Please show me any publications that demonstrate significantly better usability of Apple products relative to Linux or Windows.
What, I have to prove my subjective experience? Not to mention that you're losing the plot here... first it was "Apple's attacking FOSS". I challenged that, so you claimed that Apple was specifically attacking the usability and performance of FOSS. Now you're saying that I need to prove that Apple's software is more usable than Linux and Windows. Last time I checked, Windows wasn't FOSS. Let's get back to the original point here, how is Apple claiming something that even many Linux boosters agree with "an attack on FOSS"?
I meant what I said: if IBM has wronged you, then speak out about it, nothing more and nothing less.
It's kind of hard to tell just what you mean, you're changeable as the wind and as hard to pin down.
Se let's get back on track. first it was "Apple's attacking FOSS". I pointed out that Apple's released an awful lot of OSX as open source software, far beyond what they need to by even the most generous interpretation of any of the licenses involved. You complained that you don't happen to use any of it. My response about IBM was not an attack on IBM, it was simply astonishment that you'd so moved the goalposts. If a company doesn't release anything YOU don't PERSONALLY use, they're attacking FOSS? Is that what you mean?
I never claimed that Apple is "opposed" to FOSS
What does "Bad licensing choices in the *BSD operating systems have supported both Apple and Microsoft in their efforts to hurt FOSS." mean then?
"Easily misinterpreted marketroid" comments is exactly what the problem is
Ah, gotcha. I'm sure you'll be right there on the attack next time Eben Moglen or RMS or ESR or Theo run off some easily misinterpreted marketroid comment. No? Just Theo?
Basically, the bottom line is this... everyone talks up their own stuff, and tells us how it's better than the competition. Apple does it, the FSF does it, Gentoo and OpenBSD and Ubuntu and Mozilla and Intel and IBM and me and you too. If you don't, you get buried. Pushing your stuff isn't "an effort to hurt FOSS" any more than the KDE vs Gnome rivalry is "an effort to hurt FOSS".
Companies that want to hurt FOSS don't use much, either, because once you split from the community you have to duplicate the efforts of the community... which is why Microsoft replaced the BSD TCP stack and has barely updated the BSD userland utilities they still use in years. What Microsoft's got from BSD is pretty much limited to Interix, and they're using GPL code in there as well.
Companies that benefit the most from FOSS are the ones that stay part of the community. The BSD license gives companies a taste. Some of them don't take the bait, some do. When they do, FOSS wins. When they don't, they don't get much out of it in the long run, so who cares? In the long run everyone wins. IBM took the bait, Apple took the bait, Microsoft nibbled and swam away. That's way better than I would have expected when I started getting excited by open source (or what would be tagged 'open source' 20 years later ) and open systems.
Setting the broadcast bit should not break your DHCP server.
That's also true. I'm not defending the DHCP servers in question, though I would like a bit more details about what the actual bug there involves, and why they didn't respond. Alas, googling brings up more smoke than light.
My point is that Microsoft is not blameless here. If you know anything more about their decision I'd really like to know what led them to start setting this flag now, when it's been discouraged since 1993* in booth DHCP and BOOTP, and since their TCP stack doesn't actually seem to need it. This is unfortunately symptomatic of Microsoft's general approach to standards.
* And its interesting to note that it's only required for Infiniband because of the size of Infiniband addresses... NOT because of the reason the flag was originally created.
I wouldn't call libertarianism a meta-philosophy. It's just another spectrum, the libertarian-authoritarian spectrum, separate from the left-right economic spectrum.
That's another way to describe it, but most libertarians I know don't seem terribly opposed to authoritarianism in general, just specifically against a strong and centralized government. Apart from the left libertarians most of them seem to think private sector authoritarianism is just fine, and many seem to think it would just vanish away when the goad of Big Government evaporates (the same kind of wishful thinking typical of classic Marxists and many Anarchists).
I do like the multi-spectrum meta-philosophy, though, and use it quite a lot myself.
I wrote: "I can't abide it as a server, of course, but servers aren't really Apple's market."
I'd like to clarify this, since I didn't notice until after I posted that the antecedent of "it" is uncertain. I'm referring to OS X here, not Linux. Linux is a perfectly competent server OS, and I've used it in that role in the past when appropriate. I (obviously) prefer BSD (FreeBSD to be specific, though I did give OpenBSD a trial), but that's a preference... not a litmus test.
Like when Apple employees make unsubstantiated claims about OS X usability and performance relative to other systems, misrepresent FOSS inventions as their own, and misrepresent OS X as a "better Linux" and a superset of Linux when selling to universities and companies.
