RedHat != GNU/Linux Re: After my last month...
on
Halloween VII
·
· Score: 2, Interesting
I can't comment specifically on sound and video editing, but I've never had any of the dependency problems you mention since switching to Debian at work and Gentoo at home.
I can already hear people saying, "But those aren't particularly user-friendly distributions either," and there is some validity to that; you have to know a bit to set them up. But my point is that we have options, and once they're set up and running a Gentoo or Debian system can be quite friendly to the casual user.
I've flushed an incredible amount of time down the toilet over the past few days playing Tribes II on my Gentoo box (works just spiffy, thanks), and plan to get Unreal set up with my next round of mad money (demo worked flawlessly). I was a Red Hat cheerleader for years, they're still good for a lot, but it's the great boon and curse of any set of flexible, powerful tools that when one approach doesn't work, there are infinitely many more left over to try.
Sorry your experience didn't go so well, yes that still happens far too often, yes Windows is still easier on the initial upswing, but no, this particular at-least-as-lazy-as-the-next-guy user will not be going back to Windows any time soon, because my current setup works a lot better than any Windows box I've used ever has. YMMV.
Re:Aren't APPS the real issue?
on
Halloween VII
·
· Score: 1
I don't have much on my machine which looks at all to me like "the same" software put out by Microsoft.
There are metaphors in common, like the whole idea of a "desktop" user interface, but most of those predate Microsoft's entry into graphical environments.
Microsoft clones do exist. I find KDE to be much like Windows in many respects, and of course there are the growing number of tools which will read Microsoft formatted documents, but interoperability isn't "sameness".
Appropriation does take place. I'm a GNOME user, and I understand that the GNOME team has incorporated an increasing number of techiques invented (popularized?) by Microsoft into the underpinnings of their software. But GNOME is hardly MS-like in its overall flavor IMHO, cherry picking of design patterns notwithstanding.
Miguel di Icaza has written (sorry, no URL handy) that the specific tenets of Unix programming (do one thing and do it well, make your programs filters) don't adapt well to the problems of large, complex suites of software, BUT that the spirit of those tenets and the side effects they produce (interoperation with other software, even if it hasn't been written yet; keeping functionality modular and in small chunks) need to be brought forward into the modern era of complex, object-oriented systems. Microsoft has had a hazy grasp of that issue for years now; for GNOME to be catching up and in some cases surpassing Microsoft in a game they invented is, to me, an amazing feat, and not just highly motivated me-tooism.
And don't forget what's under the gloss. The various Unix kernels are still miles ahead of Windows, the development milieu is far easier for a motivated user to interact with, and the system as a whole is infinitely more flexible than anything I've used from any vendor of proprietary operating systems. To a lot of people that doesn't matter directly, any more than access to source code does; they want a box that will turn on and do whatever they already expect a computer to do. But over the long run the ability to adapt to new situations and uses means a healthier software ecology. That, incidentally, is what most makes Microsoft evil in my eyes: having failed to squelch open source software, they're now seeking to poison the waters in which it swims, as Halloween VII and its many predecessors abundantly illustrate.
But to return to the point, Linux never shifted to being "a clone"; Linux grew to encompass systems which include not only the fine CLI-based tools that you (and I) prefer, but also systems with a wide range of other alternatives, including some which look a lot like something from Microsoft, because that is what some people genuinely want, and some of the ideas Microsoft has used are damn good ideas. I don't have to have KDE or a big ol' office suite, nor WINE nor even a GUI at all. I can use most of the latest and greatest hardware but I don't need much more than a wheezy old x86 PC for a minimal system. Those are all solid advantages provided by open code's tendency toward flexibility, which I don't foresee going away any time soon, no matter how many Windows-like applications and features happen to become available outside of the Windows platform itself.
No. M$ can make (and has made) the legal argument that implementation details must be concealed for security purposes. They've been laughed down in the technical arena, but lawyers aren't usually sysadmins or programmers. And the public equates security with secrecy as well -- so much for the court of the press.
Instead of fighting the UNIX family, they could cash in simply and easily by moving the Windows NT/XP base to a true UNIX base
Somebody correct me if I'm wrong, but isn't Microsoft prohibited by a contractual agreement from ever entering the Unix market itself?
I remember reading, years ago, that when M$ sold the rights to the SysVr4 name to SCO, part of the deal said that M$ would cease to compete in the Unix market. So, a few years later when M$ realized that owning the desktop wasn't enough, they had to base their server platform on something else. Hm, what has a track record against Unix? Let's hire a bunch of VMS hackers and put them to work on something new. Behold, Windows NT!
