A recent interview with the President of AMD said that he'd not been contacted by Apple. Apple apparently cross compiles a lot of their OS codebase on Athalons, but I'm told it is more just to keep options open in the future and to help find bugs. I'm not sure if that includes Aqua or iApps. (I doubt it)
In either case the 970 with be in the same general class as the Opteron and will offer very good SMP. Apple's problem is that between now and perhaps September they are way behind the speed curve for desktop machines. (The portable market they do much better in, IMO)
Ah yes. "The Day the Earth Froze" Sampo became a running gag, especially in USENET. Folks should go Google it in the alt.fan.mst3k group. It had a segment where Joel and the bots tried to figure out what in hell a Sampo was.
OrobosX is pretty stable from what I've seen, and I've used it a fair bit. I admit that the past while I've been running KDE as I like Konquerer. However given how few X11 apps I actually run of late, I generally don't load X11 at all.
I find that while I run a lot of shell scripts, most of the things I used X11 for (GUI administrators) have Aqua counterparts or I can roll my own with Applescript and Applescript Studio.
I think that Unix gurus who learn Applescript will be pleasantly surprised at its power. There isn't much equivalent in Linux. The O'Reilly book on it is so-so though. However you are a Unix guru and can browse the web for example scripts. Check OSX Hints a lot. The latest version adds amazing power.
$60 for an aluminum case? Time to rent some shop time. . .
Come on. That's ridiculous. The iPod is amazingly well designed and extremely sturdy feeling compared to other mp3 players. But $60?
often there is an article that ought to be cited in your work (because it was published before yours, and is related), but is in a journal unavailable at your university's library. There are thousands of journals, and their high costs (often thousands of dollars a year each) means that no library can have them all.
Have you tried interlibrary loan? You can, for a very small fee, get photocopies of articles from journals other libraries subscribe to. I used to do that all the time while in college. You can actually get full books as well. If you are referencing articles in a paper but don't read them simply because they weren't in your library then that is very deceitful. Further it is problematic scientifically since you are ignoring a possibly significant piece of evidence. What if the paper in question undercut some of your methodology? You'd never know. ..
This is an excellent point. While everyone else is pointing out that you can quote titles in a secondary way, the issue is why these sorts of errors continue to propagate. I've done fact checking, although admittedly for review in the humanities. But I do look up footnotes and references as best as possible. While my background is primarily physics, I'd always assumed that the problems of footnotes in the humanities didn't occur in physics.
This is the real scandal. What is going on with peer review? Hopefully peer review entails more than just glances through the article and seeing if it is plausible. . .
Umm they have conferences every year. While the term is still a bit of a dirty word among physicists here it is very well funded in Japan. I know that at the college I used to go to, BYU, Steve Jones was continuing research, although I don't know if he is still active in that field.
The big debate was that the experiments were very hard to conduct and the evidence was from caloromic readings while the physicists wanted various nuclear reaction evidence. The problem was that while caloromic readings were very persuasive scientists had no idea what was going on inside the palladium.
Probably the best papers are by Michael McKubre over at Stanford. Here's a good transcript of his talk at this year's conference on cold fusion. He's written a lot on the topic and gave a fascinating hour long interview to KUER in SLC. I'll not post the link to avoid slashdotting a small NPR station. For peer reviewed papers I was looking at LANL, but the links I had appear to have been moved behind a firewall. HEre are some I found doing a bit of looking:
I'm sure there are more if you do a little searching.
Re:Google contest ideas?
on
Google's new toys
·
· Score: 3, Interesting
None of them sound like entries in the Google contest. The only entry from the contest that sounded like it had relevance to business was the one that tied geographic maps to web searches. I was actually surprised they didn't implement that one.
There are a few ideas that I'd like to see Google ad, but these are interesting, if mostly proof of concepts. Froogle in particular is a good idea but doesn't allow price comparisons like MySimon does. Were they to add that it would be considerably more useful. Further it oddly seems to leave out a lot of stores, I noticed.