As for the first, I hardly think anyone in the open source community, myself not excluded, have any standing when it comes to comparing the usability of their products to Apple's. Performance? Apple's found themselves caught boasting about their hardware often enough but I hardly think that was targeted at FOSS particularly - their campaigns seem generally to target Microsoft and (until the last few years) intel. Perhaps a citation or two would jog my memory.
I'm unsure what you're referring to with "represent FOSS inventions as their own", again a citation might jog my memory, it should be easy for you to find a dozen if they're particularly targeting FOSS with this kind of thing, and not Microsoft.
As for OS X being "a better Linux", you should be proud that Apple thinks Linux is the best desktop UNIX after OS X. Of course I'd have said "a better UNIX" myself, which comes to the same thing, since Linux is UNIX and it really is hard to argue that Linux is anywhere close to OS X on the desktop. I can't abide it as a server, of course, but servers aren't really Apple's market.
In any event, this hardly counts as a "campaign against FOSS". If Apple was engaged in such a campaign it wouldn't be easily misinterpreted marketroid comments. Not from the same source as the "I'm a PC" ads, it wouldn't.
Of course if someone in the Linux community were to make a "formal mistake" in public comments or a code release I'm sure you'd be just as quick to chastise them.:)
If IBM has wronged you, speak out about it.
If you take my comment as evidence that I have a grudge against IBM, then it's no wonder that you're seeing conspiracies elsewhere. I'm a happy former customer of IBM, and I'm grateful for their excellent laptops (there's no laptop I'd rather use than the Thinkpad) and hard drives (still available under the Hitachi label). I'm also grateful to IBM for their work in user interfaces, glad that Microsoft picked up on that, and a bit upset that Microsoft abandoned IBM's UI standards in Windows 9x and that Apple never picked up on them.
Again, the fact that I'm personally not using Eclipse or SOAP simply means that I'm not personally using Eclipse or Soap. I'm not the one who is mistaking "I don't use X" for "X is worthless", or "I'm not benefitting from Y's actions" for "Y's actions don't matter".
Come on, be specific, rather than telling me why I should like Apple.
I'm a former Mac detractor and current Mac user who thinks Apple's hardware is quite horrid, but willing to put up with it for the sake of the software. I also continue to use other platforms (both open source and proprietary) where they're best suited. I don't currently have any Linux boxes in active use, but that's just the chance of the moment... I've used it in the past and I'm sure I'll be using it in the future.
As for why you should like Apple, well, I've no particular reason to care if you like them or not, I'm mostly disturbed by the fact that you're upset with them for such a bizarre and unlikely reason as their being opposed to free and open source software. For all their faults (and there are many) that's perhaps the least convincing one I could imagine.
If you were to complain about their snobbishness and elitism, or about OS X not running on generic hardware, or about their appalling 'style before everything' design and passive-aggressive attachment to one-button mice, or about the technical flaws in HFS+, or the unnecessary overhead in Quartz or MachO binaries, or anything they're actually guilty of, I'd agree... and if there were any alternative that did as good a job I'd be all over it by now. But hostility towards open source? It ain't there.
First, nerds don't tend to be Libertarians all that much. It just seems that way because pretty much all Libertarians are nerds. But then, pretty much all of the theoretically-based extreme positions are dominated by nerds, because pretty much all theoretically based ANYTHING is dominated by nerds. The kind of obsessive attention to academic details is kind of a defining feature of nerds.
Second, Libertarianism is kind of a meta-philosophy. There's left-libertarians as well as right-libertarians, though the rights tend to dominate because the heartland of libertarianism is America and even a lot of self-identified "left wing" groups in America would be classified as right-wing elsewhere in the world.
MS is properly implementing a known, albeit outdated, DHCP option.
Only if it's connecting over infiniband or firewire. Over ethernet it SHOULD set that bit to zero.
A client that cannot receive unicast IP datagrams until its protocol software has been configured with an IP address SHOULD set the BROADCAST bit in the 'flags' field to 1 in any DHCPDISCOVER or DHCPREQUEST messages that client sends. The BROADCAST bit will provide a hint to the DHCP server and BOOTP relay agent to broadcast any messages to the client on the client's subnet. A client that can receive unicast IP datagrams before its protocol software has been configured SHOULD clear the BROADCAST bit to 0. The BOOTP clarifications document discusses the ramifications of the use of the BROADCAST bit [21]. -- RFC 1531, 1993 (!)
Flags. To work around some clients that cannot accept IP unicast datagrams before the TCP/IP software is configured, DHCP uses the flags field. The leftmost bit is defined as the BROADCAST (B) flag. The remaining bits of the flags field are reserved for future use. They must be set to zero by clients and ignored by servers and relay agents. -- Microsoft Technet.
