They're using an extremely broad definition of "derived". From that interview it's finally clear what they're trying to claim.
They're saying that they have rights to any technology that any Unix company ever added to Unix. So the JFS, for example, which was added by IBM to their Unix derivative, can't be added by IBM to any other software (including OS/2 I suppose, which is where the Linux version actually came from).
I really doubt that IBM was stupid enough to sign something that broad. In fact, it would be far more viral than the GPL. If I incorporate my proprietary code into GPLd software, I can still retain copyright to the code and continue to use it in my own projects. Apparently not so with SysV code.
From what I could figure out before it was Slashdotted, the advantage of legs is mobility. RHex , for example, can run over obsticles that are the same height as it.
Wheels are limited to obsticles smaller then the radius at low speeds, and considerably smaller than than that at high speeds. No idea what the actual formula is though - anybody know?
I'm not sure about numbers, but it's certainly popular with me. I've been waiting for months for XD2.
Red Carpet has been unable to download the packages for the last few hours, so I guess there are enough people like me to swamp their servers.
Just because your crowd doesn't use something, doesn't mean it's not popular. I don't know anybody who uses a Mac, much less Apple's new music thingy, but apparently that's popular too. Go figure.
Your comment illustrates the nature of the problem. In the Open Source world the creation of the software separate from the support.
The Apache guys rightly consider their job done once they've put the patch on their web site. It's up to the distributor, or whoever else wants to make money from support, to deal with it from here.
Proprietary software writers, however, want to support the products themselves. That's good sometimes, but it means the end-user has to deal with each software vendor separately in order to manage patches.
I think the problem is that people _do_ touch-type, but they're thinking "cut", not "control-x" or looking down at the keyboard. I certainly do that.
Similarly, if you ask me what the movement keys are in vi (or Nethack - where I really learned this skill many years ago) I'd have to think really hard to remember. But put me in front of a keyboard and it's automatic, at least with Qwerty.
Yeah, but their competitors before were either small companies (Netscape) playing on MS turf (the desktop) or too dense to figure out they were being screwed until it was too late (IBM).
Problem for MS is that there aren't a lot of those companies left. The small guys stay out of the way, or are already out of business, and the big guys don't trust MS.
Of the three big pushes MS is putting on right now:
Smartphones: very strong, very popular incumbant - MS failing badly.
Consoles: very strong, very popular incumbant - MS throwing enough money at it to put in a good showing, but still not even close to winning
Online access: large but unstable incumbant, MS doing well by some measures, but everybody (including MS) getting destroyed by telcos and cable companies that are taking over the market.
I suppose you could add to that the server OS market: MS looked like it was going to take over, and had lots of momentum, but the old guard (Unix, OS/390) held out long enough for a different kind of competitor (Linux) to start pushing back. It's unlikely that MS will grow their server market share any further, and it looks like they're headed for a gradual decline.
Yeah, I thought that was interesting too. Given the limited availability of SCO code, I can think of two explanations that are much more likely:
both SCO and Linux developers studied the BSD code
the SCO code was (perhaps accidentally) stolen from Linux
Anyway, if I had discovered that my proprietary software product contained GPL code, I'd discretely try to replace my code, since I'd have much more to lose from legal action. If somebody can prove that SCO stole GPL code, it might be possible to force them to GPL their entire kernel.
I think there a lot of people (like me, for example) who aren't really into games, but wouldn't mind being able to do some laptop racing on a long flight. FreeCiv used to fill that role for me, but I'm bored with it. Something like Midnight Racer could keep me entertained for a while.
Depends on your machine I think. If Moz feels fast for you then Phoenix probably won't make much different.
I do perceive a difference in speed, but it's hard to tell why. I doubt the actual page-rendering is faster, but the cleaner XUL code for the interface renders faster. Even if it's just a few milliseconds, it makes a big difference in the feel of the application.
Put WinXP, Gnome, KDE, and Win95 next to each other and click around a little. (Make sure you don't set up some wierdo theme - just use the defaults.)