The web annotation tool is also a good idea. But I have a feeling that it needs to be better integrated with the google bar. (BTW - when is a Mac version coming?) Annotation has been discussed for quite a while. I have a wrox book at home on XML Metadata that discusses a lot of this. Unfortunately in practice it has never been applied to the general web.
There are lots of features of the original hypertext research in the 60's that are yet to be applied to the web. Google utilizes some concepts for its ranking, such as "links to" information. Wouldn't it be great to have a Google bar monitoriing your pages and make suggestings for helpful pages that link to the page you are on? There are many ideas like that which would improve ones browsing experience.
There are a few privacy concerns, but I'm exciting that Google offers these tests. The news site has become my favorite site on the web. It tends to get better stories than the other meta-news site I used to visit: DrudgeReport.com
Do I see some bitterness in the physics community? It is seen nowadays as very important for humanity to spend more money on the life sciences and less on physics. And the physics guys do not like it!
I think the problem is that basic science isn't funded well, regardless of discipline. Yes work in the life sciences is funded, but often with an emphasis on technology rather than understanding. You hear about all this work in neurology, genetics and so forth because of potential things that can be sold. Yet there are plenty of issues that don't get funded which are foundational.
One can argue that the advance in other areas of technology arose because funding at basic research was so good. It then had a "bubble up" effect.
Having said that though, lets be honest. Physics, even basic theoretical physics, was largely subsidized by the cold war. There were a lot of theoreticians doing what they liked on the side while getting paid for research in more applied areas. Further the biggest area of physics research right now is material science. And it isn't as if there is this huge funding shortage there.
So in a sense the poster is right. There is a bit of jealousy now that physics isn't quite the high rolling area it was up through the end of the cold war. However part of the problem is that in physics, most of the easy to answer problems are solved. What's left is [i]so[/i] complex and difficult and [i]so[/i] expensive that one has to be somewhat cautious in funding. I mean do we really need to spend a few billion dollars on the next collider to find the weight of the latest theoretical particle? Especially when the real work is in what is more fundamental. And there is still a ways to go before the superstring theorists and quantum loop theorists have much to give us that can be reasonably tested.
Having said all that one branch of physics/chemistry is about to make a comeback. Apparently a lot of the cold fusion stuff has been doing quite well the past few years. It has been duplicated in over 1000 different experiments. The remaining problems are in getting the proper impurities into palladium and doing some admittedly difficult material science work. (Which isn't to say one is even remotely close to making it a power source) So while the life sciences are deservedly getting the press today, energy problems along with global warming will push things back into the court of physicists and chemist within a few years.
As mentioned, the military NEEDS the space shuttle. Remember how much involvement they had in its design. They made sure the payload bay could fit its spying equipment. While many of these systems can be launched via cheaper means, recall that many shuttle missions are still primarily military in nature. With the war on terror there will be that much of an impetus to have better spy equipment and probably better GPS equipment and communication equipment.
The problem really is NASA as being more than simply a military shipping company.
The whole way that NASA has run the Space Station program and the Mars program have been dismall. On the one hand the public doesn't really give a damn about science, so if NASA pursued fairly cheap science projects with a good return, they'd dry up quickly. At the same time though the space station is a solution in search of a problem - and one that is VERY expensive.
To a degree all of this was just to help keep Soviet scientists around in Russia and not heading to the mid-east to develop nasty weapons. Further the military clearly had motives in keeping the Space Shuttle running. However now the Russians can't do much and haven't been able to move into commercial projects. Even in NASA the shuttles are wearing out with no replacements on the horizon.
The big question is whether all of these problems are a good thing or a bad thing. When you consider the BILLIONS AND BILLIONS of dollars spent on all this, one can ask what the return has been. (Say it in a Carl Sagan voice) There are plenty of good scientific projects. Further R&D on making space flight cheaper is a big deal. But space research itself needs to be seriously rethought.