"Linux-only mobos"?? Are you out of your mind? It's remarkable enough that hardware vendors are even mentioning Linux in their product documentation these days. I don't shop hardware all that much but I've never seen anything marketed as "Linux compatible" let alone "Designed for BlueShoe Linux 7.2" or whatever.
Hardware is a low-margin business, you have to sell a fuck-ton, make that a metric fuck-ton, of any product in order to turn a profit. Hardware manufacturers are NOT going to target a minority market for anything, it's asking to lose your shirt.
Microsoft is responsible for a great number of jobs, conducts research that would be too expensive for almost anyone else...
It seems likely that this would be true to say of any company that attained monopoly status in any field of any significance.
I seriously doubt that any of our anti-monopoly laws would have passed if the only motivation ware to keep bullies like Microsoft from kicking sand in their competitors' faces; competition (and thus losers as well as winners) is a vital component of capitalism.
Our anti-monopoly laws exist to promote the health of markets, something that markets don't always do very well on their own, and the present state of the technology market begs for intervention precisely because of the effect Microsoft is having on our industry.
No one wants to compete against Microsoft, because it means losing, period, end of story. So why would anyone invest R&D money in any product which stands any chance of competing against a Microsoft endeavor? Look at the companies Microsoft competes with now: all the major ones existed before Microsoft's rise to dominance (Sun, Oracle, IBM) or got into their fields before Microsoft established its presence there (AOL/T-W).
So: you hear that Microsoft is investing in research in field X, and you're not any of those companies and you don't already have a vested interest in field X. What do you do? You stay the hell away from field X, and you look for some other way to make money. Over time this is bound to have a depressing effect on the industry as any one company, no matter how much they spend, is still a monoculture. Microsoft has no incentive to promote technologies which compete against one another, so the first thing they settle on becomes their own ad hoc standard. If nobody else is trying to set the same standard, the ad hoc standard eventually becomes legitimized.
They look like the industry powerhouse because they're the biggest one left standing in a smoldering, splintered shack of an industry.
Yes, anything you do to knock down Microsoft causes a dip in the NASDAQ. But that's because the market and the technology industry have developed/been pushed into far too great a dependency on this single entity, and too many people can no longer even envision the possibility of any alternative.
Knock them down hard enough, and after the dust settles, the rodents will come out of their burrows to bury their towering bones, and just maybe we'll have ourselves a vibrant, dare I say relevant field again.
The article asked about guarding against worms in principle. Diversity is, in principle, more healthy than monoculture, specifically in the area of resistance to infection.
As other posts have mentioned, there are several worms currently active in the world of UNIX. They don't make the kind of news that Windows worms make because today's UNIX world is not a monoculture. If you find a vulnerability in a Windows system, chances are very good that the next Windows system you find will have the same vulnerability. That's not true when you substitute "UNIX" for Windows, or "Linux", or "Red Hat", or even "Red Hat N.M".
The same vulnerability will sometimes exist across Unices, but writing an automated process that successfully jumps from one flavor of UNIX to another is still tough. You have to either know in advance what kind of system you're infecting, or write your code so that it can run just about anywhere (hard to imagine much beyond a basic Bourne shell script that will do so reliably, and even then...).
Yet another strike in favor of open, diverse methods as opposed to closed, monolithic ones. Sure it's confusing sometimes, but fortunately it's even more confusing for the bad guys and their evil robots.
...and open source developers don't necessarily fit the stereotype of the college student volunteering time on a kewl project.
I think that many of the people who work on free software do it because they or their employers simply see that approach as the most effective way to get the quality software they need. They have problems they need to solve, and this has nothing to do with the stereotypical "volunteer" open source developer scratching a personal itch. A lot of Linux-related work comes out of NASA and other large organizations which need software to get their work done, can't buy what they need off the shelf, and have no motivation for keeping their code secret. Under those circumstances, why not make the source open?
The AT&T/UNIX example is a classic. Had AT&T been allowed to market UNIX as a product, we'd probably all be using some sort of crappy VMS descendant.;)
After 25 years or so of the closed-source experiment, people are beginning to realize that the closed-source approach has its limitations; so, the alternatives are getting attention. But we shouldn't be surprised that people do this, any more than we're surprised that scientists and economists publish in journals.
There are a lot of people out there who don't know how to compile the source they got from someone to get a program to run. To be an efficient Linux user, this is one of the many special skills you have to have.
