Investigation doesn't mean guilt, or that it was even a crime.
I didn't imply either of that. I was responding to a categorical statement "no crime was committed" by citing a long and ongoing investigation as proof that some very serious people do think that a crime may have been committed. As I put it another way, if it was so clear that no crime was committed, then the Bush government is wasting a lot of taxpayer money investigating it.
Joe Wilson admits that Plame was not covert when the story broke. Therefore, no crime was committed, therefore, jailed reporter is just being an idiot by not coughing up the source.
If there was clearly no crime committed, why is there an ongoing Justice Department investigation serious enough that John Ashcroft had to recuse himself from it? Why were Air Force One phone records subpoenaed? Either the Bush government likes to waste taxpayer money, or they are taking the matter pretty seriously, don't you think?
First of all, because nobody can follow the CSS spec 100%, especially since CSS 2.1 hasn't even reached Recommendation status. IE is the worst of the breed beyond any doubt, but it's not fair to ask it to achieve 100% when nobody else has.
Secondly, if you don't want to throw away your HTML engine and rewrite it from scratch, you are usually saddled with many old assumptions that predate CSS. HTML has plenty of special cases (not to mention bugs) that a browser like IE7 needs to continue supporting.
We can excuse free software
No, we cannot. The "free beer" aspect of free software was not supposed to be an impediment or excuse to missing features or low quality. In fact, free software was supposed to be better because people wrote it for the love of the art without real world constraints like time or money.
May I suggest a slight modification to your statement? How about "I code to W3C standards, except where what I need to accomplish can't be done within W3C, or the standard solution won't work in IE, in which case I code to IE and document the deviation"? IE is the poster boy of non-compliance, but IE6 with the proper DOCTYPE is usable, so the far more interesting question is why you might disagree with my version.
Also, which IE? I have written pages that work correctly in Firefox, Safari, Opera, and IE6, but not in IE5.5.
I think I'd have to recommend keeping how's very close to the actual implementation, because what can otherwise happen is that code will be modified but the comments remain as before which can be tr?s confusing in the future!
That would be ideal. The problem is that in most real cases, the cost of maintaining the "how" comments becomes prohibitive, and they lapse into the confusion you fear. In some cases, such comments are worse than not having any.
As I've said, if your code is not clear and readable enough for a competent maintainer to figure out the "how" based on the code, aided by the "what" and "why" comments, it's already a pretty bad sign. High level computer languages, by virtue of their precision, is often a better tool than English to describe the "how".
a) What you're doing.
b) Why you're doing it.
c) How you're doing it.
This is good in theory, but in practice the "how" will become obsolete much more quickly than the "what" or "why". The "how" is also redundant for most cases, because hopefully the maintainer can understand the code as it was written. Thus, I would have to recommend documenting the "how" only for the most confusing or subtle code. Put another way, if you really have to document the "how", consider a different solution.
Of course it's unfair. One group gets to force the citizens to pay for what it has produced while another group must rely on their voluntary purchase.
So the US Postal Service is in unfair competition against FedEx and UPS? Or perhaps the police are in unfair competition against security guards? How about public schools putting private schools out of business? (Note that there are examples of successful businesses in every one of the above.)
Like it or not, there will always be some overlap between the public and private sectors, and in a democracy, voters decide where that overlap ends. Perhaps you would like to rephrase your objections.
I doubt you would like to compete against a government that can simply take whatever money it needs from it's "customers".
What I don't assume to have is a God-given right to make a profit doing any particular thing. I think a pay-per-extinguish service instead of a public fire department might be profitable (especially when several neighbors bid for my services while the houses burn). That doesn't mean the government's fire department is unfairly competing against me, does it?
Here's how it might go if I was the one writing the compiler. The first version is meant to achieve correctness. This means that I will write the simplest (if slowest) code to make sure that my compiler and test suite give me the correct results. The speed is irrelevant because I don't ever intend to use that code for one of our real CPUs. Next, for each supported CPU I'll write a version that is optimal, essentially bumping my original slow code to the 'else' section of a long strand of 'if' statements.
