IANAL, but acceptance of the GPL is required for more than just distribution; it's also requried for creation of a derivative work, and it could well be argued that their patches are a derivative work of the kernel (especially as they are useless without the kernel).
More importantly if they have ever distributed the kernel to anyone, by any means, then they also have likewise accepted the license.
Did you know that SMB is merely an implementation of the CIFS standard?
Regardless, there's nothing stopping you from implementing your own remote filesystem technology that works just as well across platforms, and making it free. Why isn't there a good open source NT NFS client? Why hasn't someone put together a good WebDAV filesystem for NT? It isn't because it's impossible to make.
The technical braindamages of the.doc format likewise have nothing to do with their business decisions regarding file format compatibility. They could still force people to upgrade without using a badly done format like the doc format. Nonetheless the doc format issue is a bigger straw man yet than anything I said, as the doc format has been fully reverse-engineered and is well implemented by OO.o.
The Outlook/Exchange integration issue is likewise not relevant, as Exchange is probably not widely used in non-corporate settings. Once again, there are fairly widely deployed Notes installations. Why isn't there an open source solution that matches either of these in terms of integration? I'm not sure.
I think the common point here is that the open source community hasn't even tried to address these issues, so calling them examples of unfair competition by Microsoft is at best a hypothesis. There hasn't been anything for them to compete against.
Yes, the consumer is hurt by the lack of choices here. But the open source community hasn't done much to provide them with serious choices on these issues (except in the case of OpenOffice.org, which was a donation by a company).
I didn't say that Apple was righteous. I said that people who compare systems independently should be smart enough to not pay attention to vendor claims like that.
Understood - 4hrs is what I get out of my TiBook - but when comparing to other manufacturers people will look at ACTUAL time of the Apple compared to QUOTED time of the other laptop, as if x86 vendors never lie or exaggurate their specs.
"Dude! My friend's Apple with the five year old battery only gets a half hour, but this Censpleeno gets 7.5 hours! Dude, Apples suck! Mac really needs to go Intel!"
So, in effect, what you're saying is that Microsoft's practices are hurting open source software, and so the open source community is entitled to affirmative action^W^Whelp from the government to succeed. Well, I'm sorry, but I think that's a bullshit position.
I'm not sure in what way open source software is on an unequal ground when competing in the marketplace. It's possible to develop open source software for Windows (witness Mozilla and OpenOffice); it's possible to get Microsoft to agree to participate in specification processes (witness HTML and their efforts to standardize C# via ECMA). And it's not the fact that it's open source that hurts adoption of OSS on Windows platforms (witness Apache on Windows). When OSS is the logical choice, people choose it, just like people choose independent vendors as well (like Sun).
However, what explains the vast number of installations of IIS? Is it because Microsoft forced them to use IIS? Is it because IIS works over MS-IP and only serves MSML? No, it's because IIS is simply easier to set up, for the most part. If Microsoft can compete with Open Source software where it does have an equal footing, then they're obviously not inferior.
The Microsoft infrastructure, which you address as if it's the Matrix, is also very minimal. Microsoft has never been one to support its developers too much. Most of the developers which produce these Windows-only applications (I assume that's what you're referring to; about the only other barrier for most people is Office, which is nicely replaced with OpenOffice) because it's the smart economic decision, and secondly because developing for Linux is hard. That's right, it's hard. Of course, someone'll jump in and say "Oh, but if you use Qt it's easy!", but now you've got a per-head license almost as expensive as that of Visual Studio to pay for; another will say "GTK is easy", but compiling GTK2 programs statically can be difficult, and there's a big lack of documentation. And you still don't get any of the libraries that Microsoft or Apple provide.
But far be it from me to suggest that open-source software might be inferior to MS software in some ways. After all, any such infierority is caused by the fact that we simply can't compete once Microsoft dominates with their evil proprietary ways, right?
The Pentium 4 already does do 36-bit addressing, and the OS could be modified to take advantage of this by setting a 4GB (more realistically, 2GB) virtual address space for each process but allowing the VM to address the entire 16GB range.
This is approximately what Apple is doing for the G5 under Jaguar; user processes will still have a 32-bit virtual address space but the VM will be capable of addressing the full 64-bit (rather, 48-bit external address space) space. What they do under Panther is unknown for now.