If it is just minor stuff, why is the OSX patch so large? That suggests that a lot of this "minor stuff" was really improving OSX performance and compatibility.
How many people have used it for Carbon development under OSX? We've been thinking of developing some programs under Kylix/C++ Builder but if QT offers better crossplatform development we might head there. Any good links for a review?
While in theory it isn't that hard, in practice it can be. I've noticed that, oddly, the place where most have problems is in using proper Carbon events. Yet that is probably the #1 facet of the program that will affect performance and perceived fluidity of use.
I honestly don't think that most people would complain too much if a program was fairly good but didn't use all the features of the OS. I mean most people were happy with Quark that barely used any OS7 feature.
The big question really is how good Quark will be. After all the disparaging comments by the Quark head, I'm none to optimistic that they'll do that good a job. This is an excellent way for Adobe to gain market share.
Gimp does have a lot of features. However I find the interface kind of lame. Whether it'll do enough for you depends upon what you are using it for. I don't know the current applications Apple supplies with their various models, but my dual 867 came with numerous graphics programs that were able to cope with all my editing needs. When I needed more I found a good deal on Photoshop and haven't regreted it. I looked at Gimp, but wasn't that happy with it.
Others will disagree. But that's why I did mention both it and Open Office. Personally I'd never use either seriously. I installed Open Office just to get a feel for it and in the off chance it had good file filters. (It didn't)
It is sad they disappeared. They were often a great source to find odds and ends for older systems. For folks like computer unsavy older people, older Macs worked great. Most of them never use the Internet and just need simple word processing.
Having said all that though with eBay getting bigger and bigger it is often cheaper to find legacy machines there. Any business that depended upon legacy machines and legacy parts either would go out of business or just end up marketing with eBay. But it would be hard to keep the margins if you weren't buying large stocks of those parts.
I suspect it is in part to help derive survival equipment - but not likely. In general the technique is to not be in the same place at the same time. If both of you have peepers then you better hope that your companion didn't get buried and can dig you out soon. Black Diamond's
Avalung will help a bit in that it keeps the snow in front of your mouth from freezing. That means you can continue to breath while buried. (You suffocate because your breath freezes the snow in front of you) But I can't see what this guy's actions can do to improve those sorts of technologies.
More likely it is to just build up data on avalanche flow so backcountry skiers and climbers know what to predict. Any skier will dig pits and the like to get an idea of what is going on. This will possibly give even more information. In theory anyway - in practice I think the general knowledge is available right now. It is just that many want to go out in semi-safe or even unsafe conditions.
Actually quite a few people have asked me for suggestions on "how to" problems that end up needing Fink. For instance how do you open a.ps file in Preview? Easy if you have Ghostscript. You can even rig an Applescript so that doubleclicking on them opens them up in Preview. http://www.macosxhints.com/article.php?story=20021 125060127218
So I have to disagree with the utility bit. Yeah chances are people will never run an X11 app. Even when you ask people for good X11 apps to demo, they end up coming up with Open Office, AbiWord or Gimp. Yet there are far superior programs available for native OSX. When I tried to come up with progams I'd use, they typically already had versions with Aqua front ends. (i.e. GNUplot) I thought I'd use X11 a lot. However I tend to do the "gee whiz" try things out and then promptly forget about them.
For other Fink utilities though I've been pleasantly surprised. I really have used the shell for a lot of things. Some utilities are now standard on OSX. (i.e. Python) However they really do install a lot of useful utilities - especially if you are doing a lot of web work.
Dumb question, but for those programs that need configuration files and environment variables updated, does the binary version do this properly? Just curious. I'd assume yes but was curious of how they do this if they don't run the installer.
i.e. "make" to compile, "make install" to install. Or did I just answer my own question?
Actually until the mid 80's I believe that Canada had far, far more liberal gun laws than the United States. I remember my Dad actually bringing a.44 magnum carry on while flying Air Canada during the late 70's. He just asked the pilot if it was OK and they said it was fine so long as the pilot kept it in the cockpit.
