player).
Does anyone else think this looks *exactly* like Windows Explorer? Particularly with the revised design in Windows 2000? Where's the innovation?
Anyone know if Transmeta's chips will be overclockable?
I'd highly disagree. All of my practice hours in Unreal Tournament go into earning the big bucks in the big tournaments. Sure, it's not *physically* demanding, it's mentally demanding. Would you say Gary Karposav doesn't deserve the checks he gets from winning chess?
Napster is an incredibly simple system: one of the reasons it is so revered.
You're using gcc? God... I gave up on gcc years ago for W32 programs. Visual C++ is much better.
Granted, this could be hacked pretty easily. But I'm sure the majority of users wouldn't know how, so I'm at a bit of a loss to see where the willful loss is here.
Echo echo echo...
If you're a true American, like myself, you'd shoot holes in the bucket with your handgun so it would empty itself out. Sheesh. :)
This beat my last college, Boston College, hands down. Working in the student computing lab here, the administrators wouldn't even think to use a system like Carnivore.
I've now started using Windows 2000, and am pretty impressed. It does crash, but it's a well-documented visual bug (playing around with OpenGL with beta Voodoo 3 drivers), and only if I attempt a set group of tasks. It runs games extremely well, in some cases better than their Windows 98 counterparts (e.g. Unreal Tournament). I also can use Visual C++ to quickly create W32 apps, and list them as shareware for hundreds of millions of "normal" computer users to use (instead of just Freshmeat users, which though cool, don't represent the average user).
I agree - I am nothing but impressed with W2K Advanced Server. I have extensive experience with Solaris/HP/AIX in a production environment, and we have been playing with W2K recently.
I decided to port some production solaris code to W32 to do some 'real world' tests. These apps take large (12+ GB) files from a mainframe and process them for a datawarehouse. They are C++ programs that do file processing --- file in (read) --- manipulate the data (process) --- and file out (write) (the port was simple - no code changes). They are very processor intensive (not so much disk).
We have an 8 way P800 (for W2K AS) and a 24x E10000 and several K class HP's. I moved the data files to the W2K box and ran the fileproc app 8 times at low priority each working on a different file at the same time. This used 100% of the box. Because it was running at low priority, all other box functions worked beautifully. You could not tell they were running from a system perspective. So far just like HPUX or Solaris from a scheduling perspective. A single file test on NT4 Server has same result - the scheduling piece is not new...
The good news is that the W2K box ran all 8 programs in 2 hr 47 minutes. The same result took 5 hr 56 min on the E10000 (6 mths old - 100% of 8 procs) and 14 hr 45 min on HP UX 11 on a 6 way K. Couldn't test AIX box :( I am predicting the results would be similar to the K
Pretty telling - we are now in the process of moving all our mainframe file manipulation software and reporting to W32. The current plan is to sell the 10000 and buy another 8 way or possibly Datacenter. The money saved on hardware is incredible.
BTW - for fun I tried this on Linux box (RH6.1), but the SMP and filesystem (can't handle file's that big) problems prevented any sort of real tests to work. Did work on a 2 way box, but the results were uninspiring (1 program took 4:23) on a cut 2 gb file. Same code - all optimized for the platform and processor.
And for those that haven't used Winders since 3.1 (most of slashdot) - I can have a terminal from anywhere in the world to do remote admin. All the arguments are gone guys - I seriously believe the days of UNIX are numbered. Please check your own facts - ours are strong enough to phase unix out of our shops in the next 12 months.
Personally, I like Windows 2000. I think it's a very capable system. But getting an honest answer here is like going to a Microsoft convention and saying "What do you think of Linux?"
Yes. I made a copy for a friend a couple of days ago. Grab a serial off the web and you're all set.