Debian 3.0? That should only be a few years away given Debian's version numbering scheme. Especially since 2.2r0 just finally made it out, and Debian versions seem to have an extremely long lifetime.
Seriously, though. I have to wonder how much this "foundation" will help given the fact it's staffed by members of a lot of very large companies.
Besides, if I remember correctly, there was a group that existed for a while that was trying to make a GPL'ed QT clone.. After TrollTech freed up the license terms to their "free" version of QT, interest died down. Although I'd prefer if it were a little more "free" (I think a BSD-style license would be most applicable to something like this), but Trolltech wants to protect their "professional" edition and the customers they've sold this product to. It's fine and dandy to say "make money from services", except for the fact this hasn't really made money for any single Linux company yet.
KDE and Gnome have both evolved in similar directions, and I honestly don't understand this hostility between them. KDE on one hand, incorporates a nearly complete standards-compliant web browser, including JavaScript, etc. Gnome, on the other hand, has always seemed to try to be customizable above all else.
About the only plus I can think of is free air conditioning (a non-trivial consideration in Phoenix, where I live!), because the exhaust (non-toxic nitrogen gas) is actually quite cool. But in a conventional electric vehicle, air conditioning takes a small percentage of the power necessary to actually move the car, so it's not impossible to have the creature comforts we're used to.
Then compare it to where I live -- in Canada on the prairies... We can have bitterly cold winters (-40 degrees, Celsius or Fahrenheit, whichever you prefer), and I would not want to be without a heater. Although you could heat via the energy created by the released gas, it would not be efficient.
Plus you need to factor the fact that oxygen depravation because of the diluting of the oxygen content in the car. It can seriously slow down one reaction time, and induce drowsiness. Unless they're planning on mixing oxygen into the canisters, although that might make them somewhat volitile.
Yes, except for the fact that you're going to have to deal with things like condensation and freezing. Plus the fact that compressing the nitrogen in the first place consumes energy that must come from somewhere.
XF4 completely redesigned most of how XFree86 worked. It was pretty much inevitable that some things would stop working. As it stands, most cards are not 3D accelerated (and some still aren't 2D accelerated), including some that were under the rather kludgy 3.3.x utah-glx system.
And 4.0 has been plaqued by problems since it's release. It didn't support many of the video cards that 3.3.6 did, had security problems, and wouldn't even build right on Debian. 4.0.1 is a step in the right direction, but I'm still going to treat it with the caution. I'm confident they will get it right eventually, but with any change this drastic, there's bound to be problems.
If this document is meant to be a myth dispeller, it should also include sections that address Linux's weaknesses. The released version of the Linux kernel still has SMP scalability problems, for example, and boasting about 2.4.x when it's not released is no excuse. It's important to note where Linux's strengths and weaknesses are. Linux can beat an NT machine as a server with only 64MB of RAM, but once you start throwing gigabytes worth of RAM things start to go the other way, with NT beating out Linux instead.
Another place that Linux is weak on is on desktop systems, due largely to the fact that whereas Windows and Macintosh both have very stringent UI guidelines, Linux has virtually none. KDE and Gnome are trying to change this, although I feel the somewhat hostile environment between the two is working more to make things worse than improve it. As a developer, I should be able to choose either GTK+ or QT and have the resultant program perform similarly on either platform. And as a user, I should be able to expect that any program I install should try to act reasonably with the preferences I have set in my "prefered" desktop environment. For example, I'd love to see a unification of UI "themes", and have a toolkit independent preferences system where both KDE and Gnome settings are stored via their respective control centres, and shared between both where applicable.
Portions of this document are no better than the FUD it attempts to refute, IMO, due to exaggerations on the behalf of the author. There's no harm in admitting weakness, in fact, it usually shows that you're trying to be honest and unbiased.
Despite the hard feelings I can't help but get the feeling that they should try to co-operate a little more. There's no real good reason why KDE and Gnome/GTK can't share things like themes, and general user preferences.. For a short while the teams appeared to co-operate to come up with a unified component system. I firmly believe that there's no valid reason why the UI settings and preferences I configure in either KDE or Gnome can't be shared between both, giving KDE/Qt apps running on a Gnome desktop a similar "feel" to then as the native Gnome/GTK apps, and vice versa.