WinXP is less like Win95 than either Gnome or KDE. You could just as easily argue that the retraining costs for XP would be greater than for Linux because MS gratuitously messed with the user interface.
As for interoperability - it's pretty straightforward and you only have to do it once. After that you duplicate the configuration on the rest of the machines.
The GPL was the license that gave them the right to distribute Linux and all the other GPL code in their distribution. So yes, they are now subject to the terms of the GPL.
That's great for you (although I dispute your claim that Linux doesn't have top-notch support). But the point for software vendors is that they can't afford not to support Linux. Certainly the high-end is a very profitable niche, but you can't support a large company on a niche market.
Back to my original statement: every single new Unix product comes with Linux support. Every major old Unix product now supports Linux.
A year ago application support was a major reason for not using Linux. That's just not an issue anymore, and in fact it's turning the other way - if you deploy Linux you'll have a greater choice of applications than with AIX or HP. With Solaris it's probably a toss-up.
There are still some barriers, and scalability is one. It's not an issue for most applications today, but there are a few that require big boxes. By this summer Linux will scale nicely to 16 processors. Two years after that it will outperform AIX on Regatta-class machines.
When that point comes, if Sun hasn't made Linux scale on their high-end machines, Sun will start losing their big-iron accounts.
Ok. I'll grant you that. I didn't know they were funding the NFS V4 stuff.
I still wouldn't count this as a major thing though. It's important from a Linux perspective because Linux NFS has traditionally sucked badly, but's it's hardly a big effort from Sun - three developers, jointly funded by Network Appliance.
It scales enough. Big servers are cool, but it's not where most of the market is. Take a look around at your major software vendors. Search for "Linux" on their web sites. A year ago you might not have found much, but times have changed.
And stuff on Freshmeat counts more than you think. Most companies use a mix of free and proprietary software. The fact that I can find a quick and dirty network monitoring tool on Freshmeat and know it will work on our Linux boxen is a big deal.
The point remains that they contribute to the technology around operating system, not the OS itself. They work on stuff that would benefit Solaris as well.
I don't think I've ever seen an e-mail from anybody at Sun on linux-kernel. There might be some contributions that I haven't noticed, but I doubt it's anything serious.
I think it's impossible to say for sure. But just because some exec. gives you the party line, doesn't mean that the people who really run the show don't know what's going on.
Of course, there's no guarantee that they do know what's going on either - certainly SCO is just as clueless as they appear.
Remember IBM in the mid-90's? - the people we knew there were pretty clueless, but Armonk managed to turn it around. I never would have believed it was possible.
Sun has some bright people too. They realize that they've got a problem, and you can see them trying to execute some strategy with their low-end offerings. I'm not sure what the strategy is exactly, but there's something going on. We'll have to see how it pans out.
OS/2 still alive in various areas, but it's not bringing IBM any new business.
Same with Solaris in five years or so. I'm sure it will exist as a product for at least a decade, but within five years there won't be any reason for somebody to use it in a new project.
Sure there are some good reasons to go with Solaris today for certain applications, but they're becoming less all the time.
When I started with Unix SunOS was king, and all the applications you got anywhere always worked on SunOS. If you were lucky there might be a port to your alternative Unix (AIX, 386BSD, Irix), but you could always be confident that there would be support for Sun.
Now things have turned the other way. Every new Unix application is available for Linux. The old ones that still matter are being ported rapidly. It's getting so that Linux is the only "no-brainer" deployment. Everything else requires thought. Is AIX supported? Is Solaris supported? Who knows - just use Linux because you know it will work.
Sun's big contribution to Linux is OpenOffice. Their efforts on Linux proper have been pretty limited anyway.
Honestly, though, I don't think will effect their Linux strategy either. It's just a short-term marketing/PR stunt.
Despite what they say, I really doubt that Sun thinks they can keep people on Solaris long-term. They're just not that dumb. More likely they're trying to keep customers from defecting for a few years while they work on improving the upper layers of their environment (Java, SunONE). Then they can switch the bottom layer to Linux but keep some proprietary advantages.
