This data is not being shown in a way that one could say 'windows is more secure' or 'linux is more secure'. You would need a random sampling of an equal number of distributions, windows, linux, bsd, others.. What is not disclosed here (in the article at least) is the ratio of linux servers to windows servers running apps on the internet. I would bet that the majority of e-business systems out there are Java or C++ code running on linux or a proprietary unix distro.
So the numbers here are not suprising, nor do they mean anything significant. The responsibility for understanding and enforcing security on a Linux server is up to the administrator, and the level of knowledge in users out there might be falling behind. Microsoft security is a matter of running around to every windows machine on your network installing patches and cleaning viruses. (or at least thats how it seems to a linux user observing a windows network admin). Microsoft also does a good job of warning admins of security flaws when they are uncovered. There are plenty of security flaws in server applications for linux, but there are also many more of them then are available for windows. You have more of a choice, some of the choices are not so good... Maybe the reason that there are more successful attacks on linux servers is the fact that many admins just let them run, assuming that Linux is Secure, and they don't have to worry much about it.
C++ needs a 'sometimes' keyword. it would allow you to easily create random behavior in any code without messing with random numbers and the like. It would take an integer probability, (the higher, the more likely)
for example:
/* Theres a 30% chance that a grue will appear */ sometimes(30) { room->injectGrue( ); }
Yeah, but if you told me this great software only runs on Oracle or SQL Server, i'd say 'screw you guys' and look elsewhere. If your market is, say, CAD, then of course, you are usually going to be working with Wintel clients. Most industries are slow to leave familiar platforms, especially when 90% of the software that targets an industry is not portable. I can understand writing software that only runs on windows, at least there is a large user base in most industries. What I don't get is companies that still write software to run on archaic databases and operating systems that almost no one uses and most people have never heard of. Like PICK. or Universe.
al-Qaeda or Jemaah Islamiah are the proud new owners of a customs mainframe. Can't imagine what they are planning to get out of that data. Maybe they are just looking for some nice boxen to run their CRM and ERP systems on?
Working at Midway's Chicago office, I was witness to the demise of many arcade titles. It was a sad, sad time, those of us fortunate enough to survive it will never be whole again. Midway's main development office is a labrynth of halls and cubicles, filled with blinking, bleeping arcade prototypes, their guts hanging out in bundles. And who can forget Rosyln Dugas? The goddess of hardware! Some still remain in those halls, slaving away on more modern titles, the sequels of successful franchises, maybe a few new concepts.
But anyway, too bad "The Grid" hasn't made it out of the arcade. The MK team built the whole thing in on a 60 some mhz TI DSP chip (in it's native microcode) and a voodoo 3 card, good luck porting that... its one of the best FPS games I've ever played.
We were working on online gaming at Midway before my team was laid off. After killing off Wavenet (networked arcade games, I came on just as it died), the business heads put us to work on MTN (Midway Tournament Network). After they killed that too, we started working on XBOX and PS2 network stuff, but by then most of us had been laid off. Midway had decided that networking would not be a big thing in the game market for another 5-7 years. (this was in 2001)
Maybe they were right? Lag is my biggest problem when playing online. Playing head to head twitch games is not so much fun when your commands preceed your character's actions by a noticeable interval. There is only so much you can do in software to make up for lag conditions, any game that has a critical timing element will suffer at some point. Some games are just not very networkable.
Too bad the game is so buggy. I like the game so far, but the bugs in the qualifying stages keep me from getting past the first stages. I qualified for sharpshooter, but every once in a while I get booted down to marksman training and have to qualify all over again. The rifle action is the most realistic I have seen in a game.
Most of my cousins and army friends (who are all 10+ years in the force) warn that recruiters will lie to you left and right. Maybe they did 10 years ago, or maybe this is just a myth/rumor that won't go away. I dont know what they tell people. I think some people go in to a recruiter not knowing what to expect. You might think you're really talented and belong in the special forces or in an elite recon unit, but the reality is, you probably aren't, and you probably don't. You have to work your way to it, and that means humping through lots of shit to get there. (and theres a good chance you never will)
Bill gates admits Windows is a dying operating system. What can you do with windows that can't be done (better) with OS X? or Gnome for that matter? Besides hosting viruses I mean. Next windows release? Microsoft Trusted Virus Server!
how about "pump-up-your-stock-ware"? If they were attempting to do anything but that, they would have been very, very far behind in their development process at the outset. Of course, I could write an x86 while reading the morning paper before breakfast.