I have seen no FOSS contributions by Apple that I actually use
That doesn't mean anything. IBM's the fair haired boy in the Linux community right now, but I don't think I use any of their open source contributions. Not being a self-centered son-of-a-bitch I don't interpret that to mean "IBM's contributions are worthless" though.
I see plenty of attempts by them to badmouth FOSS operating systems
From Apple? Like what?
The only argument that holds any water is whether removal of the copyright notices is permitted by law; if it isn't, then it will get fixed--hopefully in a way that still imposes GPL-like restrictions on it,
Friend, I already pointed out that they don't need to remove the BSD license to add another license to a derived work. That's, in fact, exactly what you've been COMPLAINING about being the problem with the BSDL... that it doesn't prohibit derived works from being licensed with a more restrictive license such as the GPL. You haven't even been paying attention to *yourself*.
Bad licensing choices in the *BSD operating systems have supported both Apple and Microsoft in their efforts to hurt FOSS.
I won't argue with you about Microsoft, but you're making things up about Apple.
Apple has bent over backwards to support FOSS:
* They're releasing their modified versions of the BSD-licensed code they're using, just as if it was GPL. * They're releasing almost all their base OS code under a FOSS license. * They're releasing almost all their utilities and command line applications under a FOSS license, which they wouldn't have to do even if they used a GPL kernel. * They've put some of their GPL-licensed code in a public repository, which (no matter what some people were claiming) really is going far beyond what the GPL requires. * They've continued to do all this despite being attacked for things that EVERY TIME turned out not to just be following the letter of the licenses but went far beyond that and clearly followed the spirit of open source.
If the BSD code was GPL licensed, Apple wouldn't have been able to use it. Their business model doesn't support a completely open source kernel and changing their business model to allow that... right after pulling out of their disasterous attempt to become a purely software company... would not have been survivable. And even if it was, they have to be able to license proprietary drivers. They might have been able to arrange some kind of hack to mix proprietary drivers in a GPL Linux kernel, but those kind of shenanigans were in the future when they started down this path... and I can't imagine Jobs betting the company on that.
As a result Apple's gone from a completely closed company to one that's a solid and even enthusiastic supporter of open source, and the BSD license has done more to change Apple's licensing policies than the GPL could have.
FOSS is not one big, happy family,
Clearly. I'm not sure why you don't see this as a problem.
and many FOSS developers neither want to be friends of people on the *BSD project
Clearly. I'm not sure why you don't see this is a problem.
nor do they give a damn about what Theo considers "ethical".
They removed the BSD license from files that *weren't* dual-licensed. That's not just unethical, it's illegal.
If the GPL software wins, evidently, its license and/or its technology were stronger.
No, just the license. The GPL acts like the "applications barrier to entry" that makes Microsoft the dominant player in desktop software. It distorts the market, giving GPLed technology an advantage regardless of its technical merits.
This has not been good for the desktop operating system and office automation markets, why do you think it would be good for open source?
I really don't care about the history
Obviously. If you don't care about it, though, you shouldn't comment on it.
Very good, you had me going for almost half an hour, before I noticed that text was not in all the files that were modified and had the BSD license or Reyk's license (which is even less restrictive than the BSD one) stripped out.
Very good, you had me going for almost half an hour, before I noticed that text was not in all the files that were modified and had the BSD license stripped out.
Apple has 10% or less of the personal computer market, yet iTunes has a dominant position in the online music market. To me, this kind of implies that the majority of the people who are impacted by this (and thus going to complain about it, or decide that this entitles them to rip off copies of NBC products) are not actually Mac users at all. But maybe I'm making a mistake in not generalizing from a single data point, what do you think?
That seems like an odd thing to include, since the current BSDL already permits redistribution under the terms of the GPL version 2. That kind of dual-licensing is normally only used when the licenses aren't compatible in that way.
Who put that text in there? It appears in files with multiple original authors. What was the history of this particular boilerplate, and was it introduced by someone who had the right to do so?
The ability to take BSD code and use it without being forced to give it back is what BSD licenses are all about. If GPL'ed projects find it preferable to fork, lock out, and not give back, that's no worse than if commercial companies do it
Actually, there's a difference.
Commercial software can't compete with open source software on an otherwise level playing field, so commercial products are only direct competitors to open source ones if the commercial product has some overwhelming benefit (for example OS X). The BSD license fits this situation very well: a commercial product can use BSD code, but if it's competing with open source products it has to be a much better product. The result is that the BSD license promotes a meritocracy.
GPL software competes directly with non-GPL open source software, so when code goes under the GPL it doesn't vanish, like it would if it went commercial, it remains in competition with the original software.
As for the rest of your article... the history of the debate and the players aren't who you seem to think they are.