1. It's not called masq. It was called IP Masquerading to distingush it from NAT. Linux (as of 2.2) doesn't include any TRUE network address translation services, but rather just a port-based NAT derrivative. NAT, in it's truest form, relys on a pool of public IP addresses that are dynamically assigned (and translated) for internal addresses, and doesn't suffer from the problem that IP masq has in dealing with listening connections.
2. A 486 is more than up for the job. A 486-DX2 running Linux kernel version 2.2.x with ISA NICs will become saturated at about the 3-4Mbit/sec mark. As long as you never see more than that much traffic, you'll be fine.
3. Safety first. I agree that keeping your firewall clean and efficient is very important. However, I find the claims that Linux is less secure than BSD more than a bit bogus. Almost all those server daemons that have had buffer overflows on Linux can be compiled and install into OpenBSD with the same buffer overflows. Security is a journey not a destination is true in ALL cases, even OpenBSD. An incompetent (or inexperienced) administrator can easily turn a secure machine into one that's wide open for anyone to break into.
Most people usually end up compromised because of services that they either never used or never knew about, and therefore didn't bother maintaining. Due to the shortsightedness of most Linux distributors, you'll probably end up "cleaning" dozens of packages out that are completely worthless. Ideally, your result should be a machine that's not listening to anything on the public interface.
4. Raise Hell About Gnome Security Issues. Absolutely! A TCP/IP port should never be opened unless there's a very good reason why this service needs to be advertised to the world. Most of the time, this is just lazy coding, and a place where other types of sockets would probably serve better.
The maintainer for XF86 in Debian stated a long time ago that he would not support 4.0 due to the fact it didn't support enough of the legacy hardware that 3.3.6 did support, and that it was plagued with many problems in general. Since then, he's started work on getting 4.0.1 into Debian, and preliminary.debs are already available at http://www.debian.org/~branden/
And as for the elitist attitude, it's pretty common on IRC, especially after being pestered by morons who do not care to bother try finding the answer to their question themselves.
The reasons for KDE not being in Debian quite a while ago (the old "free vs free" arguement) and have mostly been resolved for KDE2, but KDE2 is still too unstable to even be in the woody "unstable" branch. They may make it in there yet, but KDE2 needs to ship first. KDE1 will probably never make it in. I'm surprised Helix Gnome hasn't made it into contrib on woody or even potato for that matter.
But honestly, Debian just seems ideal for a server -- unlike the other distros, it installs completely basic (not bare enough, IMO, but better than anyone else), and you add daemons afterwards. The APT system is wonderful for when you want a package installed, but don't want to have to mess around with vague dependancies -- it's what RPM should've been. Fancy GUIs are fine and dandy on a personal workstation, but on a server, they introduce unneeded complexity and potential for error.
Considering the fairly sorry state of 3D acceleration right at this moment, it's not all that surprising. Sure we have XFree86 4.0, but it's been plagued with problems, incompatibilities, etc.
It's almost as if the story was acting upon itself. It covers a shoddy website, yet the story itself was posted poorly. I think the slashdot people should file a class action suit -- NO MORE TYPOS!@#$%&*!
Yeah, but unless I'm mistaken, a trademark is usually associated with a particular product, too. Yet, the IOC went after little companies like Olympic Pizza, etc before to get them to change their name. I guess they fail to realize that the word "olympic" is extremely common and that the world doesn't revolve around them. Pity the poor fool who gets in the way of the money making machine, since even if it wasn't intentional, it's always the "little guy" who gets screwed in the deal.
On the internet, yes. But on corporate LANs, it's hard to say when an exec will decide on a whim that IPv6 is where it's at, and wonder why the network's not running on IPv6 yesterday.
"What do you mean our workstations won't work with it? I don't care about that -- this article says that IPv6 is where it's at, and I want it now."
And ever notice they usually have engineered benchmarks that prove it?
"Windows 2000 is faster than NT 4.0!" Yeah, but only if you run it on a system that's lower than the recommended specs.
I think MS has also proven that NT 4.0 is faster than Win9x is faster than NT 4.0. How, I'm not sure. I guess what they say really is true -- There are three kinds of lies, lies, damn lies, and benchmarks.
I think my favorite part of that The Onion story is:
"Think of this as a partnership," Gates said. "Like the ones and zeroes of the binary code itself, we must all work together to make the promise of the computer revolution a reality. As the world's richest, most powerful software company, Microsoft is number one. And you, the millions of consumers who use our products, are the zeroes."