Incorrect. The GPL is an end-user license agreement. Also, Mandrake, and RedHat both have a EULA section in the installation.
No, you're incorrect here. No license is required to use software published under the GPL, just as no license is required to use a book purchased from the store. The GPL only extends to users rights not granted to the user by the Copyright Act under certain terms and conditions; namely, the right to copy and modify the software.
Nope. DEC (er, Compaq) pulled out of the project, amazingly enough. MS was perfectly willing to ship Windows 2000 on Alpha, but Compaq suddenly got cold feet.
FWIW I think some internal release candidates of 2K supported Alpha. And given that Win2K is available on Itanic, chances are it's still just as portable.
"Random"? Excuse me, sir, but in title all words but articles, conjunctions, and short prepositions are capatalized. This includes the word "Interview" and "Designers".
Next time make sure you're right before correcting someone, you under-haired excuse for a Wookie.
You idiot. It is perfectly appropriate to abbreviate the "PowerPC 970" to "970". The other reading is syntactically valid, funny, and completely semantically wrong.
When did brain cells become a short supply around here?
Shove it. Comparing ICC on Intel to GCC on IBM is *not* a valid test of CPU capability.
In fact, the only valid test of CPU capability is to optimize the algorithm for the particular processor by hand and then run it - and on those tests, the G5 might very well smoke the P4. I speak from experience - AltiVec is really quite amazing compared to the wretched horror that is SSE/SSE2 and should perform quite nicely on those G5s.
Apple used GCC 3.3 to run the SPECmarks on the Intel processors too, so the results are indeed valid.
A comparison between ICC on the Intel and GCC on the G5 is interesting to determine how fast your code might run out-of-the-box on a P4 vs G5 but ultimately flawed for comparing CPU performance.
The problem is that the economies of scale that apply to the G4 low-power variants don't apply to the Apple G4s. For those Moto has just about two customers, Apple and Cisco - and Cisco seems happy just to use 1GHz parts.
IBM, on the other hand, has already said they will be supplying a stock northbridge and working with south bridge vendors (MAI, Marvell, et all) to supply chipsets for 970-based computers. I'm just wondering how long until a 970-based Amiga is available;-)
Don't be so sure. There's been rumors that the PP970 may indeed be cheaper than the current G4 line. Remember, Moto doesn't sell a lot of 1.42 GHz G4s to its primary customers, but IBM is looking forward to selling a lot of 970s. Mass production on that delicious new East Fishkill fab could make it significantly cheaper than whatever Moto manages to push out.
Have you even looked at the issues? Yes, you can steal BSD. It's not in the public domain - it's under a liberal license, but one that still requires attribution of use. If you break the terms of that contract, then you are indeed stealing it, and that's exactly what AT&T did to UC over the decades by not attributing their copyright in UNIX.
That's why UC won the AT&T / USL vs UC suit of the early 1990s - so much of BSD had been put into AT&T illegally that UC would have had a heckuva lawsuit against AT&T should they have chosen. Instead, AT&T let them rewrite three files and continue on their merry way.
That's a good idea! Fred Brooks should patent the idea of the over-budget over-time software project, and then sue SCO on the basis of their filed claims that UNIX has been under development for 20 years.
In effect, that and the intense pressure on SPS to perform up to profit numbers and reduce headcount combined to do just that. When Apple was stuck with Moto as an exclusive supplier (which the introduction of AltiVec effectively did), justifying cost spent on improving performance for a customer who will order the same number of chips anyway is hard to do.
As far as the shared / separate FSB issues, I'm not too familiar with this myself. Does the cache coherency protocol used by the G4 assume that the second processor is connected to the same FSB?
Slashdot Mirror
It's actually fellatious in that the more they push that license the more I want to shove my dick in their mouths.
That doesn't sound like such a wise idea. I'd bet SCO bites.
IANAL, but acceptance of the GPL is required for more than just distribution; it's also requried for creation of a derivative work, and it could well be argued that their patches are a derivative work of the kernel (especially as they are useless without the kernel).
More importantly if they have ever distributed the kernel to anyone, by any means, then they also have likewise accepted the license.