This has the net effect of a compatible but undetected CPU falling into the slow 'else' case. Sure, they might complain, but it doesn't do me enough good to fix that and test it on competing CPUs. I might lose some customers, but I'll lose a great deal more by actively helping my competitor.
While I sympathize with AMD, it's plainly unreasonable to force Intel to support AMD chips with the same vigor as they do their own. I think a nice compromise would be for Intel to accept input from AMD (which must do its own testing of the Intel compiler) indicating which of AMD's CPUs map exactly to which Intel CPU. AMD would then shoulder all support expenses of using Intel's compiler on AMD chips, particularly if AMD's mapping was not entirely correct. I doubt AMD will get a better deal (in practice) than that even if they win in court.
Then you are not fundamentalist. You are a person who is trying to understand the universe through religion. But just for the sake of argument:
1. The existence and complexity of the universe
You are making a leap of faith here, assuming that complexity requires design. But I don't want to argue about that.
2. The accuracy of the biblical record
The bible contains many fantasies. For example, do you believe that the children of Adam and Eve are incestuous? Also, do you believe in Noah's Ark, literally? Do you believe that Moses parted the Red Sea? These are all major events in the Bible that are unlikely to be literally accurate.
3. The evidence of the changed lives of the people who knew Jesus personally
How many people's lives changed when they encountered Buddhism, Islam, or maybe Hinduism?
4. The evidence of the way that my relationship with Jesus Christ has changed my life.
Again, nothing I wish to argue. If you're interested enough, you can see that the points I don't argue cover what science simply doesn't know (and maybe cannot know).
Naturalists and creationists share the same evidence, and we each have things that we cannot fully explain.
I wasn't talking about a naturalist or creationist, but a fundamentalist. A fundamentalist can fully explain everything. As I've said, their knowledge of the truth is both absolute and complete.
Based on our current knowledge of science, all that is needed for that fertilized egg to grow up like you or me is to have food and shelter.
Well, not quite. For an arbitrary embryo to grow up like me, he or she would have to be born into a "regular" family and be able to afford to go to college. Also, in many parts of the world, food and shelter are not available. It takes quite a bit of resources.
Is it ok to kill them for the furthering of scientific goals?
Yes, I think so. If you believe in God, then you must believe that he made one egg cell per month per woman. Roughly guessing that we have three billion women and twenty years of fertility each, we're talking 720 billion eggs (and God knows how many trillion sperms) per century or so. I don't think you can argue that most of them were not intended to die unproductively. That is, I know about "fill the earth and multiply", but 700 billion people on this earth should be ridiculous.
So if God intended for them to die, why is it immoral to "recycle" them to improve the life that had been produced?
This is really not comparable to the heliocentric issue.
I was responding to your "moral guidelines", which sought to quash scientific discovery. I can't be sure, but I'd say it's likely that they were at least as sure of their moral superiority as you are of yours when they persecuted Copernicus. Point is, morality frequently lags, because it tends to be controlled by conservative and slower-moving entities. So how do we know it's not just lagging right now?
I would certainly be described as one of those "Fundamental" Christians, and I must respectfully take issue with your point.
People who believe as I do do not deny the evidence. We collect evidence and draw inferences from it to see how that fits into our view of how the universe works.
Either you ignore contrarian evidence, or you are not fundamentalist. A fundamentalist is not open to new evidence or ideas, because he or she is already in possession of absolute and complete truth. You cannot reason with a fundamentalist, by definition, and they cannot lose an argument because there isn't an argument.
Anybody who is willing to accept that he or she may be wrong about God is not fundamentalist. (There's a big difference between thinking you are right and not accepting that you may be wrong.)
Embryonic stem cell research is different.