No, these are skills you need in order to be a Linux/Unix administrator or consultant. I work in an office full of people who don't know their way around the guts of Unix, but they all sit in front of Linux desktops all day and get along just fine, because they're able to offload the low-level stuff to a dedicated admin while they concentrate on getting their work done.
WELL, you might say, what about home users? They don't hire admins. Well in many cases they don't do their own Windows administration either. They take the box out to a service technician, or put the touch on their techno-geek teenager to work on it when it breaks. When my parents' AOLStation gets sick, my phone rings.
And the sort of Windows administration that most people do for themselves can be done just as easily on a Linux box using package management utilitys and GUI management tools like the stuff that has been coming with Red Hat for years now.
Well, so the old story goes, if they really are decent *n*x talent, one warm body will go a lot further than it would in a Windows shop.
Anecdotal evidence: I work in a software development shop. Mostly Linux, some OpenBSD, some Solaris, two or three Windows laptops, one Macintoy for the interface designer. Pound for pound I spend a lot more time fixing the Windows boxen; next comes Solaris; then the OBSD and Linux machines which require next to no day-to-day fixing. I do all I can to ignore the Mac.
This release by the NSA lends legitimacy to the security claims of Linux proponents. Old-timers can feel good because
the NSA "endorses" the use of Linux and the rest of us can finally get a chance to use Linux in government projects.
I wouldn't break out the party hats just yet. This is a preliminary release of code not yet integrated into a usable distribution. I see the significance of this as more like a seed than a full grown tree, it will eventually populate the Linux world with improved security options, but it one cannot assume that any entity, least of all another tentacle of the US gub'ment, will take this as an occasion to get a clue about how to use technology.
They did it by reverse engineering. The WINE guys have said themselves that WINE may never be truly finished precisely because so much of the Win32 API is secret, and also because they must duplicate not on the API as specified (somewhere) but also the implementations in the API as implemented in the various incarnations of Windows.
"*NIX would be even better if some of these common modules would seperate engine from interface, just like the OS is seperated from the GUI."
Agreed, and this seems to echo some of the sentiments (if not the letter) of the talk by Miguel deIcaza posted here a few days ago. The principles that made Unix beautiful to begin with appear to be better exemplified these days by (GAAAK!) IE than by KDE or StarOffice, and I admire the design choices being made by the GNOME and GIMP teams which appear aimed at bringing "small is beautiful" and "laziness is laudible" up from the low-level, "traditional" *n*x utilities and into the application space.
Re-use the code; make the same bits able to function in many different (unforeseen) situations.
> Now, now. You'll never beat Microsoft if that's your attitude.;-)
Winky-thing noted; but just because some people might take you seriously I feel the urge to chime in.
If we're going to beat Microsoft (and who really cares at this point if we do, but that's different flamebait) we won't do it by playing their game; they already own it. Unix-like systems have thrived thus far not on integration between components, but on flexibility that can be made to serve the purpose of integration. Everybody understands i/o redirection, for example, and that's one thing that makes all these command line tools so startlingly powerful in the hands of somebody who has taken the time to learn them well enough to manage their own "integration".
We need a complete, stable browser _badly_; if Opera can be made embeddable, or if it has good support for some form of IPC (haven't begun to look at it yet myself, my download's stalled at 44%), then so much the better, but the closest competetor right now doesn't have those features and also doesn't work very well.
Let KDE and GNOME finish up being compatible with each other, and then let's start asking for compatibility from arbitrary applications.
I've heard of this situation from a friend of mine; if it's what I think it is, your problem is not Linux-specific, rather it is a problem in the way that NetMeeting is implemented which will break any kind of NAT (of which the IPMASQ code in Linux is a specialization).
The problem is that NetMeeting encodes the IP address of the client host within the data portion of its packets, and the other peer then tries to send (at least some of) its responses to that IP instead of the one from which the connection appears to have originated. Since you are most likely using non-routable IP's on your local LAN, packets addressed to your machine's IP address will go nowhere.
I don't think that this is fixable at the firewall without some pretty inventive hacking, probably with knowledge of the internal structure of NetMeeting 3 IP packets as well as some other awareness of your local environment. I don't even know why it is that NetMeeting feels the need to do the job of the IP stack within its own structure; presumably they had a reason but in this case it obviously causes nasty problems.
"...Perhaps they have been using it, even in some small, obscure, or very unnoticed way, and we're seemingly not the worse for it."