Sure, but I would prefer if httpd.conf was really an XML file, instead of the psuedo-xml that is still being used in apache2. It takes getting used to.
Windows is just bad code for the most part, (not that we could see it, but you can always tell..)
Linux is, for the most part, just better code, especially underneath where it counts. Its true, most linux users (and admins, hopefully) are more security savvy, but there is also the fact that it's much more difficult to run arbitrary code as root on a unix system. Getting root the dirty way in windows is much easier. If people are dumb and allow access to WU-FTP or something, of course they are going to get hacked.
Agreed! I'd mod parent up, but I've already posted to this thread. I've been working in an environment that includes a perl application using mod_per+apache, mysql, and a java app server. The perl app is 3rd party and is generally considered well put together. Deploying the perl app and apache is always a nightmare, recently spent hours tracking down a bug in a CPAN module. (at least they proved test cases so you can see the modules fail before they install) Automating the deployment is very difficult, obviously the developers expect you to spend hours changing hard coded paths in lines of source code and config files every time you want to deploy it.
Deploying the Java app server and web app is never a problem, it usually takes about 2 minutes and can easily be automated using Ant.
Call me redundant, but Java 1.4.1 has been out for a few months, 1.4.2 was recently released, 1.5 is in development. So I can finally use Collections on BSD? Whooopee. I guess Java developers could care less about BSD anyway, speed not being the primary, all overriding, all encompassing concern.
Um, those layers of logical infrastructure and indirection are called interfaces and they are there to provide abstractions and encapsulation. Which can be very nice in large systems. I started with bash, perl, c, and PHP, but for rapid development of enterprise apps, Java won out. The only thing _close to J2EE is.NET, which is a complete rip off of J2EE (and a complete mess in places). I've worked on web systems hacked together in Perl using a bunch of CPAN modules, none of it comes close to a real development framework. Not to bash perl or anything, but just because you can do something cool using two non-descriptive characters does not mean you should. I dont get paid for using less keystrokes.
Mark Heise keeps brining up 'derivative works' and has mentioned obfuscated code a few times. How can anyone claim rights to derivitave works? Who decides if an idea is derivitave or original, and how can that be proven by looking at code? And unless they can de-obfuscate the obfuscated code, claiming stake to that should not hold up either. SCO is swarming with lawyers suddenly, they have publicly announced that litigation will be their primary source of income (cause SCO Unix is garbage and everyone knows it). Really, they should just become a law firm themselves. What we are missing these days is a death squad that hunts for swarms of overambitious lawyers and massacres the lot of them before they start breeding or litigating. I'll volunteer my services, any one else?
that the movie industry can no longer fool the masses into paying for movies that should have died on the cutting room floor. Maybe now studios will consider spending money on movies that will be worth watching instead of blowing it all on big name err.. talent and overblown ad campaigns. Or not. Making the free flow of information a scapegoat for diminishing profits is certainly not the answer, but then we are talking about an industry that brought us great films like 'Daredevil' and 'Batman Returns again and again and again...
(I wouldnt mind seeing a few more Aliens movies though..)
can be found
here
If the russians have learned from the past at all, they will build it in some remote, dark crater not unlike the Ukraine... Better make that 'biological shield' a few meters thicker.
how microsoft determines responsibility for crashes:
3rd party code makes windows API function call->API function bombs ->3rd party code bombs as a result ->3rd party code is responsible!
I dont know what the intentions for this featur are, but I know Ill be using it to track my drugs and arms runners. No more calling them like 'Where you at?' or 'Where's my guns at?'. Now I'll just know. And when they mess up a deal or get locked up, I can send a cleaner and know where the cleaner is at too!
Have to agree. Libraries are one thing, and more will always appear to provide solutions to new problems. One of the fundamental differnces in philosophy between Perl and (more or less) OO languages is the desire to allow globalized access to variables and functions. There are many arguments for and against this. I dont LIKE it, but I think in the context of a scripting language it's acceptable. You probably use namespace imports and keep everything fully qualified and readable for your own sanity and mercy towards others. This is good practice, but the language does not enforce it. Therein lies another difference: Perl enforces almost nothing. If something can be done, folks are going to do it.