First... nobody's got a foot to stand on in the "who's been bitching the hardest" stakes. Let's not go down the road of complaining about what high profile people said or didn't say, or the language they use... please. I'm not going to even mention specific cases because they're beside the point.
Second... Theo isn't a spokesman for anything but his particular project. There isn't any spokesman, really... the BSD community doesn't have a single figurehead or leader. There's Theo on the OpenBSD project, Jordan on FreeBSD, the ex-CSRG folks like Kirk McKusick, but no equivalent of Linus or RMS.
No where have I ever seen mention that a dual license requires acceptance of BOTH licenses. It allows you to accept and use either license.
That's a completely different kind of "dual license".
You're thinking of the case where one copyright owner chooses to release a work under two licenses, and explicitly notes that you have the choice of using either.
This is a case where the resulting work has more than one copyright on it, and each copyright owner applies a different license to the work. BOTH licenses remain in effect. The BSDL "allows relicensing" only in so far as it allows the creation of modified works that have more restrictions than the BSDL. The resulting work remains copyrighted by the original author *as well as* the person who created the derived work, and remains subject to both licenses.
It's not true: he can modify and distribute under BSD the original code that was released under BSD, he can't distribute as BSD whatever was added and licensed under GPL. So none is stealing his work, they are just licensing their intellectual work as they feel it's better.
They can only do that if they are the copyright holders for the entire work.
If the work is licensed under the BDSL then he CAN NOT remove that fact by *relicensing* it under the GPL. In combining the GPL code with the BSDL code he has created a *dual licensed* work. There's no way around that.
If this was not true then (a) there wouldn't be any BSD, because this is what the USL violated that made them back down in the USL-CSRG case, and (b) Linux might also be in trouble, because in the aftermath of the USL-CSRG case a lot of old AT&T code was released, and that code was a hugely effective part of the smackdown laid on SCO in the SCO-Linux case.
If Jobs REALLY had balls, he'd do the Zune right...
What does "doing the Zune right" mean?
I don't get the whole "Zune" thing. It's just a repackaged Toshiba MP3 player with crippled Wifi for astroturfing. You can get MP3 players from Korea that do things like hook together through their USB ports and swap the song you're playing at USB2 speeds, which is way better than passing around a demo version that self destructs over radio.
There's nothing unique or innovative about it, what's the big deal?
Not possible. You're never not dependent on something or someone.
Don't be silly. You're taking the word "dependence" far beyond its meaning in this discussion here. If you have two independent mechanisms for completing a task, either one of which is sufficient, then you're not *dependent* on either mechanism.
If there's a hiccup in the grid it will get a lot of attention. If your little hut is washed into the sea no one may ever know.
So take both ****ing pills already. Wake up and smell the magic smoke... YOU CAN HAVE BOTH.
What we're talking about is the trend of the future
What we're talking about is the past, 25 years ago, and what broke us out were "powerful" personal computers that are beaten by five year old PDAs today.
When a multimedia application begins playback, the multimedia APIs it uses call the MMCSS service to boost the priority of the playback thread into the realtime range, which covers priorities 16-31, for up to 8ms of every 10ms interval of the time, depending on how much CPU the playback thread requires. Because other threads run at priorities in the dynamic priority range below 15, even very CPU intensive applications won't interfere with the playback.
An audio playback thread should require negligable CPU time. All it should be doing is queueing up buffers for the I/O device. Unless the audio buffer size is fixed and ludicrously small, the playback thread shouldn't even need to wake up at all in any given 10ms interval, nore require a fraction of that time when it does wake up.
While DPCs execute with interrupts enabled, they take precedence over all thread execution, regardless of priority, on the processor on which they run, and can therefore impede media playback threads.
This seems like an odd design. The whole point of a multithreaded kernel is to avoid this kind of thing. Standard UNIX TCP processing operates in this kind of regime, sure, but it was designed for a monolithic kernel. The Windows TCP stack started out that way, but people from Microsoft have claimed that the BSD networking code was turfed from Windows NT back in the '90s.
Because the standard Ethernet frame size is about 1500 bytes, a limit of 10,000 packets per second equals a maximum throughput of roughly 15MB/s.
Assuming all packets are maximum size (eg, bulk file transfer). Some workloads have average packet size well below that. I've found Microsoft estimates for streaming audio packet size below 1000 bytes.
Personally, I would prefer it if Microsoft retained the BSD stack. Saying that they were using one of the most reliable and best supported stacks out there rather than doing what Allen Cox did in Linux and reimplement it would hardly count as "bashing" Microsoft. Unfortunately, that didn't happen.