That claim is awfully hard to get out of.
Did you know that SMB is merely an implementation of the CIFS standard?
.doc format likewise have nothing to do with their business decisions regarding file format compatibility. They could still force people to upgrade without using a badly done format like the doc format. Nonetheless the doc format issue is a bigger straw man yet than anything I said, as the doc format has been fully reverse-engineered and is well implemented by OO.o.
Regardless, there's nothing stopping you from implementing your own remote filesystem technology that works just as well across platforms, and making it free. Why isn't there a good open source NT NFS client? Why hasn't someone put together a good WebDAV filesystem for NT? It isn't because it's impossible to make.
The technical braindamages of the
The Outlook/Exchange integration issue is likewise not relevant, as Exchange is probably not widely used in non-corporate settings. Once again, there are fairly widely deployed Notes installations. Why isn't there an open source solution that matches either of these in terms of integration? I'm not sure.
I think the common point here is that the open source community hasn't even tried to address these issues, so calling them examples of unfair competition by Microsoft is at best a hypothesis. There hasn't been anything for them to compete against.
Yes, the consumer is hurt by the lack of choices here. But the open source community hasn't done much to provide them with serious choices on these issues (except in the case of OpenOffice.org, which was a donation by a company).
I didn't say that Apple was righteous. I said that people who compare systems independently should be smart enough to not pay attention to vendor claims like that.
Understood - 4hrs is what I get out of my TiBook - but when comparing to other manufacturers people will look at ACTUAL time of the Apple compared to QUOTED time of the other laptop, as if x86 vendors never lie or exaggurate their specs.
"Dude! My friend's Apple with the five year old battery only gets a half hour, but this Censpleeno gets 7.5 hours! Dude, Apples suck! Mac really needs to go Intel!"
OK, let me spell it out for you:
Apple produces mid-range and high-end products. They do not produce any low-end products.
iBooks are mid-range laptops and compete quite well with comparable PC laptops.
PowerBooks are high-end laptops and blow away most other high-end PC laptops.
Apple also does not produce luggables, almost-PDA-sized laptops, laptops without CDs, laptops with less than 4.5 hours of battery life, or tablets.
If you stick to laptops you will find that Apple's laptops are highly competative.
Was that enough?
So, in effect, what you're saying is that Microsoft's practices are hurting open source software, and so the open source community is entitled to affirmative action^W^Whelp from the government to succeed. Well, I'm sorry, but I think that's a bullshit position.
I'm not sure in what way open source software is on an unequal ground when competing in the marketplace. It's possible to develop open source software for Windows (witness Mozilla and OpenOffice); it's possible to get Microsoft to agree to participate in specification processes (witness HTML and their efforts to standardize C# via ECMA). And it's not the fact that it's open source that hurts adoption of OSS on Windows platforms (witness Apache on Windows). When OSS is the logical choice, people choose it, just like people choose independent vendors as well (like Sun).
However, what explains the vast number of installations of IIS? Is it because Microsoft forced them to use IIS? Is it because IIS works over MS-IP and only serves MSML? No, it's because IIS is simply easier to set up, for the most part. If Microsoft can compete with Open Source software where it does have an equal footing, then they're obviously not inferior.
The Microsoft infrastructure, which you address as if it's the Matrix, is also very minimal. Microsoft has never been one to support its developers too much. Most of the developers which produce these Windows-only applications (I assume that's what you're referring to; about the only other barrier for most people is Office, which is nicely replaced with OpenOffice) because it's the smart economic decision, and secondly because developing for Linux is hard. That's right, it's hard. Of course, someone'll jump in and say "Oh, but if you use Qt it's easy!", but now you've got a per-head license almost as expensive as that of Visual Studio to pay for; another will say "GTK is easy", but compiling GTK2 programs statically can be difficult, and there's a big lack of documentation. And you still don't get any of the libraries that Microsoft or Apple provide.
But far be it from me to suggest that open-source software might be inferior to MS software in some ways. After all, any such infierority is caused by the fact that we simply can't compete once Microsoft dominates with their evil proprietary ways, right?
Impossible. 4.4 BSD/Lite 2 has those, and is completely free of SCO code.