There have been a MULTITUDE of flaws in Outlook and Outlook Express that allowed execution of arbitrary code when looking at a specially contructed (read: MS HTML extensions) HTML message. It's stupid, and most of it revolved around ActiveX.
This is the patent office we're talking about. They wouldn't know software prior art if it walked up and slapped them in the face. Sometimes I think they go out of their way to aprove absurd patents. It'd be nice if software patents weren't legal, really, it would.
My favorite little "bug" in Win9x is the fact you get ONE shot at installing the networking files... If you add a networking device, and click cancel instead of putting in the Win9x CD-ROM, it'll never try to install VREDIR.VXD, VNETSUP.VXD, and a handful of other VXD files ever again. Or the other issue where going into the network control panel in Win95 will cause it to copy the same files usually about 4 times and then tell you to reboot despite not changing anything. And of course, it always wants to overwrite the updated version of SECUR32.DLL with the one on the CD. It's slightly better with Win98, but not much. I seriously doubt any of these quirks would require substancial programming effort to fix.
I think one of the best accomplishments for Windows is the fact that all standard Windows dialogs and applets fit in 640x480 with the standard fonts/color scheme. This is one of the things that irratates me to no end when I use X. Everyone assumes you're capable of supporting 1600x1200. Some apps don't even fit in 1024x768 well, let alone the 800x600 that is the maximum for many notebook computers.
Once you learn the keyboard accelerators for Windows, there's very little you can't do in Windows with no mouse (I think the taskbar is one of the few inaccessable things that require mouse use). Some of the shortcuts may be clumsy, but at least they're there. On the other hand, I find myself searching for the mouse constantly in X because some programs and window managers just plain refuse to acknowledge that someone would rather type on the keyboard than touch a mouse. KDE/QT tends to be one of the better toolkits for this, however it's still possible and easy to find a program that won't respond to anything but a mouse.
And despite how many people love to hate Windows and it's UI elements, I have to admit, they're fairly light on screen real-estate. Windows add about 18 pixels at the top of a window, and 2 pixels at each of the sides and bottom. Most of the common controls are distingushable from each other (unlike the motif widgets which mostly all look the same). They may not be the prettiest, but they work.
I agree with the previous poster in that these real-world UIs are aweful, even more so since they almost always ignore the keyboard, and always seem awkward to use. Fighting with circular dial controls, all of which act differently, is not my idea of a good UI design.
Now, don't get me wrong. Skins aren't all bad. But unless they apply to all of the user interface, they're a bad idea except for power users.
I think Netscape 6 would be a lot better if it used the Windows common controls on Windows by default, the MacOS common controls on a Mac, etc. GTK+ is a nice toolkit and all, but when you enforce a UI on users that aren't familiar with it, you're asking for problems...
Re:Where is ESS 1868 support?
on
BeOS For Linux!
·
· Score: 1
The largest change between the 1868 and 1869 was a slight revision and the addition of more Windows friendly Plug and Pray, er, Play. The problem with the persistant IDE controller (for cards without the IDE connector) on the 1898 was also resolved on the 1869. Either way, both are extremely low cost, low quality chipsets. And both are full of bugs with their SoundBlaster Pro emulation.
Honestly, I think breaking up Microsoft would probably help them. The company has become very stagnant, and any "innovation" they come up with was either just bought or stolen from another company. Lets look at the difference between Office 97 and Office 2000 for example.. What did they really add? A 3D "office assistant", and annoying "personalized" menus. I get the distinct feeling that if they were broken into smaller companies they might have some more initiative to steal better ideas from their competitors.
Microsoft HAS done a lot of good for the PC industry, such as the widespread acceptance and use of PCs worldwide, as well as making sure their new OS releases are so unbarably slow that it pushes the industry to make faster and better products so the OS isn't quite so slow anymore.
Regardless of what you think about the user interface elements, the fact remains that Windows' UI is effecient enough to make things easier to work with. It's possible to operate just about everything in Windows via just the keyboard (I don't think you can say that about MacOS for the most part). Frankly to me, it doesn't matter what the widgets look like, as long as they work and are distinguishable enough from each other (Motif and it's clones are bad for that). Honestly I can get many things done faster in Windows than I can in the X-Window system due to these keyboard shortcuts.
So, you can laugh all you want, but it doesn't matter.