Yes, the Windows TCP stack was originally based on the BSD stack (it may have been originally based on the Lachman port of BSD networking, since that was the TCP stack available in Microsoft's Xenix). This stack did not remain untouched for long, and by the release of Windows 2000 it's unlikely that anything but the userland applications were still based on Berkeley code.
I'm not arguing that Microsoft hadn't substantially changed the stack in Vista. Just noting that it wasn't BSD code they replaced if they did so.
Slashdot Mirror
Please show me any publications that demonstrate significantly better usability of Apple products relative to Linux or Windows.
What, I have to prove my subjective experience? Not to mention that you're losing the plot here... first it was "Apple's attacking FOSS". I challenged that, so you claimed that Apple was specifically attacking the usability and performance of FOSS. Now you're saying that I need to prove that Apple's software is more usable than Linux and Windows. Last time I checked, Windows wasn't FOSS. Let's get back to the original point here, how is Apple claiming something that even many Linux boosters agree with "an attack on FOSS"?
I meant what I said: if IBM has wronged you, then speak out about it, nothing more and nothing less.
It's kind of hard to tell just what you mean, you're changeable as the wind and as hard to pin down.
Se let's get back on track. first it was "Apple's attacking FOSS". I pointed out that Apple's released an awful lot of OSX as open source software, far beyond what they need to by even the most generous interpretation of any of the licenses involved. You complained that you don't happen to use any of it. My response about IBM was not an attack on IBM, it was simply astonishment that you'd so moved the goalposts. If a company doesn't release anything YOU don't PERSONALLY use, they're attacking FOSS? Is that what you mean?
I never claimed that Apple is "opposed" to FOSS
What does "Bad licensing choices in the *BSD operating systems have supported both Apple and Microsoft in their efforts to hurt FOSS." mean then?
"Easily misinterpreted marketroid" comments is exactly what the problem is
Ah, gotcha. I'm sure you'll be right there on the attack next time Eben Moglen or RMS or ESR or Theo run off some easily misinterpreted marketroid comment. No? Just Theo?
Basically, the bottom line is this... everyone talks up their own stuff, and tells us how it's better than the competition. Apple does it, the FSF does it, Gentoo and OpenBSD and Ubuntu and Mozilla and Intel and IBM and me and you too. If you don't, you get buried. Pushing your stuff isn't "an effort to hurt FOSS" any more than the KDE vs Gnome rivalry is "an effort to hurt FOSS".
Companies that want to hurt FOSS don't use much, either, because once you split from the community you have to duplicate the efforts of the community... which is why Microsoft replaced the BSD TCP stack and has barely updated the BSD userland utilities they still use in years. What Microsoft's got from BSD is pretty much limited to Interix, and they're using GPL code in there as well.
Companies that benefit the most from FOSS are the ones that stay part of the community. The BSD license gives companies a taste. Some of them don't take the bait, some do. When they do, FOSS wins. When they don't, they don't get much out of it in the long run, so who cares? In the long run everyone wins. IBM took the bait, Apple took the bait, Microsoft nibbled and swam away. That's way better than I would have expected when I started getting excited by open source (or what would be tagged 'open source' 20 years later ) and open systems.
Setting the broadcast bit should not break your DHCP server.
That's also true. I'm not defending the DHCP servers in question, though I would like a bit more details about what the actual bug there involves, and why they didn't respond. Alas, googling brings up more smoke than light.
My point is that Microsoft is not blameless here. If you know anything more about their decision I'd really like to know what led them to start setting this flag now, when it's been discouraged since 1993* in booth DHCP and BOOTP, and since their TCP stack doesn't actually seem to need it. This is unfortunately symptomatic of Microsoft's general approach to standards.
* And its interesting to note that it's only required for Infiniband because of the size of Infiniband addresses... NOT because of the reason the flag was originally created.
I wouldn't call libertarianism a meta-philosophy. It's just another spectrum, the libertarian-authoritarian spectrum, separate from the left-right economic spectrum.
That's another way to describe it, but most libertarians I know don't seem terribly opposed to authoritarianism in general, just specifically against a strong and centralized government. Apart from the left libertarians most of them seem to think private sector authoritarianism is just fine, and many seem to think it would just vanish away when the goad of Big Government evaporates (the same kind of wishful thinking typical of classic Marxists and many Anarchists).
I do like the multi-spectrum meta-philosophy, though, and use it quite a lot myself.
I wrote: "I can't abide it as a server, of course, but servers aren't really Apple's market."
I'd like to clarify this, since I didn't notice until after I posted that the antecedent of "it" is uncertain. I'm referring to OS X here, not Linux. Linux is a perfectly competent server OS, and I've used it in that role in the past when appropriate. I (obviously) prefer BSD (FreeBSD to be specific, though I did give OpenBSD a trial), but that's a preference... not a litmus test.