Not any different from letting the embryo die any other way. A fertilized egg that dies because of birth control pills is literally flushed down the toilet, and "leftover" embryos from artificial insemination are not exactly killed with any more ceremony. Would your opinion change if there was a way to take such an embryo from certain death into a lab, instead of fertilizing an egg specifically for research? The morality of such an act seems akin to harvesting body parts from the braindead, as an embryo has no discernible brain.
I guess the question is, is experimenting on embryos absolutely immoral, or only when embryos are created for experimentation?
Science has long existed in a realm where there were moral guidelines on appropriate research.
You mean like how the first folks who figured out that the sun didn't revolve around the earth were accused of heresy? Where would we be if science always stayed within moral guidelines?
Big companies (such as Adobe) won't have problems finding PPC computers to test their programs in. This is really a small investment in exchange for all of the advantages they get by building Universal binaries.
You did not understand what I wrote. Getting Photoshop to compile on a PPC is trivial, but getting a PPC Photoshop to be ready to put in a box for sale is not.
But it's not such a big difference from the current situation where they are not able to test their programs in different PPC Mac models.
Even if you stick with gcc on both CPUs, you'll be using an entirely different code generator backend, which will have different bugs. If you use Intel's excellent compiler instead, then it's a different compiler altogether. Changing compilers is not usually a negligible risk.
Apple is pushing developers a lot to build universal binaries, so I really think that they must have worked a lot to ensure this goal.
I'm sure Apple will work very hard to make the transition as smooth as possible. I'm just pointing out to you that simply getting the code to compile for both CPUs is trivial next to getting it to ship on both.
I agree with your conclusion, but not your reasoning.
It's not compiling for PPC that will be difficult. It's testing it sufficiently to achieve release quality that is expensive. Very small vendors might just make a build and have users test it for them, but vendors with reputations usually need more assurance than "it compiles". I don't think Photoshop or Word will ever ship with "use at your own risk" PPC binaries.
But as you said, it's not at all stupid to support the PPC. This is because even when the number of PPC Macs sold drops to zero, there is still a big installed base of PPC Macs to sell to. If you look at the MacOS 8/9 section of versiontracker.com, you'll see that there are at least five (on a Sunday!) software updates a day for the past five days, when MacOS 9 was discontinued in 2002 and and the last Mac that could boot MacOS 9 was discontinued in June 2004.
First of all, you forgot about Carbon. A lot of apps are still using it.
Secondly, just because GNUstep apps run in OS X doesn't mean that the reverse is true. From the GNUstep developer FAQs:
It's easier from GNUstep to Cocoa than Cocoa to GNUstep. Cocoa is constantly changing, much faster than GNUstep could hope to keep up. They have added extensions and new classes that aren't available in GNUstep yet.
MacOS X also has various frameworks for audio, video, disc burning, etc. So no, it will definitely be a non-trivial task, along the lines of WINE.
Hell even the iPod... what truly separates it from other players? Yeah the hardware is good, and it looks slick, but it's the interface.
Yet one key aspect of that "interface" is the scroll wheel, which is hardware, just as one key aspect of the original Macintosh GUI is the hardware mouse. I find these hardware versus software arguments silly, because to me Apple is a company that is able to solve problems either in hardware or software. Therefore they are both.
Both Linux and MacOSX will run fine on supported hardware but Linux supports a lot more hardware. How exactly does that make MacOSX better?
Because of the difference in definitions of the word "supported". In MacOS, that word usually means "auto-detected, driver already present or on companion CD-ROM, plug-and-play". In Linux, it can mean exactly the same, or it can mean "look online, read config file comments, experiment, deal with lack of meaningful error messages" and more.
In the end, whether you value time or money more is entirely your own decision, and the people who find the Mac "better" probably value their time more. You don't have to agree, but it probably helps to understand why.
It's easy for geeks to criticize what we know, but look at the engineering marvel that is the Leaning Tower of Pisa. Too long ago? Well, Frank Lloyd Wright's famous Fallingwatersagged and needed millions of dollars worth in repairs. That same house had an original budget of $30,000, but ultimately cost $155,000. From the article I linked to:
As the house was being built, the problem with Wright's design was obvious to the Pittsburgh engineering concern, Metzger-Richardson. Mr. Silman said the company doubled the number of one-inch-square bars in each beam from what had been called for in Wright's design.