No spook agency worth their trenchcoats and sunglasses would use a hole like this in obvious ways. They also wouldn't risk discovery by using this back door on just anyone. Relying on the obvious exploitation of a security hole to prove its existence is bad practice.
Well, in order to transmit a network packet, you do have to store it, however briefly...
So, it is conceivable to me that this law could be abused into putting the finger to somebody who just runs a router.
Which, by the way, is not that qualitatively different from a web server in a legalistic sense; a lot of web site admins have about as much control over what gets posted on their sites as router admins do over what they transmit.
Internet content control by politicians is stupid, stupid, stupid, stupid.
Nuts. I tried to install this version, but it requires versions of gtk+ and glib that are newer than RedHat 6 -- no biggie, go to the GNOME site and -- whoah, they're not there either.
I am really bugged by the fact that GNOME stuff seems to proliferate so irregularly. Where is everybody else going to get all this up-to-date software that I can never find?
Slashdot Mirror
I can't comment specifically on sound and video editing, but I've never had any of the dependency problems you mention since switching to Debian at work and Gentoo at home.
I can already hear people saying, "But those aren't particularly user-friendly distributions either," and there is some validity to that; you have to know a bit to set them up. But my point is that we have options, and once they're set up and running a Gentoo or Debian system can be quite friendly to the casual user.
I've flushed an incredible amount of time down the toilet over the past few days playing Tribes II on my Gentoo box (works just spiffy, thanks), and plan to get Unreal set up with my next round of mad money (demo worked flawlessly). I was a Red Hat cheerleader for years, they're still good for a lot, but it's the great boon and curse of any set of flexible, powerful tools that when one approach doesn't work, there are infinitely many more left over to try.
Sorry your experience didn't go so well, yes that still happens far too often, yes Windows is still easier on the initial upswing, but no, this particular at-least-as-lazy-as-the-next-guy user will not be going back to Windows any time soon, because my current setup works a lot better than any Windows box I've used ever has. YMMV.
I don't have much on my machine which looks at all to me like "the same" software put out by Microsoft.
There are metaphors in common, like the whole idea of a "desktop" user interface, but most of those predate Microsoft's entry into graphical environments.
Microsoft clones do exist. I find KDE to be much like Windows in many respects, and of course there are the growing number of tools which will read Microsoft formatted documents, but interoperability isn't "sameness".
Appropriation does take place. I'm a GNOME user, and I understand that the GNOME team has incorporated an increasing number of techiques invented (popularized?) by Microsoft into the underpinnings of their software. But GNOME is hardly MS-like in its overall flavor IMHO, cherry picking of design patterns notwithstanding.
Miguel di Icaza has written (sorry, no URL handy) that the specific tenets of Unix programming (do one thing and do it well, make your programs filters) don't adapt well to the problems of large, complex suites of software, BUT that the spirit of those tenets and the side effects they produce (interoperation with other software, even if it hasn't been written yet; keeping functionality modular and in small chunks) need to be brought forward into the modern era of complex, object-oriented systems. Microsoft has had a hazy grasp of that issue for years now; for GNOME to be catching up and in some cases surpassing Microsoft in a game they invented is, to me, an amazing feat, and not just highly motivated me-tooism.
And don't forget what's under the gloss. The various Unix kernels are still miles ahead of Windows, the development milieu is far easier for a motivated user to interact with, and the system as a whole is infinitely more flexible than anything I've used from any vendor of proprietary operating systems. To a lot of people that doesn't matter directly, any more than access to source code does; they want a box that will turn on and do whatever they already expect a computer to do. But over the long run the ability to adapt to new situations and uses means a healthier software ecology. That, incidentally, is what most makes Microsoft evil in my eyes: having failed to squelch open source software, they're now seeking to poison the waters in which it swims, as Halloween VII and its many predecessors abundantly illustrate.
But to return to the point, Linux never shifted to being "a clone"; Linux grew to encompass systems which include not only the fine CLI-based tools that you (and I) prefer, but also systems with a wide range of other alternatives, including some which look a lot like something from Microsoft, because that is what some people genuinely want, and some of the ideas Microsoft has used are damn good ideas. I don't have to have KDE or a big ol' office suite, nor WINE nor even a GUI at all. I can use most of the latest and greatest hardware but I don't need much more than a wheezy old x86 PC for a minimal system. Those are all solid advantages provided by open code's tendency toward flexibility, which I don't foresee going away any time soon, no matter how many Windows-like applications and features happen to become available outside of the Windows platform itself.