Like when Apple employees make unsubstantiated claims about OS X usability and performance relative to other systems, misrepresent FOSS inventions as their own, and misrepresent OS X as a "better Linux" and a superset of Linux when selling to universities and companies.
:)
As for the first, I hardly think anyone in the open source community, myself not excluded, have any standing when it comes to comparing the usability of their products to Apple's. Performance? Apple's found themselves caught boasting about their hardware often enough but I hardly think that was targeted at FOSS particularly - their campaigns seem generally to target Microsoft and (until the last few years) intel. Perhaps a citation or two would jog my memory.
I'm unsure what you're referring to with "represent FOSS inventions as their own", again a citation might jog my memory, it should be easy for you to find a dozen if they're particularly targeting FOSS with this kind of thing, and not Microsoft.
As for OS X being "a better Linux", you should be proud that Apple thinks Linux is the best desktop UNIX after OS X. Of course I'd have said "a better UNIX" myself, which comes to the same thing, since Linux is UNIX and it really is hard to argue that Linux is anywhere close to OS X on the desktop. I can't abide it as a server, of course, but servers aren't really Apple's market.
In any event, this hardly counts as a "campaign against FOSS". If Apple was engaged in such a campaign it wouldn't be easily misinterpreted marketroid comments. Not from the same source as the "I'm a PC" ads, it wouldn't.
Of course if someone in the Linux community were to make a "formal mistake" in public comments or a code release I'm sure you'd be just as quick to chastise them.
If IBM has wronged you, speak out about it.
If you take my comment as evidence that I have a grudge against IBM, then it's no wonder that you're seeing conspiracies elsewhere. I'm a happy former customer of IBM, and I'm grateful for their excellent laptops (there's no laptop I'd rather use than the Thinkpad) and hard drives (still available under the Hitachi label). I'm also grateful to IBM for their work in user interfaces, glad that Microsoft picked up on that, and a bit upset that Microsoft abandoned IBM's UI standards in Windows 9x and that Apple never picked up on them.
Again, the fact that I'm personally not using Eclipse or SOAP simply means that I'm not personally using Eclipse or Soap. I'm not the one who is mistaking "I don't use X" for "X is worthless", or "I'm not benefitting from Y's actions" for "Y's actions don't matter".
Come on, be specific, rather than telling me why I should like Apple.
I'm a former Mac detractor and current Mac user who thinks Apple's hardware is quite horrid, but willing to put up with it for the sake of the software. I also continue to use other platforms (both open source and proprietary) where they're best suited. I don't currently have any Linux boxes in active use, but that's just the chance of the moment... I've used it in the past and I'm sure I'll be using it in the future.
As for why you should like Apple, well, I've no particular reason to care if you like them or not, I'm mostly disturbed by the fact that you're upset with them for such a bizarre and unlikely reason as their being opposed to free and open source software. For all their faults (and there are many) that's perhaps the least convincing one I could imagine.
If you were to complain about their snobbishness and elitism, or about OS X not running on generic hardware, or about their appalling 'style before everything' design and passive-aggressive attachment to one-button mice, or about the technical flaws in HFS+, or the unnecessary overhead in Quartz or MachO binaries, or anything they're actually guilty of, I'd agree... and if there were any alternative that did as good a job I'd be all over it by now. But hostility towards open source? It ain't there.
First, nerds don't tend to be Libertarians all that much. It just seems that way because pretty much all Libertarians are nerds. But then, pretty much all of the theoretically-based extreme positions are dominated by nerds, because pretty much all theoretically based ANYTHING is dominated by nerds. The kind of obsessive attention to academic details is kind of a defining feature of nerds.
Second, Libertarianism is kind of a meta-philosophy. There's left-libertarians as well as right-libertarians, though the rights tend to dominate because the heartland of libertarianism is America and even a lot of self-identified "left wing" groups in America would be classified as right-wing elsewhere in the world.
Only if it's connecting over infiniband or firewire. Over ethernet it SHOULD set that bit to zero.
Ah, the dangers in summarizing RFCs.
I have seen no FOSS contributions by Apple that I actually use
That doesn't mean anything. IBM's the fair haired boy in the Linux community right now, but I don't think I use any of their open source contributions. Not being a self-centered son-of-a-bitch I don't interpret that to mean "IBM's contributions are worthless" though.
I see plenty of attempts by them to badmouth FOSS operating systems
From Apple? Like what?
The only argument that holds any water is whether removal of the copyright notices is permitted by law; if it isn't, then it will get fixed--hopefully in a way that still imposes GPL-like restrictions on it,
Friend, I already pointed out that they don't need to remove the BSD license to add another license to a derived work. That's, in fact, exactly what you've been COMPLAINING about being the problem with the BSDL... that it doesn't prohibit derived works from being licensed with a more restrictive license such as the GPL. You haven't even been paying attention to *yourself*.