BTW, I love my 4 button mouse, and I won't use anything with fewer than 3 buttons now.
Re:CNN has a report on this.
on
Protesting DMCA
·
· Score: 1
Realistically, regional lockouts are NOTHING new. The video game console industry has been doing it for years, however for consoles, anti-region lockouts adaptors and modifications are much more common and often a little less controlled. On the SNES, for example, I believe Nintendo changed the cartridge design slightly to prevent you from plugging in a Japanese SFC game into a North American SNES by way or a couple plastic tabs. The Playstation has region encoding on the CD (along with copy protection) that has been defeated by gamers soldering some changes to their PSX. Both Nintendo and Sony upon learning of this happening made changes to either their games or their consoles to curtail the ability to use these bypasses.
Personally, I like DVD. The image is clear, you get aditional features with the movie, but currently, I'm just SOL if I want to use my DVD-ROM to play any movies. Of the two OS's I'm currently working with -- Windows 2000 and Debian Linux, neither support DVD decoding at this point. The idea of a binary only player for Linux bugs me somewhat due to the fact that the idea of binary compatibility is completely lost on Linux, where if it's a bug, it's going to be fixed, no matter how much shit is depending on it. In contrast to the Windows world, where Microsoft has to be extremely careful about fixing bugs in their library calls because they could easily break hundreds of programs.
Unforuntately, this is what you get when you try to apply a patent process that's incredibly old and dated to a technological area that thrives on incremental improvements. I can see the need for patents on physical devices -- often the inventor of a device doesn't have the capital needed to produce the device, however this isn't true in the web/software world where even some 13 year old kid can come up with something new and original and start making money off of it immediately.
The patent system has a lot of problems when applied to software. For one, they last too long. By the time the patent has expired, it's usually so archiac that it's been replaced by newer software technologies that perhaps didn't have such restrictions. Another problem is that when the patent HAS expired, the patent itself is worthless, since it doesn't provide and subtancial information on the implementation to make the release of "intelectual property" worth it. I read part of the Powerquest partitioning resizeing patent, and gathered absolutely no useful information from it. I gathered more information by actually using Partition Magic a few times, and observing it's behaviour. If patents on software only lasted say 2 or 3 years, and required a sample implementation to be submitted with the patent, perhaps the patent system would be of some use, however since it's not, it'll never benefit the small developers.
Slashdot Mirror
Seriously, though. I have to wonder how much this "foundation" will help given the fact it's staffed by members of a lot of very large companies.
Besides, if I remember correctly, there was a group that existed for a while that was trying to make a GPL'ed QT clone.. After TrollTech freed up the license terms to their "free" version of QT, interest died down. Although I'd prefer if it were a little more "free" (I think a BSD-style license would be most applicable to something like this), but Trolltech wants to protect their "professional" edition and the customers they've sold this product to. It's fine and dandy to say "make money from services", except for the fact this hasn't really made money for any single Linux company yet.
KDE and Gnome have both evolved in similar directions, and I honestly don't understand this hostility between them. KDE on one hand, incorporates a nearly complete standards-compliant web browser, including JavaScript, etc. Gnome, on the other hand, has always seemed to try to be customizable above all else.
Plus you need to factor the fact that oxygen depravation because of the diluting of the oxygen content in the car. It can seriously slow down one reaction time, and induce drowsiness. Unless they're planning on mixing oxygen into the canisters, although that might make them somewhat volitile.
Yes, except for the fact that you're going to have to deal with things like condensation and freezing. Plus the fact that compressing the nitrogen in the first place consumes energy that must come from somewhere.
And 4.0 has been plaqued by problems since it's release. It didn't support many of the video cards that 3.3.6 did, had security problems, and wouldn't even build right on Debian. 4.0.1 is a step in the right direction, but I'm still going to treat it with the caution. I'm confident they will get it right eventually, but with any change this drastic, there's bound to be problems.
Another place that Linux is weak on is on desktop systems, due largely to the fact that whereas Windows and Macintosh both have very stringent UI guidelines, Linux has virtually none. KDE and Gnome are trying to change this, although I feel the somewhat hostile environment between the two is working more to make things worse than improve it. As a developer, I should be able to choose either GTK+ or QT and have the resultant program perform similarly on either platform. And as a user, I should be able to expect that any program I install should try to act reasonably with the preferences I have set in my "prefered" desktop environment. For example, I'd love to see a unification of UI "themes", and have a toolkit independent preferences system where both KDE and Gnome settings are stored via their respective control centres, and shared between both where applicable.