But Wright, hearing of the change, wrote to Kaufmann complaining: "I have put so much more into this house than you or any other client has a right to expect, that if I don't have your confidence -- to hell with the whole thing." Kaufmann expressed his confidence, but the contractor's extra steel stayed.
If it had not, Mr. Silman said, the building would have collapsed already.
I've been to Fallingwater, and it's an awesome sight. But just like computer science, architecture is still learning all the time.
As does your conclusion that suddenly Apple switching to Intel will cause any change to the Mac sales.
You are entirely wrong here, as I made no conclusion. In fact, I specifically wrote: "note that I'm not making the opposite assertion. I'm pointing out that we lack data", which you apparently missed.
Are you telling me that a switch to Intel is suddenly going to cause a massive influx of new users?
No, I'm saying we don't know yet, and that Solaris is not much of a precedent guide.
What advantage does Intel have over the other supposed advantages mentioned above is is going to cause this jump?
Potential ability to dual-boot Windows, or run Windows in VM at an acceptable speed.
Potentially cheaper boxes.
Potentially faster boxes (important for high-end apps).
Will this add up to a "massive influx of new users"? I have no idea. I'm just pointing out your conclusions seem to be based on the assumption that these are non-factors, rather than facts.
Solaris, free or not, doesn't add much value over what Linux offers in terms of applications. That may be why few Linux users bothered switching to Solaris x86. MacOS X x86 offers Word, Photoshop, etc., which is more than what Solaris offers.
Now, it is entirely possible that you are right that very few Linux users would prefer to use Word or Photoshop or some other MacOS X application. What I don't know is where you get the data to make that assumption so early on.
Note that I'm not making the opposite assertion. I'm pointing out that we lack data.
If someone wants to run those applications, they are already on the Mac (or Windows
Similarly, this is a lapse in logic. If they move to a PPC Mac, they lose certain advantages of the x86 (such as dual booting). If they dual boot Windows, and MacOS allows them to not dual boot (or at least not as often), then there's considerable value in time saved. MacOS X x86 allows the former to retain whatever advantage they wanted with the x86, and the latter to avoid or decrease the need to dual boot. Why would you assume that this means nothing to users?
Either way, my entire point is that the success or failure of MacOS X cannot be predicted from that of Solaris. Your conclusions depends on unproven assumptions.
All the advantages/disadvantages of Linux over OSX remain constant regardless of what processor Apple chooses for their systems.
I'm not arguing that MacOS X is going to be a real threat to Linux on the desktop, but this point is probably wrong. If the Mac will dual-boot Windows, as reports have said it will, then the Mac is equalizing what some (think games, for one) would feel is an advantage of x86 Linux over PPC MacOS X. It's hard to say how big that market is at this point, but it's almost certainly not zero. Another sub-market that may be interested are those who just can't make do with VirtualPC, assuming VMWare begins to support the Mac.
I know the backup and support for is going to be superior to any that Apple (single vendor) will give me. [...] If I were a network admin or a CIO, I'd be looking at being vendor free as much as possible.
Your world view may be entirely valid, but it is not consistent with many companies I know of. Many people who are put in charge of information systems feel more confident with vendor support.
I still don't get the hooplah over this in terms of Linux usage. I see the two as being very distinct.
The unanswered question right now is how many people run Linux on the desktop because they are somehow tied to the x86 (Windows games, etc). This is the portion of the market that might decide that MacOS X is more comfortable (which includes all sorts of reasons, including the availability of Office and Photoshop) than Linux. Whatever Slashdot likes to think, a good number of people do prefer those commercial apps over free alternatives, and are willing to pay for them (or would rather violate copyright).
Similarly, a (likely very small) portion of MacOS X users are there because of the PowerPC CPU, and they might switch to Linux if that is no longer the case.