No. M$ can make (and has made) the legal argument that implementation details must be concealed for security purposes. They've been laughed down in the technical arena, but lawyers aren't usually sysadmins or programmers. And the public equates security with secrecy as well -- so much for the court of the press.
"Linux-only mobos"?? Are you out of your mind? It's remarkable enough that hardware vendors are even mentioning Linux in their product documentation these days. I don't shop hardware all that much but I've never seen anything marketed as "Linux compatible" let alone "Designed for BlueShoe Linux 7.2" or whatever.
Hardware is a low-margin business, you have to sell a fuck-ton, make that a metric fuck-ton, of any product in order to turn a profit. Hardware manufacturers are NOT going to target a minority market for anything, it's asking to lose your shirt.
It seems likely that this would be true to say of any company that attained monopoly status in any field of any significance.
I seriously doubt that any of our anti-monopoly laws would have passed if the only motivation ware to keep bullies like Microsoft from kicking sand in their competitors' faces; competition (and thus losers as well as winners) is a vital component of capitalism.
Our anti-monopoly laws exist to promote the health of markets, something that markets don't always do very well on their own, and the present state of the technology market begs for intervention precisely because of the effect Microsoft is having on our industry.
No one wants to compete against Microsoft, because it means losing, period, end of story. So why would anyone invest R&D money in any product which stands any chance of competing against a Microsoft endeavor? Look at the companies Microsoft competes with now: all the major ones existed before Microsoft's rise to dominance (Sun, Oracle, IBM) or got into their fields before Microsoft established its presence there (AOL/T-W).
So: you hear that Microsoft is investing in research in field X, and you're not any of those companies and you don't already have a vested interest in field X. What do you do? You stay the hell away from field X, and you look for some other way to make money. Over time this is bound to have a depressing effect on the industry as any one company, no matter how much they spend, is still a monoculture. Microsoft has no incentive to promote technologies which compete against one another, so the first thing they settle on becomes their own ad hoc standard. If nobody else is trying to set the same standard, the ad hoc standard eventually becomes legitimized.
They look like the industry powerhouse because they're the biggest one left standing in a smoldering, splintered shack of an industry.
Yes, anything you do to knock down Microsoft causes a dip in the NASDAQ. But that's because the market and the technology industry have developed/been pushed into far too great a dependency on this single entity, and too many people can no longer even envision the possibility of any alternative.
Knock them down hard enough, and after the dust settles, the rodents will come out of their burrows to bury their towering bones, and just maybe we'll have ourselves a vibrant, dare I say relevant field again.
The article asked about guarding against worms in principle. Diversity is, in principle, more healthy than monoculture, specifically in the area of resistance to infection.
As other posts have mentioned, there are several worms currently active in the world of UNIX. They don't make the kind of news that Windows worms make because today's UNIX world is not a monoculture. If you find a vulnerability in a Windows system, chances are very good that the next Windows system you find will have the same vulnerability. That's not true when you substitute "UNIX" for Windows, or "Linux", or "Red Hat", or even "Red Hat N.M".
The same vulnerability will sometimes exist across Unices, but writing an automated process that successfully jumps from one flavor of UNIX to another is still tough. You have to either know in advance what kind of system you're infecting, or write your code so that it can run just about anywhere (hard to imagine much beyond a basic Bourne shell script that will do so reliably, and even then...).
Yet another strike in favor of open, diverse methods as opposed to closed, monolithic ones. Sure it's confusing sometimes, but fortunately it's even more confusing for the bad guys and their evil robots.
I think that many of the people who work on free software do it because they or their employers simply see that approach as the most effective way to get the quality software they need. They have problems they need to solve, and this has nothing to do with the stereotypical "volunteer" open source developer scratching a personal itch. A lot of Linux-related work comes out of NASA and other large organizations which need software to get their work done, can't buy what they need off the shelf, and have no motivation for keeping their code secret. Under those circumstances, why not make the source open?
The AT&T/UNIX example is a classic. Had AT&T been allowed to market UNIX as a product, we'd probably all be using some sort of crappy VMS descendant. ;)
After 25 years or so of the closed-source experiment, people are beginning to realize that the closed-source approach has its limitations; so, the alternatives are getting attention. But we shouldn't be surprised that people do this, any more than we're surprised that scientists and economists publish in journals.
