The Institute of Creation Research and its dishonest fundamentalist members are probably going to pretend that this didn't happen. Duane Gish will go on claiming at debates that no transitionary fossils have been foudn between dinosaurs and birds, and so on.
What I worry is that people in general will hear this and say "so?" and forget about it like every other tidbit of science they've ever heard. Then later on they're going to be watching some well-funded, slickly-produced Creationist video that claims there aren't any known transitionary fossils and just swallow it. How else can Kansas be explained?
:-(
It's just good that most of the brain-disease known as creationism is confined to the USA and undeveloped countries where education is so poor people will actually fall for the literalist fundie crap.
This is a new OS. New OSes require migration. They are migrating FROM two OSes. Thus, they must make migration possible from both lest they leave users without an upgrade path.
This is a nontrivial task, but they are making it as easy as possible. Complexity is bad when it means that system behavior is complex. However, complex code written to make system behavior as simple as possible is GOOD.
Note that 450 MHz G4 Macs outfitted with 256MB of RAM using the crappy old MacOS can beat the Dell system you quoted by aproximately 100% using tests run by Henry Norr. On a modern OS like Mac Ten or LinuxPPC, likely the same tests would show even more dramatic performance gains.
If the feild is leveled on those same tests and everyone has 256 MB of RAM, then the 450 G4 is only 30-50% faster. But still, that means that the G4 is potentially faster than a single processor i386 CPU at any price, even given the crappy old MacOS. That's just going to be more true with a modern OS running on that box.
And with the advent of MP G4s and Mac Ten, I think Macs may be much faster than any i386 PC of any type. granted, this will be only for certain tasks. However, those tasks are non-trivial things like web serving, video processing, graphic manipulation and perhaps even game performance. Sure, the 1000MHz CPUs will still be faster spellchecking in MS Word, but I don't think anyone cares.
Because if you just started in 1995, then that's not that many years.
That said, your comments on preemptive multitasking and memory protection have some merit. Certainly the former is a major stumbling block to performance when multiple applications are running (which is pretty much all the time).
Your doubt about the veracity of Photoshop performance is ill-founded, however. Independant testers such as Henry Norr have found those photoshop results to be direct reflections of real-world performance for normal professional users. I have no trouble believing this because I have seen similar results with my own eye. They really are that fast. That shows the power of the processor, though, on a task not greatly slowed by the poor OS performance. Nutscrape Navigator, on the other hand, is a case in point about problems with the Mac - it's slow(even on wicked-fast G4s), it crashes a lot(often bringing the system with it), and it hogs CPU time even when idling.
In conclusion: PowerPC=very fast. MacOS subsystems=very crappy. I think that's more or less where you were going with this anyway. But those benchmarks are for real, and those Macs were not overclocked, nor were the PCs crippled.
IEEE1394 gives you isochronus transfer mode an the ability for two devices, say a iLink digital camcorder and a 1394 harddrive to communicate directly without lugging a computer around with you. Alternatively, the new Powerbooks have IEEE built in.
In any case, the isochronus transfer keeps you from dropping frames as even the fastest of asynchronous busses like SCSI can. It locks down a guaranteed bandwidth for devices that need it(and DV devices are the classic example of somethign that needs a certain amount of guaranteed bandwidth), so no spike in bus usage can lose you a frame that you can never get back.
The device-to-device communication is nice for keeping things light, but apparently not absolutely necessary when you consider laptops are not that bad to work with.
They plan on riding QT's proprietary features to glory, leverage-wise. I don't think they'll open-source it, because they're so afraid of their competitors getting ahold of some of their coolest technologies and copying them.
Now, under normal circumstances, this wouldn't be such a bad thing, but when was the last time you changed your player for some media type when you didn't have to? What would be the incentive to use QT if all the other folks out there could decode.mov files, without penalty? Right now, QT is pretty much the only way to get high quality, relatively compact video, so they can fight the "Default" guys.
Now, one might say that open-sourceing the code will give them all of open-source's advantages - but those won't even begin to matter because of the reasons above. No one on the client side cares about anything except playing that Halo trailer without having to go try and find a translation utility or swapping players.
That said, I do expect that BSD and Linux flavored QT players might be coming in the not-too-distant future - shortly after OS X, I predict. They want leverage on the authoring and server side - they know they can't move clients from Win or Linux to Apple just by withholding players.
I got it back when it first came out. I havn't really read it since, but I found it to be something less than convincing in its spiritual arguments. That's not at all to say that it's a worthless book; Davies is a better than decent expositor of physics. Still, Brian Greene's Elegant Universe is head and shoulders above the Mind of God both as an exposition and as a snapshot of the modern frontiers of physics.