The cell networks in Finland are _very_ good. You can get signal almost anywhere. Even the metro stops have signal (although that doesn't always extend the entire distance between the stops).
But there's no feeling, as near as I can tell, that you have to respond to a phone call if it's inconvenient. And really you don't make a voice call unless you think the other person wants to talk to you. SMS is much more discreet so you can send it any time - even if you think the other person might be sleeping.
I actually never bothered to get a land line when I moved here. I don't have much use for it. None of the 20-somethings I know have land lines in their homes.
It's not so much that society revolves around the device, it's that people assume that you'll be reachable regardless of location.
So if I'm meeting somebody downtown, I won't necessarily specify an exact time or place. Just before I leave I'll send a text message giving a time estimate, and then when I arrive I'll send another to say where I can be found (or to determine where I can find my friends).
If somebody didn't have a mobile phone then I probably wouldn't see them as often. It would just be too much hassle to plan ahead.
Of course, that's very much hypothetical. I've lived in Finland for a year and a half and I've never met an adult Finn without a cell-phone.
1) Linux doesn't scale to large SMP systems yet. I think 2.6 is supposed to make it nicely to 16 processors.
2) Recently most (all?) of the big Unix vendors have included mainframe-style partitioning. You can do that with Linux on IBM zSeries and pSeries (and maybe iSeries), but you need another OS acting as the executive.
I can't think of anything else off-hand. I'd say that for the vast majority of applications Linux is as good or better than commerical Unix.
Don't know - it's really hard to guestimate when you're talking just a few percentage points. I don't actually know anybody personally who uses a Mac, but I know about a dozen who use Linux almost exclusively.
Granted, these are at least semi-Geeks, but I would think the geek population is a lot more than 1%.
If you work in graphic arts you might have the opposite impression, but again it's not a representative sample. What percentage of the population works in the graphics business?
Slashdot Mirror
They're using an extremely broad definition of "derived". From that interview it's finally clear what they're trying to claim.
They're saying that they have rights to any technology that any Unix company ever added to Unix. So the JFS, for example, which was added by IBM to their Unix derivative, can't be added by IBM to any other software (including OS/2 I suppose, which is where the Linux version actually came from).
I really doubt that IBM was stupid enough to sign something that broad. In fact, it would be far more viral than the GPL. If I incorporate my proprietary code into GPLd software, I can still retain copyright to the code and continue to use it in my own projects. Apparently not so with SysV code.
From what I could figure out before it was Slashdotted, the advantage of legs is mobility. RHex , for example, can run over obsticles that are the same height as it.
Wheels are limited to obsticles smaller then the radius at low speeds, and considerably smaller than than that at high speeds. No idea what the actual formula is though - anybody know?
I'm not sure about numbers, but it's certainly popular with me. I've been waiting for months for XD2.
Red Carpet has been unable to download the packages for the last few hours, so I guess there are enough people like me to swamp their servers.
Just because your crowd doesn't use something, doesn't mean it's not popular. I don't know anybody who uses a Mac, much less Apple's new music thingy, but apparently that's popular too. Go figure.
Your comment illustrates the nature of the problem. In the Open Source world the creation of the software separate from the support.
The Apache guys rightly consider their job done once they've put the patch on their web site. It's up to the distributor, or whoever else wants to make money from support, to deal with it from here.
Proprietary software writers, however, want to support the products themselves. That's good sometimes, but it means the end-user has to deal with each software vendor separately in order to manage patches.
I think the problem is that people _do_ touch-type, but they're thinking "cut", not "control-x" or looking down at the keyboard. I certainly do that.
Similarly, if you ask me what the movement keys are in vi (or Nethack - where I really learned this skill many years ago) I'd have to think really hard to remember. But put me in front of a keyboard and it's automatic, at least with Qwerty.
Yeah, but their competitors before were either small companies (Netscape) playing on MS turf (the desktop) or too dense to figure out they were being screwed until it was too late (IBM).