That changed fast. The initial changes were fine and to a degree reasonable. However the changes of the last decade have been draconian and really are oriented towards getting rid of all handguns and even make hunting rifles much more difficult to own and use. What is worse is that people are often forced to "obey" contradictory laws with a "presumed guilty" tag put on them.
My father was actually going to try and ship all his guns down here to me, to get them out of Canada where presumably they'll eventually be banned. Unfortunately under Clinton the ATF has made it so difficult to import guns that he can't. (Even though all of them can be purchased in 5 minutes at most American stores)
Can't obtain in the sense of buy them there? Can't you just go to an other state and purchase them without problem? Or is it illegal to even own them in Massachusets? For instance what would happen to me if I moved there?
There is a lot of gun control in the northeast, for sure. I'm not sure it is fair to call that most of America though. I think most of America still has fairly liberal gun control laws.
As for the person who said that gun control in NYC corresponded with a decrease in crime, that doesn't sound right. Unless I'm mistaken NYC *always* had strong gun control. The decrease in crime tended to correspond more to the change in policing techniques the mayor and his chief imposed along with cleaning up the city. I don't think you can make a gun control argument there, especially given the change in cities that didn't enact the various policing strategies. Consider Washington DC for instance.
Apple's not entirely off with the Altivec. With all the graphics "gee whiz" features in OSX I understand it hits it fairly often. Of course that doesn't translate to your word processor or the like in that it is Apple's GUI that improves with Altivec. And even there it still is somewhat limited as the improvements with the recent moving processing to the graphics cards show. But for watching say Quicktime videos I think it actually does make a big difference. And that is something most people do.
You are right that other chips have similar code in them. Unless I'm mistaken though the Altivec really does improve things dramatically over the rated speed of the G4.
According to IBM the 970 is dual out the second half of 2003 and is roughly compatible in speed with *current* AMD/Intel offerings. Now IBM says they'll be upping the clock speed fairly quickly. Past experience suggests that IBM has done very well in this regard. (Look at the G3) However to asser that they're "all due out around the same time" is very misleading.
If you mean the AMD Hammer, you may be right. I don't know the time schedule on that chip nor its SPECMARKS off the top of my head. Whether the Hammer and 970 are roughly equal in performance is an other issue entirely.
The reasons are somewhat complex. Motorola does deserve a lot of blame. However there were also issues between IBM, Motorola and Apple regarding Altivec. The G3 was originally a Motorola chip and then IBM came on and improved it, doing an excellent job. It did not, however, have Altivec. Yet Altivec is used a lot by Apple, epecially as a way to try and tone down the PC/Mac divide. So Apple was stuck with a slowly improving G4 and a never produced G5.
Motorola's lack of delivery was for several reasons. The first were internal problems in the company. They really made numerous bad decisions and lost a lot of markets that they had owned. This led to financial issues which limited how much R&D they did. (Or so I am told) The other problem was that this was Apple's dark ages and a lot of people seriously questioned Apple's long term viability. Why apply resources into producing good chips that may never get used? Better markets were embedded systems and both IBM and Motorola seemed to push their chips in that direction.
At the same time AMD was producing good chips for the x86 leading to heightened competition. Intel had to improve and improve quickly to keep its edge. This led to some fantastic improvements in terms of price/performance on the PC side even while the Mac side was languishing.
Now things are changing. AMD has more or less conceded the desktop to Intel. They'll still make excellent chips but there won't be the degree of competition that we've seen the past 4 years or so. At the same time Apple has bounced back with OSX and IBM will be delivering the 970 chip which they claim is competitive with top P4 chips. The problem is that the 970 is coming out second half of 2003. What will happend to Apple in the meantime? No one is quite sure.
If Apple can produce better software then they can probably hold on to become competitive with XP based systems again. (I think that in the portable computer market they already are competitive, largely because of the way the PowerPC works in terms of power and heat.)