Bad licensing choices in the *BSD operating systems have supported both Apple and Microsoft in their efforts to hurt FOSS.
I won't argue with you about Microsoft, but you're making things up about Apple.
Apple has bent over backwards to support FOSS:
* They're releasing their modified versions of the BSD-licensed code they're using, just as if it was GPL.
* They're releasing almost all their base OS code under a FOSS license.
* They're releasing almost all their utilities and command line applications under a FOSS license, which they wouldn't have to do even if they used a GPL kernel.
* They've put some of their GPL-licensed code in a public repository, which (no matter what some people were claiming) really is going far beyond what the GPL requires.
* They've continued to do all this despite being attacked for things that EVERY TIME turned out not to just be following the letter of the licenses but went far beyond that and clearly followed the spirit of open source.
If the BSD code was GPL licensed, Apple wouldn't have been able to use it. Their business model doesn't support a completely open source kernel and changing their business model to allow that... right after pulling out of their disasterous attempt to become a purely software company... would not have been survivable. And even if it was, they have to be able to license proprietary drivers. They might have been able to arrange some kind of hack to mix proprietary drivers in a GPL Linux kernel, but those kind of shenanigans were in the future when they started down this path... and I can't imagine Jobs betting the company on that.
As a result Apple's gone from a completely closed company to one that's a solid and even enthusiastic supporter of open source, and the BSD license has done more to change Apple's licensing policies than the GPL could have.
FOSS is not one big, happy family,
Clearly. I'm not sure why you don't see this as a problem.
and many FOSS developers neither want to be friends of people on the *BSD project
Clearly. I'm not sure why you don't see this is a problem.
nor do they give a damn about what Theo considers "ethical".
They removed the BSD license from files that *weren't* dual-licensed. That's not just unethical, it's illegal.
If the GPL software wins, evidently, its license and/or its technology were stronger.
No, just the license. The GPL acts like the "applications barrier to entry" that makes Microsoft the dominant player in desktop software. It distorts the market, giving GPLed technology an advantage regardless of its technical merits.
This has not been good for the desktop operating system and office automation markets, why do you think it would be good for open source?
I really don't care about the history
Obviously. If you don't care about it, though, you shouldn't comment on it.
Very good, you had me going for almost half an hour, before I noticed that text was not in all the files that were modified and had the BSD license or Reyk's license (which is even less restrictive than the BSD one) stripped out.
Very good, you had me going for almost half an hour, before I noticed that text was not in all the files that were modified and had the BSD license stripped out.
I think Mac users in general need to grow up.
Apple has 10% or less of the personal computer market, yet iTunes has a dominant position in the online music market. To me, this kind of implies that the majority of the people who are impacted by this (and thus going to complain about it, or decide that this entitles them to rip off copies of NBC products) are not actually Mac users at all. But maybe I'm making a mistake in not generalizing from a single data point, what do you think?
That seems like an odd thing to include, since the current BSDL already permits redistribution under the terms of the GPL version 2. That kind of dual-licensing is normally only used when the licenses aren't compatible in that way.
Who put that text in there? It appears in files with multiple original authors. What was the history of this particular boilerplate, and was it introduced by someone who had the right to do so?
The ability to take BSD code and use it without being forced to give it back is what BSD licenses are all about. If GPL'ed projects find it preferable to fork, lock out, and not give back, that's no worse than if commercial companies do it
Actually, there's a difference.
Commercial software can't compete with open source software on an otherwise level playing field, so commercial products are only direct competitors to open source ones if the commercial product has some overwhelming benefit (for example OS X). The BSD license fits this situation very well: a commercial product can use BSD code, but if it's competing with open source products it has to be a much better product. The result is that the BSD license promotes a meritocracy.
GPL software competes directly with non-GPL open source software, so when code goes under the GPL it doesn't vanish, like it would if it went commercial, it remains in competition with the original software.
As for the rest of your article... the history of the debate and the players aren't who you seem to think they are.
First... nobody's got a foot to stand on in the "who's been bitching the hardest" stakes. Let's not go down the road of complaining about what high profile people said or didn't say, or the language they use... please. I'm not going to even mention specific cases because they're beside the point.
Second... Theo isn't a spokesman for anything but his particular project. There isn't any spokesman, really... the BSD community doesn't have a single figurehead or leader. There's Theo on the OpenBSD project, Jordan on FreeBSD, the ex-CSRG folks like Kirk McKusick, but no equivalent of Linus or RMS.