Problem for MS is that there aren't a lot of those companies left. The small guys stay out of the way, or are already out of business, and the big guys don't trust MS.
Of the three big pushes MS is putting on right now:
I suppose you could add to that the server OS market: MS looked like it was going to take over, and had lots of momentum, but the old guard (Unix, OS/390) held out long enough for a different kind of competitor (Linux) to start pushing back. It's unlikely that MS will grow their server market share any further, and it looks like they're headed for a gradual decline.
Yeah, I thought that was interesting too. Given the limited availability of SCO code, I can think of two explanations that are much more likely:
Anyway, if I had discovered that my proprietary software product contained GPL code, I'd discretely try to replace my code, since I'd have much more to lose from legal action. If somebody can prove that SCO stole GPL code, it might be possible to force them to GPL their entire kernel.
You have to start somewhere.
I think there a lot of people (like me, for example) who aren't really into games, but wouldn't mind being able to do some laptop racing on a long flight. FreeCiv used to fill that role for me, but I'm bored with it. Something like Midnight Racer could keep me entertained for a while.
Depends on your machine I think. If Moz feels fast for you then Phoenix probably won't make much different.
I do perceive a difference in speed, but it's hard to tell why. I doubt the actual page-rendering is faster, but the cleaner XUL code for the interface renders faster. Even if it's just a few milliseconds, it makes a big difference in the feel of the application.
Put WinXP, Gnome, KDE, and Win95 next to each other and click around a little. (Make sure you don't set up some wierdo theme - just use the defaults.)
WinXP is less like Win95 than either Gnome or KDE. You could just as easily argue that the retraining costs for XP would be greater than for Linux because MS gratuitously messed with the user interface.
As for interoperability - it's pretty straightforward and you only have to do it once. After that you duplicate the configuration on the rest of the machines.
They distributed the code.
The GPL was the license that gave them the right to distribute Linux and all the other GPL code in their distribution. So yes, they are now subject to the terms of the GPL.
I was thinking this too. Or possibly a PS2+: same software, same performance, smaller form factor with built-in broadband, lower manufacturing costs.
That's great for you (although I dispute your claim that Linux doesn't have top-notch support). But the point for software vendors is that they can't afford not to support Linux. Certainly the high-end is a very profitable niche, but you can't support a large company on a niche market.
Back to my original statement: every single new Unix product comes with Linux support. Every major old Unix product now supports Linux.
A year ago application support was a major reason for not using Linux. That's just not an issue anymore, and in fact it's turning the other way - if you deploy Linux you'll have a greater choice of applications than with AIX or HP. With Solaris it's probably a toss-up.
There are still some barriers, and scalability is one. It's not an issue for most applications today, but there are a few that require big boxes. By this summer Linux will scale nicely to 16 processors. Two years after that it will outperform AIX on Regatta-class machines.
When that point comes, if Sun hasn't made Linux scale on their high-end machines, Sun will start losing their big-iron accounts.
Ok. I'll grant you that. I didn't know they were funding the NFS V4 stuff.
I still wouldn't count this as a major thing though. It's important from a Linux perspective because Linux NFS has traditionally sucked badly, but's it's hardly a big effort from Sun - three developers, jointly funded by Network Appliance.
It scales enough. Big servers are cool, but it's not where most of the market is. Take a look around at your major software vendors. Search for "Linux" on their web sites. A year ago you might not have found much, but times have changed.
And stuff on Freshmeat counts more than you think. Most companies use a mix of free and proprietary software. The fact that I can find a quick and dirty network monitoring tool on Freshmeat and know it will work on our Linux boxen is a big deal.
Dunno. Linux and *BSD have a huge political advantage: big companies can support them without feeling they're supporting the competition.
Also, they'd have the same problem that Netscape had. The code-base is so huge that it would take years for developers to really understand it.
The point remains that they contribute to the technology around operating system, not the OS itself. They work on stuff that would benefit Solaris as well.