Gee- ZDNET and CNet both have managed not to mention this at all. IPv6, for example- talk about a buzzword feature! Yet not a peep.
But perhaps I ask too much. Maybe they will bring themselves to mention this sometime soon. Or maybe OSS will continue to be synonymous with Linux as if Linus invented Open Source. Not that I dislike Linux or Linus, but it irks me that it hogs the spotlight so. BSD is STILL more mature in many(most?) ways, and, as this pays testement, is still improving.
Many of these problems already have pertinent strategies worked out over the years, but I don't think you mean to suggest that OSS simply consists of making source code available to customers and working collaboratively. You know that there are new challenges involved here, not least of which is the influx of people who have NOT been programming for 20 years.
I didn't mean to suggest that every aspect of OSS is totally different from traditional programming. I certainly don't mean to suggest that/. or any other forum is anything like totally new.
All I'm saying is that things are chaotic now. BSD had years to mature in relative obscurity, but Linux has drawn a great deal of interest lately and powerful forces have grown interested in it. Strains that were less evident before are now going to have to be overcome with some new tactics, or at least by adapting the old tactics to a new world and teaching them to the young.
Did you notice that my post was two paragraphs long? I sketched a simple idea. That idea is that OSS is both new and rebellious. Things that are new and/or rebillious take a while to mature before they are traditional and no longer rebellious, and thus no longer suffer from the weaknesses of things that are new and rebellious. I mostly feared that what I was saying would already be clear to many people and thus worthless to post. You, on the other hand, have managed to disagree with aspects of my post that are so self-evident as to be nearly redundant.
You seem to think that OSS is traditional. Perhaps you feel it is a more traditional model than that of closed source internal projects. That can be argued, but in the end it's not relevant to the discussion of OSS vs proprietary software. Massive collaborative projects of today's Linux have no tradition that I know of, and they are what is meant by OSS, not simply that of programmers coding in their spare time.
As for your indictment of my use of the word "enterprise" - perhaps you should look up the word perfore you impugn my usage.
Further, a rebellion is not defined as a collection of rebellious people. What I said about fractious people populating the OSS rebellion is NOT a tautology. Peopel who are not normally rebellious may join a rebellion, and people with rebellious proclivities may choose not to rebel in fact.
Your assumption that proprietary software will always constitute the heart of mainstream programming projects is perhaps true, but perhaps it is not. You make a claim there that many would argue with. I might, and I might not. We shall have to simply wait and see for the final word on that.
Finally, it would appear that I have to concede that some people want to be the asshole, since you provide such a good example with your post. However, the fact is that any project manager who was forced to listen to your specious arguments day after day would either throw you off the team or find a way to ignore you.
So is the project manager, assuming there is one, supposed to e-mail her project members some literacy? Or perhaps you can only send that via snail-mail or ocean liner. Or maybe she should just throw anyone who cannot already rival Carl Sagan off the project?
Don't you think that project managers and the ad-hoc teams that constitute most of the open-source free labor out there do what they can to support literacy? That part is a no-brainer. What is more difficult is HOW to teach these footsoldiers of the OSS movement to read and write well. Just saying that it should be done as if it were a trivial task is awfully denigrating to all those out there who have been worrying at this problem for years.
Traditional programming process has had almost a half century to mature, where open source has had maybe five years to climb out of the ad-hoc status it has held so far. So naturally there will be some jaggies along the way.
Perhaps even more important, though, is that OSS is, right now, almost as much a rebellion as an enterprise all its own. Rebellions always attract fractious people who have an emotional attachment to "the cause". When OSS becomes mainstream and people start calmly devoting more attention to processs design than holy wars, most of the flame wars will settle down, IMHO. No one wants to be known as the asshole.
I see that they only commit you to two years of AOHell, but I don't know if that'd be a big factor. The reality is that the people buying these cheapo internet boxes are usually ultra-low-level consumers. They've never heard of Linux. Also, who's going to stock these in the retail stores where a lot of these consumer PCs are still sold? I think most of the sales are going to be to family and freinds of Linux advocates and to the Linux advocates themselves who need an extra PC. But how many people is that? Too soon, is my judgement. -N
"Squashing competition is a legitemate part of that, unless it strays into illegality(eg. a monopoly)."
That leaves out Microsoft.