WELL, you might say, what about home users? They don't hire admins. Well in many cases they don't do their own Windows administration either. They take the box out to a service technician, or put the touch on their techno-geek teenager to work on it when it breaks. When my parents' AOLStation gets sick, my phone rings.
And the sort of Windows administration that most people do for themselves can be done just as easily on a Linux box using package management utilitys and GUI management tools like the stuff that has been coming with Red Hat for years now.
Well, so the old story goes, if they really are decent *n*x talent, one warm body will go a lot further than it would in a Windows shop.
Anecdotal evidence: I work in a software development shop. Mostly Linux, some OpenBSD, some Solaris, two or three Windows laptops, one Macintoy for the interface designer. Pound for pound I spend a lot more time fixing the Windows boxen; next comes Solaris; then the OBSD and Linux machines which require next to no day-to-day fixing. I do all I can to ignore the Mac.
They did it by reverse engineering. The WINE guys have said themselves that WINE may never be truly finished precisely because so much of the Win32 API is secret, and also because they must duplicate not on the API as specified (somewhere) but also the implementations in the API as implemented in the various incarnations of Windows.
"*NIX would be even better if some of these common modules would seperate engine from interface, just like the OS is seperated from the GUI."
Agreed, and this seems to echo some of the sentiments (if not the letter) of the talk by Miguel deIcaza posted here a few days ago. The principles that made Unix beautiful to begin with appear to be better exemplified these days by (GAAAK!) IE than by KDE or StarOffice, and I admire the design choices being made by the GNOME and GIMP teams which appear aimed at bringing "small is beautiful" and "laziness is laudible" up from the low-level, "traditional" *n*x utilities and into the application space.
Re-use the code; make the same bits able to function in many different (unforeseen) situations.
> Now, now. You'll never beat Microsoft if that's your attitude. ;-)
Winky-thing noted; but just because some people might take you seriously I feel the urge to chime in.
If we're going to beat Microsoft (and who really cares at this point if we do, but that's different flamebait) we won't do it by playing their game; they already own it. Unix-like systems have thrived thus far not on integration between components, but on flexibility that can be made to serve the purpose of integration. Everybody understands i/o redirection, for example, and that's one thing that makes all these command line tools so startlingly powerful in the hands of somebody who has taken the time to learn them well enough to manage their own "integration".
We need a complete, stable browser _badly_; if Opera can be made embeddable, or if it has good support for some form of IPC (haven't begun to look at it yet myself, my download's stalled at 44%), then so much the better, but the closest competetor right now doesn't have those features and also doesn't work very well.
Let KDE and GNOME finish up being compatible with each other, and then let's start asking for compatibility from arbitrary applications.
-m
I've heard of this situation from a friend of mine; if it's what I think it is, your problem is not Linux-specific, rather it is a problem in the way that NetMeeting is implemented which will break any kind of NAT (of which the IPMASQ code in Linux is a specialization).
The problem is that NetMeeting encodes the IP address of the client host within the data portion of its packets, and the other peer then tries to send (at least some of) its responses to that IP instead of the one from which the connection appears to have originated. Since you are most likely using non-routable IP's on your local LAN, packets addressed to your machine's IP address will go nowhere.
I don't think that this is fixable at the firewall without some pretty inventive hacking, probably with knowledge of the internal structure of NetMeeting 3 IP packets as well as some other awareness of your local environment. I don't even know why it is that NetMeeting feels the need to do the job of the IP stack within its own structure; presumably they had a reason but in this case it obviously causes nasty problems.
Maybe there's a proxy....?
"...Perhaps they have been using it, even in some small, obscure, or very unnoticed way, and we're seemingly not the worse for it."
No spook agency worth their trenchcoats and sunglasses would use a hole like this in obvious ways. They also wouldn't risk discovery by using this back door on just anyone. Relying on the obvious exploitation of a security hole to prove its existence is bad practice.
-m
Well, in order to transmit a network packet, you do have to store it, however briefly...
So, it is conceivable to me that this law could be abused into putting the finger to somebody who just runs a router.
Which, by the way, is not that qualitatively different from a web server in a legalistic sense; a lot of web site admins have about as much control over what gets posted on their sites as router admins do over what they transmit.
Internet content control by politicians is stupid, stupid, stupid, stupid.
Nuts. I tried to install this version, but it requires versions of gtk+ and glib that are newer than RedHat 6 -- no biggie, go to the GNOME site and -- whoah, they're not there either.
I am really bugged by the fact that GNOME stuff seems to proliferate so irregularly. Where is everybody else going to get all this up-to-date software that I can never find?