Slashdot Mirror
This data is not being shown in a way that one could say 'windows is more secure' or 'linux is more secure'. You would need a random sampling of an equal number of distributions, windows, linux, bsd, others.. What is not disclosed here (in the article at least) is the ratio of linux servers to windows servers running apps on the internet. I would bet that the majority of e-business systems out there are Java or C++ code running on linux or a proprietary unix distro. So the numbers here are not suprising, nor do they mean anything significant. The responsibility for understanding and enforcing security on a Linux server is up to the administrator, and the level of knowledge in users out there might be falling behind. Microsoft security is a matter of running around to every windows machine on your network installing patches and cleaning viruses. (or at least thats how it seems to a linux user observing a windows network admin). Microsoft also does a good job of warning admins of security flaws when they are uncovered. There are plenty of security flaws in server applications for linux, but there are also many more of them then are available for windows. You have more of a choice, some of the choices are not so good... Maybe the reason that there are more successful attacks on linux servers is the fact that many admins just let them run, assuming that Linux is Secure, and they don't have to worry much about it.
for example:
kaboom! /. effect strikes again. Anyone find a cache of this?
Yeah, but if you told me this great software only runs on Oracle or SQL Server, i'd say 'screw you guys' and look elsewhere. If your market is, say, CAD, then of course, you are usually going to be working with Wintel clients. Most industries are slow to leave familiar platforms, especially when 90% of the software that targets an industry is not portable. I can understand writing software that only runs on windows, at least there is a large user base in most industries. What I don't get is companies that still write software to run on archaic databases and operating systems that almost no one uses and most people have never heard of. Like PICK. or Universe.
al-Qaeda or Jemaah Islamiah are the proud new owners of a customs mainframe. Can't imagine what they are planning to get out of that data. Maybe they are just looking for some nice boxen to run their CRM and ERP systems on?
Working at Midway's Chicago office, I was witness to the demise of many arcade titles. It was a sad, sad time, those of us fortunate enough to survive it will never be whole again. Midway's main development office is a labrynth of halls and cubicles, filled with blinking, bleeping arcade prototypes, their guts hanging out in bundles. And who can forget Rosyln Dugas? The goddess of hardware! Some still remain in those halls, slaving away on more modern titles, the sequels of successful franchises, maybe a few new concepts.
But anyway, too bad "The Grid" hasn't made it out of the arcade. The MK team built the whole thing in on a 60 some mhz TI DSP chip (in it's native microcode) and a voodoo 3 card, good luck porting that... its one of the best FPS games I've ever played.
We were working on online gaming at Midway before my team was laid off. After killing off Wavenet (networked arcade games, I came on just as it died), the business heads put us to work on MTN (Midway Tournament Network). After they killed that too, we started working on XBOX and PS2 network stuff, but by then most of us had been laid off. Midway had decided that networking would not be a big thing in the game market for another 5-7 years. (this was in 2001) Maybe they were right? Lag is my biggest problem when playing online. Playing head to head twitch games is not so much fun when your commands preceed your character's actions by a noticeable interval. There is only so much you can do in software to make up for lag conditions, any game that has a critical timing element will suffer at some point. Some games are just not very networkable.
..your code generator can refactor generated code when it recognizes duplication.
Too bad the game is so buggy. I like the game so far, but the bugs in the qualifying stages keep me from getting past the first stages. I qualified for sharpshooter, but every once in a while I get booted down to marksman training and have to qualify all over again. The rifle action is the most realistic I have seen in a game.
Most of my cousins and army friends (who are all 10+ years in the force) warn that recruiters will lie to you left and right. Maybe they did 10 years ago, or maybe this is just a myth/rumor that won't go away. I dont know what they tell people. I think some people go in to a recruiter not knowing what to expect. You might think you're really talented and belong in the special forces or in an elite recon unit, but the reality is, you probably aren't, and you probably don't. You have to work your way to it, and that means humping through lots of shit to get there. (and theres a good chance you never will)
This must eat quite a few CPU cycles. Any idea what kind of machines/software the RIAA is using to perform these operations?
Bill gates admits Windows is a dying operating system. What can you do with windows that can't be done (better) with OS X? or Gnome for that matter? Besides hosting viruses I mean. Next windows release? Microsoft Trusted Virus Server!
how about "pump-up-your-stock-ware"? If they were attempting to do anything but that, they would have been very, very far behind in their development process at the outset. Of course, I could write an x86 while reading the morning paper before breakfast.