Portions of this document are no better than the FUD it attempts to refute, IMO, due to exaggerations on the behalf of the author. There's no harm in admitting weakness, in fact, it usually shows that you're trying to be honest and unbiased.
Despite the hard feelings I can't help but get the feeling that they should try to co-operate a little more. There's no real good reason why KDE and Gnome/GTK can't share things like themes, and general user preferences.. For a short while the teams appeared to co-operate to come up with a unified component system. I firmly believe that there's no valid reason why the UI settings and preferences I configure in either KDE or Gnome can't be shared between both, giving KDE/Qt apps running on a Gnome desktop a similar "feel" to then as the native Gnome/GTK apps, and vice versa.
2. A 486 is more than up for the job. A 486-DX2 running Linux kernel version 2.2.x with ISA NICs will become saturated at about the 3-4Mbit/sec mark. As long as you never see more than that much traffic, you'll be fine.
3. Safety first. I agree that keeping your firewall clean and efficient is very important. However, I find the claims that Linux is less secure than BSD more than a bit bogus. Almost all those server daemons that have had buffer overflows on Linux can be compiled and install into OpenBSD with the same buffer overflows. Security is a journey not a destination is true in ALL cases, even OpenBSD. An incompetent (or inexperienced) administrator can easily turn a secure machine into one that's wide open for anyone to break into.
Most people usually end up compromised because of services that they either never used or never knew about, and therefore didn't bother maintaining. Due to the shortsightedness of most Linux distributors, you'll probably end up "cleaning" dozens of packages out that are completely worthless. Ideally, your result should be a machine that's not listening to anything on the public interface.
4. Raise Hell About Gnome Security Issues. Absolutely! A TCP/IP port should never be opened unless there's a very good reason why this service needs to be advertised to the world. Most of the time, this is just lazy coding, and a place where other types of sockets would probably serve better.
And as for the elitist attitude, it's pretty common on IRC, especially after being pestered by morons who do not care to bother try finding the answer to their question themselves.
The reasons for KDE not being in Debian quite a while ago (the old "free vs free" arguement) and have mostly been resolved for KDE2, but KDE2 is still too unstable to even be in the woody "unstable" branch. They may make it in there yet, but KDE2 needs to ship first. KDE1 will probably never make it in. I'm surprised Helix Gnome hasn't made it into contrib on woody or even potato for that matter.
But honestly, Debian just seems ideal for a server -- unlike the other distros, it installs completely basic (not bare enough, IMO, but better than anyone else), and you add daemons afterwards. The APT system is wonderful for when you want a package installed, but don't want to have to mess around with vague dependancies -- it's what RPM should've been. Fancy GUIs are fine and dandy on a personal workstation, but on a server, they introduce unneeded complexity and potential for error.
Groupwise 5.2 and 5.5 included a MAPI provider so you could use VBS-worm heaven Outlook or any other MAPI client.
Considering the fairly sorry state of 3D acceleration right at this moment, it's not all that surprising. Sure we have XFree86 4.0, but it's been plagued with problems, incompatibilities, etc.
It's almost as if the story was acting upon itself. It covers a shoddy website, yet the story itself was posted poorly. I think the slashdot people should file a class action suit -- NO MORE TYPOS!@#$%&*!
Yeah, but unless I'm mistaken, a trademark is usually associated with a particular product, too. Yet, the IOC went after little companies like Olympic Pizza, etc before to get them to change their name. I guess they fail to realize that the word "olympic" is extremely common and that the world doesn't revolve around them. Pity the poor fool who gets in the way of the money making machine, since even if it wasn't intentional, it's always the "little guy" who gets screwed in the deal.
"What do you mean our workstations won't work with it? I don't care about that -- this article says that IPv6 is where it's at, and I want it now."
"Windows 2000 is faster than NT 4.0!"
Yeah, but only if you run it on a system that's lower than the recommended specs.
I think MS has also proven that NT 4.0 is faster than Win9x is faster than NT 4.0. How, I'm not sure. I guess what they say really is true -- There are three kinds of lies, lies, damn lies, and benchmarks.
Maybe I'm just bitter, but after reading the same lame jokes for the hundredth time, I'm not finding them very funny anymore.