The Pentium 4 already does do 36-bit addressing, and the OS could be modified to take advantage of this by setting a 4GB (more realistically, 2GB) virtual address space for each process but allowing the VM to address the entire 16GB range. This is approximately what Apple is doing for the G5 under Jaguar; user processes will still have a 32-bit virtual address space but the VM will be capable of addressing the full 64-bit (rather, 48-bit external address space) space. What they do under Panther is unknown for now.
Incorrect. The GPL is an end-user license agreement. Also, Mandrake, and RedHat both have a EULA section in the installation. No, you're incorrect here. No license is required to use software published under the GPL, just as no license is required to use a book purchased from the store. The GPL only extends to users rights not granted to the user by the Copyright Act under certain terms and conditions; namely, the right to copy and modify the software.
Oh, and that system controller is all soft and squishy?
Nope. DEC (er, Compaq) pulled out of the project, amazingly enough. MS was perfectly willing to ship Windows 2000 on Alpha, but Compaq suddenly got cold feet. FWIW I think some internal release candidates of 2K supported Alpha. And given that Win2K is available on Itanic, chances are it's still just as portable.
"Random"? Excuse me, sir, but in title all words but articles, conjunctions, and short prepositions are capatalized. This includes the word "Interview" and "Designers". Next time make sure you're right before correcting someone, you under-haired excuse for a Wookie.
You idiot. It is perfectly appropriate to abbreviate the "PowerPC 970" to "970". The other reading is syntactically valid, funny, and completely semantically wrong. When did brain cells become a short supply around here?
Join the Army, they said.
See the world, they said.
I'd rather be sailing.
Shove it. Comparing ICC on Intel to GCC on IBM is *not* a valid test of CPU capability.
In fact, the only valid test of CPU capability is to optimize the algorithm for the particular processor by hand and then run it - and on those tests, the G5 might very well smoke the P4. I speak from experience - AltiVec is really quite amazing compared to the wretched horror that is SSE/SSE2 and should perform quite nicely on those G5s.
Apple used GCC 3.3 to run the SPECmarks on the Intel processors too, so the results are indeed valid.
A comparison between ICC on the Intel and GCC on the G5 is interesting to determine how fast your code might run out-of-the-box on a P4 vs G5 but ultimately flawed for comparing CPU performance.
The problem is that the economies of scale that apply to the G4 low-power variants don't apply to the Apple G4s. For those Moto has just about two customers, Apple and Cisco - and Cisco seems happy just to use 1GHz parts. IBM, on the other hand, has already said they will be supplying a stock northbridge and working with south bridge vendors (MAI, Marvell, et all) to supply chipsets for 970-based computers. I'm just wondering how long until a 970-based Amiga is available ;-)
Don't be so sure. There's been rumors that the PP970 may indeed be cheaper than the current G4 line. Remember, Moto doesn't sell a lot of 1.42 GHz G4s to its primary customers, but IBM is looking forward to selling a lot of 970s. Mass production on that delicious new East Fishkill fab could make it significantly cheaper than whatever Moto manages to push out.
Last I checked, the local Apple store had precisely 0 shelves of PC software. Must not be any PC software on the market!
That's why UC won the AT&T / USL vs UC suit of the early 1990s - so much of BSD had been put into AT&T illegally that UC would have had a heckuva lawsuit against AT&T should they have chosen. Instead, AT&T let them rewrite three files and continue on their merry way.
Hey, I can confirm that. So can you. Go crack open an iBook.
Go download iPulse from the Iconfactory. It uses a totally vector-based and non-rectangular UI. It's quite pretty, too.
That's a good idea! Fred Brooks should patent the idea of the over-budget over-time software project, and then sue SCO on the basis of their filed claims that UNIX has been under development for 20 years.
In effect, that and the intense pressure on SPS to perform up to profit numbers and reduce headcount combined to do just that. When Apple was stuck with Moto as an exclusive supplier (which the introduction of AltiVec effectively did), justifying cost spent on improving performance for a customer who will order the same number of chips anyway is hard to do.
As far as the shared / separate FSB issues, I'm not too familiar with this myself. Does the cache coherency protocol used by the G4 assume that the second processor is connected to the same FSB?