Sure, but I would prefer if httpd.conf was really an XML file, instead of the psuedo-xml that is still being used in apache2. It takes getting used to.
Windows is just bad code for the most part, (not that we could see it, but you can always tell..) Linux is, for the most part, just better code, especially underneath where it counts. Its true, most linux users (and admins, hopefully) are more security savvy, but there is also the fact that it's much more difficult to run arbitrary code as root on a unix system. Getting root the dirty way in windows is much easier. If people are dumb and allow access to WU-FTP or something, of course they are going to get hacked.
Agreed! I'd mod parent up, but I've already posted to this thread. I've been working in an environment that includes a perl application using mod_per+apache, mysql, and a java app server. The perl app is 3rd party and is generally considered well put together. Deploying the perl app and apache is always a nightmare, recently spent hours tracking down a bug in a CPAN module. (at least they proved test cases so you can see the modules fail before they install) Automating the deployment is very difficult, obviously the developers expect you to spend hours changing hard coded paths in lines of source code and config files every time you want to deploy it. Deploying the Java app server and web app is never a problem, it usually takes about 2 minutes and can easily be automated using Ant.
Call me redundant, but Java 1.4.1 has been out for a few months, 1.4.2 was recently released, 1.5 is in development. So I can finally use Collections on BSD? Whooopee. I guess Java developers could care less about BSD anyway, speed not being the primary, all overriding, all encompassing concern.
Um, those layers of logical infrastructure and indirection are called interfaces and they are there to provide abstractions and encapsulation. Which can be very nice in large systems. I started with bash, perl, c, and PHP, but for rapid development of enterprise apps, Java won out. The only thing _close to J2EE is .NET, which is a complete rip off of J2EE (and a complete mess in places). I've worked on web systems hacked together in Perl using a bunch of CPAN modules, none of it comes close to a real development framework. Not to bash perl or anything, but just because you can do something cool using two non-descriptive characters does not mean you should. I dont get paid for using less keystrokes.
This needs to be publicised, as anything which slows down spam can only be a good thing.
You mean anything short of dragging these people out into the street and hanging them from a light post.
Mark Heise keeps brining up 'derivative works' and has mentioned obfuscated code a few times. How can anyone claim rights to derivitave works? Who decides if an idea is derivitave or original, and how can that be proven by looking at code? And unless they can de-obfuscate the obfuscated code, claiming stake to that should not hold up either. SCO is swarming with lawyers suddenly, they have publicly announced that litigation will be their primary source of income (cause SCO Unix is garbage and everyone knows it). Really, they should just become a law firm themselves. What we are missing these days is a death squad that hunts for swarms of overambitious lawyers and massacres the lot of them before they start breeding or litigating. I'll volunteer my services, any one else?
that the movie industry can no longer fool the masses into paying for movies that should have died on the cutting room floor. Maybe now studios will consider spending money on movies that will be worth watching instead of blowing it all on big name err.. talent and overblown ad campaigns. Or not. Making the free flow of information a scapegoat for diminishing profits is certainly not the answer, but then we are talking about an industry that brought us great films like 'Daredevil' and 'Batman Returns again and again and again... (I wouldnt mind seeing a few more Aliens movies though..)
can be found here
If the russians have learned from the past at all, they will build it in some remote, dark crater not unlike the Ukraine... Better make that 'biological shield' a few meters thicker.
how microsoft determines responsibility for crashes:
3rd party code makes windows API function call->API function bombs ->3rd party code bombs as a result ->3rd party code is responsible!
I dont know what the intentions for this featur are, but I know Ill be using it to track my drugs and arms runners. No more calling them like 'Where you at?' or 'Where's my guns at?'. Now I'll just know. And when they mess up a deal or get locked up, I can send a cleaner and know where the cleaner is at too!
Have to agree. Libraries are one thing, and more will always appear to provide solutions to new problems. One of the fundamental differnces in philosophy between Perl and (more or less) OO languages is the desire to allow globalized access to variables and functions. There are many arguments for and against this. I dont LIKE it, but I think in the context of a scripting language it's acceptable. You probably use namespace imports and keep everything fully qualified and readable for your own sanity and mercy towards others. This is good practice, but the language does not enforce it. Therein lies another difference: Perl enforces almost nothing. If something can be done, folks are going to do it.
Just be glad it does not require closing if(coditions) with an 'fi' statement. Christ. I could complain all day about shell languages.