There have been a MULTITUDE of flaws in Outlook and Outlook Express that allowed execution of arbitrary code when looking at a specially contructed (read: MS HTML extensions) HTML message. It's stupid, and most of it revolved around ActiveX.
This is the patent office we're talking about. They wouldn't know software prior art if it walked up and slapped them in the face. Sometimes I think they go out of their way to aprove absurd patents. It'd be nice if software patents weren't legal, really, it would.
My favorite little "bug" in Win9x is the fact you get ONE shot at installing the networking files... If you add a networking device, and click cancel instead of putting in the Win9x CD-ROM, it'll never try to install VREDIR.VXD, VNETSUP.VXD, and a handful of other VXD files ever again. Or the other issue where going into the network control panel in Win95 will cause it to copy the same files usually about 4 times and then tell you to reboot despite not changing anything. And of course, it always wants to overwrite the updated version of SECUR32.DLL with the one on the CD. It's slightly better with Win98, but not much.
I seriously doubt any of these quirks would require substancial programming effort to fix.
Once you learn the keyboard accelerators for Windows, there's very little you can't do in Windows with no mouse (I think the taskbar is one of the few inaccessable things that require mouse use). Some of the shortcuts may be clumsy, but at least they're there. On the other hand, I find myself searching for the mouse constantly in X because some programs and window managers just plain refuse to acknowledge that someone would rather type on the keyboard than touch a mouse. KDE/QT tends to be one of the better toolkits for this, however it's still possible and easy to find a program that won't respond to anything but a mouse.
And despite how many people love to hate Windows and it's UI elements, I have to admit, they're fairly light on screen real-estate. Windows add about 18 pixels at the top of a window, and 2 pixels at each of the sides and bottom. Most of the common controls are distingushable from each other (unlike the motif widgets which mostly all look the same). They may not be the prettiest, but they work.
I agree with the previous poster in that these real-world UIs are aweful, even more so since they almost always ignore the keyboard, and always seem awkward to use. Fighting with circular dial controls, all of which act differently, is not my idea of a good UI design.
Now, don't get me wrong. Skins aren't all bad. But unless they apply to all of the user interface, they're a bad idea except for power users.
I think Netscape 6 would be a lot better if it used the Windows common controls on Windows by default, the MacOS common controls on a Mac, etc. GTK+ is a nice toolkit and all, but when you enforce a UI on users that aren't familiar with it, you're asking for problems...
The largest change between the 1868 and 1869 was a slight revision and the addition of more Windows friendly Plug and Pray, er, Play. The problem with the persistant IDE controller (for cards without the IDE connector) on the 1898 was also resolved on the 1869. Either way, both are extremely low cost, low quality chipsets. And both are full of bugs with their SoundBlaster Pro emulation.
Microsoft HAS done a lot of good for the PC industry, such as the widespread acceptance and use of PCs worldwide, as well as making sure their new OS releases are so unbarably slow that it pushes the industry to make faster and better products so the OS isn't quite so slow anymore.
So, you can laugh all you want, but it doesn't matter.
BTW, I love my 4 button mouse, and I won't use anything with fewer than 3 buttons now.
Personally, I like DVD. The image is clear, you get aditional features with the movie, but currently, I'm just SOL if I want to use my DVD-ROM to play any movies. Of the two OS's I'm currently working with -- Windows 2000 and Debian Linux, neither support DVD decoding at this point. The idea of a binary only player for Linux bugs me somewhat due to the fact that the idea of binary compatibility is completely lost on Linux, where if it's a bug, it's going to be fixed, no matter how much shit is depending on it. In contrast to the Windows world, where Microsoft has to be extremely careful about fixing bugs in their library calls because they could easily break hundreds of programs.
The patent system has a lot of problems when applied to software. For one, they last too long. By the time the patent has expired, it's usually so archiac that it's been replaced by newer software technologies that perhaps didn't have such restrictions. Another problem is that when the patent HAS expired, the patent itself is worthless, since it doesn't provide and subtancial information on the implementation to make the release of "intelectual property" worth it. I read part of the Powerquest partitioning resizeing patent, and gathered absolutely no useful information from it. I gathered more information by actually using Partition Magic a few times, and observing it's behaviour. If patents on software only lasted say 2 or 3 years, and required a sample implementation to be submitted with the patent, perhaps the patent system would be of some use, however since it's not, it'll never benefit the small developers.