Apple has been on a buying spree the past year of high end video systems. Presumably they'll be emphasizing OSX versions of that software. So they absolutely need the 970 sooner than later. Further the 970 can multi-process better than the Intel chips. (I can't speed for the AMD MP chips) If Apple can deliver 970 systems on time with a batch of well written high end graphics and video software, then they have a shot at reclaiming this market niche.
This is the problem Apple faces. The 970 will hopefully improve this. Unfortunately the 970 will reportably be speed equivalent to the current top Intel chips but come out 6 - 10 months later. By then Intel will have improved for sure. Presumably Apple will issue dual 970 systems. They purportedly do SMP quite well. We'll see.
The bigger issue is that in the remaining 10 months before 970 systems come out Apple will be falling further and further behind in the markets they want to reconquer: video and graphics.
Slashdot Mirror
In either case the 970 with be in the same general class as the Opteron and will offer very good SMP. Apple's problem is that between now and perhaps September they are way behind the speed curve for desktop machines. (The portable market they do much better in, IMO)
For a great site with MST3K Sampo sounds and images visit http://www.painstick.com/mst_spotlight_earthfroze_ downloads.htm
OrobosX is pretty stable from what I've seen, and I've used it a fair bit. I admit that the past while I've been running KDE as I like Konquerer. However given how few X11 apps I actually run of late, I generally don't load X11 at all.
I find that while I run a lot of shell scripts, most of the things I used X11 for (GUI administrators) have Aqua counterparts or I can roll my own with Applescript and Applescript Studio.
I think that Unix gurus who learn Applescript will be pleasantly surprised at its power. There isn't much equivalent in Linux. The O'Reilly book on it is so-so though. However you are a Unix guru and can browse the web for example scripts. Check OSX Hints a lot. The latest version adds amazing power.
$60 for an aluminum case? Time to rent some shop time. . . Come on. That's ridiculous. The iPod is amazingly well designed and extremely sturdy feeling compared to other mp3 players. But $60?
Have you tried interlibrary loan? You can, for a very small fee, get photocopies of articles from journals other libraries subscribe to. I used to do that all the time while in college. You can actually get full books as well. If you are referencing articles in a paper but don't read them simply because they weren't in your library then that is very deceitful. Further it is problematic scientifically since you are ignoring a possibly significant piece of evidence. What if the paper in question undercut some of your methodology? You'd never know. . .
This is the real scandal. What is going on with peer review? Hopefully peer review entails more than just glances through the article and seeing if it is plausible. . .
The big debate was that the experiments were very hard to conduct and the evidence was from caloromic readings while the physicists wanted various nuclear reaction evidence. The problem was that while caloromic readings were very persuasive scientists had no idea what was going on inside the palladium.
Probably the best papers are by Michael McKubre over at Stanford. Here's a good transcript of his talk at this year's conference on cold fusion. He's written a lot on the topic and gave a fascinating hour long interview to KUER in SLC. I'll not post the link to avoid slashdotting a small NPR station. For peer reviewed papers I was looking at LANL, but the links I had appear to have been moved behind a firewall. HEre are some I found doing a bit of looking:
Anomalous Behavior of the Pd/D System
Correlation of Excess Power and Helium Production During D2O and H20 Electrolysis Using Palladium Cathods
The Emergence of a Coherent Explanation for Anomalies Observed in D/Pd and H/Pd System
Some Thoughts on the Nature of the Nuclear-Active Regions in Palladium
I'm sure there are more if you do a little searching.
There are a few ideas that I'd like to see Google ad, but these are interesting, if mostly proof of concepts. Froogle in particular is a good idea but doesn't allow price comparisons like MySimon does. Were they to add that it would be considerably more useful. Further it oddly seems to leave out a lot of stores, I noticed.
The web annotation tool is also a good idea. But I have a feeling that it needs to be better integrated with the google bar. (BTW - when is a Mac version coming?) Annotation has been discussed for quite a while. I have a wrox book at home on XML Metadata that discusses a lot of this. Unfortunately in practice it has never been applied to the general web.