The reason being a monopoly or squashing competition on their own are not illegal is because either one can be countered without the presence of the other, as long as one's product and business practices are sound. However, monopolies who squash competition are impossible to beat without huge blunders on the monopolist's part. This mucks up the gears of capitalism and cannot be allowed.
Amazon.com has nothing like a monopoly. At best, they have a strong market position. This is one of the few things they can do to assure that they keep it. -N
As was mentioned above, a company's purpose is to make money for its shareholders. Worrying about how nice they are being to the rest fo the field is only important so far as the state of the rest of the field effects the company's P&Ls. Squashing competition is a legitemate part of that, unless it strays into illegality(eg. a monopoly).
I think Jeff Bezos is attempting to avoid being a bad guy, because that would be bad for his company's image. I doubt his real concern is the freedom of the rest of the internet. And that's good, with respect to his responsibilities. So, if we give commentary that we really want a company to listen to, I think we should talk about it in terms of benefit to the company, not some moral code. I think Tim's done a reasonably good job of this, but even he strays a little too much into the assumption that Amazon.com and Jeff Bexos should somehow be concerned about the welfare of the rest of the world.
If we social critics and OSSers and etc would keep this in mind, then maybe companies would listen to us more often. Notice how well it worked in this instance. I bet if Tim had come up with some concrete and convincing exampleds of how hording patents could hurt Amazon.com specifically, then Jeff would have been been more enthusiastic in his agreement. More of the conflict of interest would have been removed.
This all seems laughably pie-in-the-sky to me. maybe, just maybe, they'll turn out to be right, but I'm not very impressed. They havn't really shown very much at all, if this article is to be used as the final source (a situation to which we are doomed for lack of references). More importantly, they seem to get bogged down in claiming that their threory solves non-existent "problems". As has been noted above, the "criticality" of the present is an experiencial artifact, not a fundamental feature of the universe. Also, the randomness of the future versus the recordedness of the past is another artifact of our experience(which is, remember, informed by very limited information), not a necessity of physics.
Superstring theory, on the other hand, seems to have tremendous power to resolve all the REAL problems of modern physics, if we can find ways to make the calculations tractable. I think the best bet for these guys is if their ideas fit in with M theory at some point. I suppose that's still a possibility. -N
Slashdot Mirror
The Institute of Creation Research and its dishonest fundamentalist members are probably going to pretend that this didn't happen. Duane Gish will go on claiming at debates that no transitionary fossils have been foudn between dinosaurs and birds, and so on.
:-(
What I worry is that people in general will hear this and say "so?" and forget about it like every other tidbit of science they've ever heard. Then later on they're going to be watching some well-funded, slickly-produced Creationist video that claims there aren't any known transitionary fossils and just swallow it. How else can Kansas be explained?
It's just good that most of the brain-disease known as creationism is confined to the USA and undeveloped countries where education is so poor people will actually fall for the literalist fundie crap.
I SAID NT!
This is a new OS. New OSes require migration. They are migrating FROM two OSes. Thus, they must make migration possible from both lest they leave users without an upgrade path.
This is a nontrivial task, but they are making it as easy as possible. Complexity is bad when it means that system behavior is complex. However, complex code written to make system behavior as simple as possible is GOOD.
See what I mean?
Note that 450 MHz G4 Macs outfitted with 256MB of RAM using the crappy old MacOS can beat the Dell system you quoted by aproximately 100% using tests run by Henry Norr. On a modern OS like Mac Ten or LinuxPPC, likely the same tests would show even more dramatic performance gains.
If the feild is leveled on those same tests and everyone has 256 MB of RAM, then the 450 G4 is only 30-50% faster. But still, that means that the G4 is potentially faster than a single processor i386 CPU at any price, even given the crappy old MacOS. That's just going to be more true with a modern OS running on that box.
And with the advent of MP G4s and Mac Ten, I think Macs may be much faster than any i386 PC of any type. granted, this will be only for certain tasks. However, those tasks are non-trivial things like web serving, video processing, graphic manipulation and perhaps even game performance. Sure, the 1000MHz CPUs will still be faster spellchecking in MS Word, but I don't think anyone cares.
-N
Because if you just started in 1995, then that's not that many years.
That said, your comments on preemptive multitasking and memory protection have some merit. Certainly the former is a major stumbling block to performance when multiple applications are running (which is pretty much all the time).
Your doubt about the veracity of Photoshop performance is ill-founded, however. Independant testers such as Henry Norr have found those photoshop results to be direct reflections of real-world performance for normal professional users. I have no trouble believing this because I have seen similar results with my own eye. They really are that fast. That shows the power of the processor, though, on a task not greatly slowed by the poor OS performance. Nutscrape Navigator, on the other hand, is a case in point about problems with the Mac - it's slow(even on wicked-fast G4s), it crashes a lot(often bringing the system with it), and it hogs CPU time even when idling.