No where have I ever seen mention that a dual license requires acceptance of BOTH licenses. It allows you to accept and use either license.
That's a completely different kind of "dual license".
You're thinking of the case where one copyright owner chooses to release a work under two licenses, and explicitly notes that you have the choice of using either.
This is a case where the resulting work has more than one copyright on it, and each copyright owner applies a different license to the work. BOTH licenses remain in effect. The BSDL "allows relicensing" only in so far as it allows the creation of modified works that have more restrictions than the BSDL. The resulting work remains copyrighted by the original author *as well as* the person who created the derived work, and remains subject to both licenses.
It's not true: he can modify and distribute under BSD the original code that was released under BSD, he can't distribute as BSD whatever was added and licensed under GPL. So none is stealing his work, they are just licensing their intellectual work as they feel it's better.
They can only do that if they are the copyright holders for the entire work.
If the work is licensed under the BDSL then he CAN NOT remove that fact by *relicensing* it under the GPL. In combining the GPL code with the BSDL code he has created a *dual licensed* work. There's no way around that.
If this was not true then (a) there wouldn't be any BSD, because this is what the USL violated that made them back down in the USL-CSRG case, and (b) Linux might also be in trouble, because in the aftermath of the USL-CSRG case a lot of old AT&T code was released, and that code was a hugely effective part of the smackdown laid on SCO in the SCO-Linux case.
So they removed mention.
Which violates the BSDL.
The GPL does not replace the BSDL. The resulting work remains subject to both licenses. The GPL can't change that.
Could be. NBC is in bed with MS and has been for years.
If Jobs REALLY had balls, he'd do the Zune right...
What does "doing the Zune right" mean?
I don't get the whole "Zune" thing. It's just a repackaged Toshiba MP3 player with crippled Wifi for astroturfing. You can get MP3 players from Korea that do things like hook together through their USB ports and swap the song you're playing at USB2 speeds, which is way better than passing around a demo version that self destructs over radio.
There's nothing unique or innovative about it, what's the big deal?
Not possible. You're never not dependent on something or someone.
Don't be silly. You're taking the word "dependence" far beyond its meaning in this discussion here. If you have two independent mechanisms for completing a task, either one of which is sufficient, then you're not *dependent* on either mechanism.
If there's a hiccup in the grid it will get a lot of attention. If your little hut is washed into the sea no one may ever know.
So take both ****ing pills already. Wake up and smell the magic smoke... YOU CAN HAVE BOTH.
What we're talking about is the trend of the future
What we're talking about is the past, 25 years ago, and what broke us out were "powerful" personal computers that are beaten by five year old PDAs today.
When a multimedia application begins playback, the multimedia APIs it uses call the MMCSS service to boost the priority of the playback thread into the realtime range, which covers priorities 16-31, for up to 8ms of every 10ms interval of the time, depending on how much CPU the playback thread requires. Because other threads run at priorities in the dynamic priority range below 15, even very CPU intensive applications won't interfere with the playback.
An audio playback thread should require negligable CPU time. All it should be doing is queueing up buffers for the I/O device. Unless the audio buffer size is fixed and ludicrously small, the playback thread shouldn't even need to wake up at all in any given 10ms interval, nore require a fraction of that time when it does wake up.
While DPCs execute with interrupts enabled, they take precedence over all thread execution, regardless of priority, on the processor on which they run, and can therefore impede media playback threads.
This seems like an odd design. The whole point of a multithreaded kernel is to avoid this kind of thing. Standard UNIX TCP processing operates in this kind of regime, sure, but it was designed for a monolithic kernel. The Windows TCP stack started out that way, but people from Microsoft have claimed that the BSD networking code was turfed from Windows NT back in the '90s.
Because the standard Ethernet frame size is about 1500 bytes, a limit of 10,000 packets per second equals a maximum throughput of roughly 15MB/s.
Assuming all packets are maximum size (eg, bulk file transfer). Some workloads have average packet size well below that. I've found Microsoft estimates for streaming audio packet size below 1000 bytes.
Personally, I would prefer it if Microsoft retained the BSD stack. Saying that they were using one of the most reliable and best supported stacks out there rather than doing what Allen Cox did in Linux and reimplement it would hardly count as "bashing" Microsoft. Unfortunately, that didn't happen.
Yes, the Windows TCP stack was originally based on the BSD stack (it may have been originally based on the Lachman port of BSD networking, since that was the TCP stack available in Microsoft's Xenix). This stack did not remain untouched for long, and by the release of Windows 2000 it's unlikely that anything but the userland applications were still based on Berkeley code.
I'm not arguing that Microsoft hadn't substantially changed the stack in Vista. Just noting that it wasn't BSD code they replaced if they did so.