There are lots of features of the original hypertext research in the 60's that are yet to be applied to the web. Google utilizes some concepts for its ranking, such as "links to" information. Wouldn't it be great to have a Google bar monitoriing your pages and make suggestings for helpful pages that link to the page you are on? There are many ideas like that which would improve ones browsing experience.
There are a few privacy concerns, but I'm exciting that Google offers these tests. The news site has become my favorite site on the web. It tends to get better stories than the other meta-news site I used to visit: DrudgeReport.com
One can argue that the advance in other areas of technology arose because funding at basic research was so good. It then had a "bubble up" effect.
Having said that though, lets be honest. Physics, even basic theoretical physics, was largely subsidized by the cold war. There were a lot of theoreticians doing what they liked on the side while getting paid for research in more applied areas. Further the biggest area of physics research right now is material science. And it isn't as if there is this huge funding shortage there.
So in a sense the poster is right. There is a bit of jealousy now that physics isn't quite the high rolling area it was up through the end of the cold war. However part of the problem is that in physics, most of the easy to answer problems are solved. What's left is [i]so[/i] complex and difficult and [i]so[/i] expensive that one has to be somewhat cautious in funding. I mean do we really need to spend a few billion dollars on the next collider to find the weight of the latest theoretical particle? Especially when the real work is in what is more fundamental. And there is still a ways to go before the superstring theorists and quantum loop theorists have much to give us that can be reasonably tested.
Having said all that one branch of physics/chemistry is about to make a comeback. Apparently a lot of the cold fusion stuff has been doing quite well the past few years. It has been duplicated in over 1000 different experiments. The remaining problems are in getting the proper impurities into palladium and doing some admittedly difficult material science work. (Which isn't to say one is even remotely close to making it a power source) So while the life sciences are deservedly getting the press today, energy problems along with global warming will push things back into the court of physicists and chemist within a few years.
The problem really is NASA as being more than simply a military shipping company.
To a degree all of this was just to help keep Soviet scientists around in Russia and not heading to the mid-east to develop nasty weapons. Further the military clearly had motives in keeping the Space Shuttle running. However now the Russians can't do much and haven't been able to move into commercial projects. Even in NASA the shuttles are wearing out with no replacements on the horizon.
The big question is whether all of these problems are a good thing or a bad thing. When you consider the BILLIONS AND BILLIONS of dollars spent on all this, one can ask what the return has been. (Say it in a Carl Sagan voice) There are plenty of good scientific projects. Further R&D on making space flight cheaper is a big deal. But space research itself needs to be seriously rethought.
How many people have used it for Carbon development under OSX? We've been thinking of developing some programs under Kylix/C++ Builder but if QT offers better crossplatform development we might head there. Any good links for a review?
I honestly don't think that most people would complain too much if a program was fairly good but didn't use all the features of the OS. I mean most people were happy with Quark that barely used any OS7 feature.
The big question really is how good Quark will be. After all the disparaging comments by the Quark head, I'm none to optimistic that they'll do that good a job. This is an excellent way for Adobe to gain market share.
Others will disagree. But that's why I did mention both it and Open Office. Personally I'd never use either seriously. I installed Open Office just to get a feel for it and in the off chance it had good file filters. (It didn't)
Having said all that though with eBay getting bigger and bigger it is often cheaper to find legacy machines there. Any business that depended upon legacy machines and legacy parts either would go out of business or just end up marketing with eBay. But it would be hard to keep the margins if you weren't buying large stocks of those parts.
More likely it is to just build up data on avalanche flow so backcountry skiers and climbers know what to predict. Any skier will dig pits and the like to get an idea of what is going on. This will possibly give even more information. In theory anyway - in practice I think the general knowledge is available right now. It is just that many want to go out in semi-safe or even unsafe conditions.