In conclusion: PowerPC=very fast. MacOS subsystems=very crappy. I think that's more or less where you were going with this anyway. But those benchmarks are for real, and those Macs were not overclocked, nor were the PCs crippled.
-N
IEEE1394 gives you isochronus transfer mode an the ability for two devices, say a iLink digital camcorder and a 1394 harddrive to communicate directly without lugging a computer around with you. Alternatively, the new Powerbooks have IEEE built in.
In any case, the isochronus transfer keeps you from dropping frames as even the fastest of asynchronous busses like SCSI can. It locks down a guaranteed bandwidth for devices that need it(and DV devices are the classic example of somethign that needs a certain amount of guaranteed bandwidth), so no spike in bus usage can lose you a frame that you can never get back.
The device-to-device communication is nice for keeping things light, but apparently not absolutely necessary when you consider laptops are not that bad to work with.
-N
They plan on riding QT's proprietary features to glory, leverage-wise. I don't think they'll open-source it, because they're so afraid of their competitors getting ahold of some of their coolest technologies and copying them.
.mov files, without penalty? Right now, QT is pretty much the only way to get high quality, relatively compact video, so they can fight the "Default" guys.
Now, under normal circumstances, this wouldn't be such a bad thing, but when was the last time you changed your player for some media type when you didn't have to? What would be the incentive to use QT if all the other folks out there could decode
Now, one might say that open-sourceing the code will give them all of open-source's advantages - but those won't even begin to matter because of the reasons above. No one on the client side cares about anything except playing that Halo trailer without having to go try and find a translation utility or swapping players.
That said, I do expect that BSD and Linux flavored QT players might be coming in the not-too-distant future - shortly after OS X, I predict. They want leverage on the authoring and server side - they know they can't move clients from Win or Linux to Apple just by withholding players.
At least, I hope they're not that stupid.
-N
I got it back when it first came out. I havn't really read it since, but I found it to be something less than convincing in its spiritual arguments. That's not at all to say that it's a worthless book; Davies is a better than decent expositor of physics. Still, Brian Greene's Elegant Universe is head and shoulders above the Mind of God both as an exposition and as a snapshot of the modern frontiers of physics.
-N
Gee- ZDNET and CNet both have managed not to mention this at all. IPv6, for example- talk about a buzzword feature! Yet not a peep.
But perhaps I ask too much. Maybe they will bring themselves to mention this sometime soon. Or maybe OSS will continue to be synonymous with Linux as if Linus invented Open Source. Not that I dislike Linux or Linus, but it irks me that it hogs the spotlight so. BSD is STILL more mature in many(most?) ways, and, as this pays testement, is still improving.
*sigh* - At least there will always be Slashdot.
-N
Many of these problems already have pertinent strategies worked out over the years, but I don't think you mean to suggest that OSS simply consists of making source code available to customers and working collaboratively. You know that there are new challenges involved here, not least of which is the influx of people who have NOT been programming for 20 years.
/. or any other forum is anything like totally new.
I didn't mean to suggest that every aspect of OSS is totally different from traditional programming. I certainly don't mean to suggest that
All I'm saying is that things are chaotic now. BSD had years to mature in relative obscurity, but Linux has drawn a great deal of interest lately and powerful forces have grown interested in it. Strains that were less evident before are now going to have to be overcome with some new tactics, or at least by adapting the old tactics to a new world and teaching them to the young.
-N
Did you notice that my post was two paragraphs long? I sketched a simple idea. That idea is that OSS is both new and rebellious. Things that are new and/or rebillious take a while to mature before they are traditional and no longer rebellious, and thus no longer suffer from the weaknesses of things that are new and rebellious. I mostly feared that what I was saying would already be clear to many people and thus worthless to post. You, on the other hand, have managed to disagree with aspects of my post that are so self-evident as to be nearly redundant.
You seem to think that OSS is traditional. Perhaps you feel it is a more traditional model than that of closed source internal projects. That can be argued, but in the end it's not relevant to the discussion of OSS vs proprietary software. Massive collaborative projects of today's Linux have no tradition that I know of, and they are what is meant by OSS, not simply that of programmers coding in their spare time.
As for your indictment of my use of the word "enterprise" - perhaps you should look up the word perfore you impugn my usage.