I don't think I've ever seen an e-mail from anybody at Sun on linux-kernel. There might be some contributions that I haven't noticed, but I doubt it's anything serious.
I think it's impossible to say for sure. But just because some exec. gives you the party line, doesn't mean that the people who really run the show don't know what's going on.
Of course, there's no guarantee that they do know what's going on either - certainly SCO is just as clueless as they appear.
Remember IBM in the mid-90's? - the people we knew there were pretty clueless, but Armonk managed to turn it around. I never would have believed it was possible.
Sun has some bright people too. They realize that they've got a problem, and you can see them trying to execute some strategy with their low-end offerings. I'm not sure what the strategy is exactly, but there's something going on. We'll have to see how it pans out.
OS/2 still alive in various areas, but it's not bringing IBM any new business.
Same with Solaris in five years or so. I'm sure it will exist as a product for at least a decade, but within five years there won't be any reason for somebody to use it in a new project.
Sure there are some good reasons to go with Solaris today for certain applications, but they're becoming less all the time.
When I started with Unix SunOS was king, and all the applications you got anywhere always worked on SunOS. If you were lucky there might be a port to your alternative Unix (AIX, 386BSD, Irix), but you could always be confident that there would be support for Sun.
Now things have turned the other way. Every new Unix application is available for Linux. The old ones that still matter are being ported rapidly. It's getting so that Linux is the only "no-brainer" deployment. Everything else requires thought. Is AIX supported? Is Solaris supported? Who knows - just use Linux because you know it will work.
Sun's big contribution to Linux is OpenOffice. Their efforts on Linux proper have been pretty limited anyway.
Honestly, though, I don't think will effect their Linux strategy either. It's just a short-term marketing/PR stunt.
Despite what they say, I really doubt that Sun thinks they can keep people on Solaris long-term. They're just not that dumb. More likely they're trying to keep customers from defecting for a few years while they work on improving the upper layers of their environment (Java, SunONE). Then they can switch the bottom layer to Linux but keep some proprietary advantages.
The cell networks in Finland are _very_ good. You can get signal almost anywhere. Even the metro stops have signal (although that doesn't always extend the entire distance between the stops).
But there's no feeling, as near as I can tell, that you have to respond to a phone call if it's inconvenient. And really you don't make a voice call unless you think the other person wants to talk to you. SMS is much more discreet so you can send it any time - even if you think the other person might be sleeping.
I actually never bothered to get a land line when I moved here. I don't have much use for it. None of the 20-somethings I know have land lines in their homes.
It's not so much that society revolves around the device, it's that people assume that you'll be reachable regardless of location.
So if I'm meeting somebody downtown, I won't necessarily specify an exact time or place. Just before I leave I'll send a text message giving a time estimate, and then when I arrive I'll send another to say where I can be found (or to determine where I can find my friends).
If somebody didn't have a mobile phone then I probably wouldn't see them as often. It would just be too much hassle to plan ahead.
Of course, that's very much hypothetical. I've lived in Finland for a year and a half and I've never met an adult Finn without a cell-phone.
It'll work, but there's not much point. The locking overhead kills any extra performance you might expect from adding more processors.
I'm not really sure where the performance tops out, but the consensus seems to be that 2.4.x doesn't scale well beyond 4 processors.
1) Linux doesn't scale to large SMP systems yet. I think 2.6 is supposed to make it nicely to 16 processors.
2) Recently most (all?) of the big Unix vendors have included mainframe-style partitioning. You can do that with Linux on IBM zSeries and pSeries (and maybe iSeries), but you need another OS acting as the executive.
I can't think of anything else off-hand. I'd say that for the vast majority of applications Linux is as good or better than commerical Unix.
Don't know - it's really hard to guestimate when you're talking just a few percentage points. I don't actually know anybody personally who uses a Mac, but I know about a dozen who use Linux almost exclusively.
Granted, these are at least semi-Geeks, but I would think the geek population is a lot more than 1%.
If you work in graphic arts you might have the opposite impression, but again it's not a representative sample. What percentage of the population works in the graphics business?