Whoops. Sorry about that. Should have previewed. The link was http://www.macosxhints.com/article.php?story=20021 125060127218
So I have to disagree with the utility bit. Yeah chances are people will never run an X11 app. Even when you ask people for good X11 apps to demo, they end up coming up with Open Office, AbiWord or Gimp. Yet there are far superior programs available for native OSX. When I tried to come up with progams I'd use, they typically already had versions with Aqua front ends. (i.e. GNUplot) I thought I'd use X11 a lot. However I tend to do the "gee whiz" try things out and then promptly forget about them.
For other Fink utilities though I've been pleasantly surprised. I really have used the shell for a lot of things. Some utilities are now standard on OSX. (i.e. Python) However they really do install a lot of useful utilities - especially if you are doing a lot of web work.
i.e. "make" to compile, "make install" to install. Or did I just answer my own question?
That changed fast. The initial changes were fine and to a degree reasonable. However the changes of the last decade have been draconian and really are oriented towards getting rid of all handguns and even make hunting rifles much more difficult to own and use. What is worse is that people are often forced to "obey" contradictory laws with a "presumed guilty" tag put on them.
My father was actually going to try and ship all his guns down here to me, to get them out of Canada where presumably they'll eventually be banned. Unfortunately under Clinton the ATF has made it so difficult to import guns that he can't. (Even though all of them can be purchased in 5 minutes at most American stores)
There is a lot of gun control in the northeast, for sure. I'm not sure it is fair to call that most of America though. I think most of America still has fairly liberal gun control laws.
As for the person who said that gun control in NYC corresponded with a decrease in crime, that doesn't sound right. Unless I'm mistaken NYC *always* had strong gun control. The decrease in crime tended to correspond more to the change in policing techniques the mayor and his chief imposed along with cleaning up the city. I don't think you can make a gun control argument there, especially given the change in cities that didn't enact the various policing strategies. Consider Washington DC for instance.
You are right that other chips have similar code in them. Unless I'm mistaken though the Altivec really does improve things dramatically over the rated speed of the G4.
If you mean the AMD Hammer, you may be right. I don't know the time schedule on that chip nor its SPECMARKS off the top of my head. Whether the Hammer and 970 are roughly equal in performance is an other issue entirely.
Motorola's lack of delivery was for several reasons. The first were internal problems in the company. They really made numerous bad decisions and lost a lot of markets that they had owned. This led to financial issues which limited how much R&D they did. (Or so I am told) The other problem was that this was Apple's dark ages and a lot of people seriously questioned Apple's long term viability. Why apply resources into producing good chips that may never get used? Better markets were embedded systems and both IBM and Motorola seemed to push their chips in that direction.
At the same time AMD was producing good chips for the x86 leading to heightened competition. Intel had to improve and improve quickly to keep its edge. This led to some fantastic improvements in terms of price/performance on the PC side even while the Mac side was languishing.
Now things are changing. AMD has more or less conceded the desktop to Intel. They'll still make excellent chips but there won't be the degree of competition that we've seen the past 4 years or so. At the same time Apple has bounced back with OSX and IBM will be delivering the 970 chip which they claim is competitive with top P4 chips. The problem is that the 970 is coming out second half of 2003. What will happend to Apple in the meantime? No one is quite sure.
If Apple can produce better software then they can probably hold on to become competitive with XP based systems again. (I think that in the portable computer market they already are competitive, largely because of the way the PowerPC works in terms of power and heat.)
Apple has been on a buying spree the past year of high end video systems. Presumably they'll be emphasizing OSX versions of that software. So they absolutely need the 970 sooner than later. Further the 970 can multi-process better than the Intel chips. (I can't speed for the AMD MP chips) If Apple can deliver 970 systems on time with a batch of well written high end graphics and video software, then they have a shot at reclaiming this market niche.
The bigger issue is that in the remaining 10 months before 970 systems come out Apple will be falling further and further behind in the markets they want to reconquer: video and graphics.