Further, a rebellion is not defined as a collection of rebellious people. What I said about fractious people populating the OSS rebellion is NOT a tautology. Peopel who are not normally rebellious may join a rebellion, and people with rebellious proclivities may choose not to rebel in fact.
Your assumption that proprietary software will always constitute the heart of mainstream programming projects is perhaps true, but perhaps it is not. You make a claim there that many would argue with. I might, and I might not. We shall have to simply wait and see for the final word on that.
Finally, it would appear that I have to concede that some people want to be the asshole, since you provide such a good example with your post. However, the fact is that any project manager who was forced to listen to your specious arguments day after day would either throw you off the team or find a way to ignore you.
-N
So is the project manager, assuming there is one, supposed to e-mail her project members some literacy? Or perhaps you can only send that via snail-mail or ocean liner. Or maybe she should just throw anyone who cannot already rival Carl Sagan off the project?
Don't you think that project managers and the ad-hoc teams that constitute most of the open-source free labor out there do what they can to support literacy? That part is a no-brainer. What is more difficult is HOW to teach these footsoldiers of the OSS movement to read and write well. Just saying that it should be done as if it were a trivial task is awfully denigrating to all those out there who have been worrying at this problem for years.
-N
Traditional programming process has had almost a half century to mature, where open source has had maybe five years to climb out of the ad-hoc status it has held so far. So naturally there will be some jaggies along the way.
Perhaps even more important, though, is that OSS is, right now, almost as much a rebellion as an enterprise all its own. Rebellions always attract fractious people who have an emotional attachment to "the cause". When OSS becomes mainstream and people start calmly devoting more attention to processs design than holy wars, most of the flame wars will settle down, IMHO. No one wants to be known as the asshole.
-N
I see that they only commit you to two years of AOHell, but I don't know if that'd be a big factor. The reality is that the people buying these cheapo internet boxes are usually ultra-low-level consumers. They've never heard of Linux. Also, who's going to stock these in the retail stores where a lot of these consumer PCs are still sold? I think most of the sales are going to be to family and freinds of Linux advocates and to the Linux advocates themselves who need an extra PC. But how many people is that? Too soon, is my judgement. -N
The subject line says it all.
-N
Did you read my post?
"Squashing competition is a legitemate part of that, unless it strays into illegality(eg. a monopoly)."
That leaves out Microsoft.
The reason being a monopoly or squashing competition on their own are not illegal is because either one can be countered without the presence of the other, as long as one's product and business practices are sound. However, monopolies who squash competition are impossible to beat without huge blunders on the monopolist's part. This mucks up the gears of capitalism and cannot be allowed.
Amazon.com has nothing like a monopoly. At best, they have a strong market position. This is one of the few things they can do to assure that they keep it.
-N
As was mentioned above, a company's purpose is to make money for its shareholders. Worrying about how nice they are being to the rest fo the field is only important so far as the state of the rest of the field effects the company's P&Ls. Squashing competition is a legitemate part of that, unless it strays into illegality(eg. a monopoly).
I think Jeff Bezos is attempting to avoid being a bad guy, because that would be bad for his company's image. I doubt his real concern is the freedom of the rest of the internet. And that's good, with respect to his responsibilities. So, if we give commentary that we really want a company to listen to, I think we should talk about it in terms of benefit to the company, not some moral code. I think Tim's done a reasonably good job of this, but even he strays a little too much into the assumption that Amazon.com and Jeff Bexos should somehow be concerned about the welfare of the rest of the world.
If we social critics and OSSers and etc would keep this in mind, then maybe companies would listen to us more often. Notice how well it worked in this instance. I bet if Tim had come up with some concrete and convincing exampleds of how hording patents could hurt Amazon.com specifically, then Jeff would have been been more enthusiastic in his agreement. More of the conflict of interest would have been removed.
That's what I think, anyway.
-N
This all seems laughably pie-in-the-sky to me. maybe, just maybe, they'll turn out to be right, but I'm not very impressed. They havn't really shown very much at all, if this article is to be used as the final source (a situation to which we are doomed for lack of references). More importantly, they seem to get bogged down in claiming that their threory solves non-existent "problems". As has been noted above, the "criticality" of the present is an experiencial artifact, not a fundamental feature of the universe. Also, the randomness of the future versus the recordedness of the past is another artifact of our experience(which is, remember, informed by very limited information), not a necessity of physics.
Superstring theory, on the other hand, seems to have tremendous power to resolve all the REAL problems of modern physics, if we can find ways to make the calculations tractable. I think the best bet for these guys is if their ideas fit in with M theory at some point. I suppose that's still a possibility.
-N