People who use Linux for philosophical reasons will obviously not switch just because Apple changed CPUs.
it has a lot of ugly nextstep-isms though, including the heavy leanings toward objc and bundles.
Assuming we're talking about the same thing, bundles are a compromise between the need for easy installation of applications and "flat" file systems. What would you have preferred under those requirements?
Also, do you have a more objective criticism than "ugly"? I'm not saying you don't, but many times it's just another word for "unfamiliar to me".
They included it as a component of their case to establish _criminal intent_. In other words, if he's hiding something, he knows he has something to hide, therefore he knows he's doing something wrong...
No, it would be wrong to conclude criminal intent based on the fact that his drive is encrypted. The prosecution should have to establish that he otherwise would not have encrypted his drive, because lots of people encrypt their drives for little or no reason. It's a few clicks away in MacOS X, for example.
Consider two MacOS X users who both accidentally download the same illegal pornographic picture. One of them has File Vault enabled. Should he therefore be treated as having criminal intent by virtue of that single fact?
Let's blame trucks for bringing illegal immigrants across borders, hyperdermic needles for heroin use, beer cans for alcohol abuse, cameras for pornography, voice boxes for the rise in bad language and linear time for people getting older and dying.
Trucks also carry food to starving refugees. Needles are also used to inject lifesaving drugs. Cameras are aos used to record vacations, prevent cheating in casinos, and deter robbery of convenience stores. Voices are also used to sing and say pleasant things.
So to complete your argument, the overwhelming positive and pre-dominant use of BitTorrent (which I am by no means suggesting does not exist) would be...?
Slashdot Mirror
I didn't imply either of that. I was responding to a categorical statement "no crime was committed" by citing a long and ongoing investigation as proof that some very serious people do think that a crime may have been committed. As I put it another way, if it was so clear that no crime was committed, then the Bush government is wasting a lot of taxpayer money investigating it.
If there was clearly no crime committed, why is there an ongoing Justice Department investigation serious enough that John Ashcroft had to recuse himself from it? Why were Air Force One phone records subpoenaed? Either the Bush government likes to waste taxpayer money, or they are taking the matter pretty seriously, don't you think?
First of all, because nobody can follow the CSS spec 100%, especially since CSS 2.1 hasn't even reached Recommendation status. IE is the worst of the breed beyond any doubt, but it's not fair to ask it to achieve 100% when nobody else has.
Secondly, if you don't want to throw away your HTML engine and rewrite it from scratch, you are usually saddled with many old assumptions that predate CSS. HTML has plenty of special cases (not to mention bugs) that a browser like IE7 needs to continue supporting.
We can excuse free software
No, we cannot. The "free beer" aspect of free software was not supposed to be an impediment or excuse to missing features or low quality. In fact, free software was supposed to be better because people wrote it for the love of the art without real world constraints like time or money.
May I suggest a slight modification to your statement? How about "I code to W3C standards, except where what I need to accomplish can't be done within W3C, or the standard solution won't work in IE, in which case I code to IE and document the deviation"? IE is the poster boy of non-compliance, but IE6 with the proper DOCTYPE is usable, so the far more interesting question is why you might disagree with my version.
Also, which IE? I have written pages that work correctly in Firefox, Safari, Opera, and IE6, but not in IE5.5.
That would be ideal. The problem is that in most real cases, the cost of maintaining the "how" comments becomes prohibitive, and they lapse into the confusion you fear. In some cases, such comments are worse than not having any.
As I've said, if your code is not clear and readable enough for a competent maintainer to figure out the "how" based on the code, aided by the "what" and "why" comments, it's already a pretty bad sign. High level computer languages, by virtue of their precision, is often a better tool than English to describe the "how".
a) What you're doing.
b) Why you're doing it.
c) How you're doing it.
This is good in theory, but in practice the "how" will become obsolete much more quickly than the "what" or "why". The "how" is also redundant for most cases, because hopefully the maintainer can understand the code as it was written. Thus, I would have to recommend documenting the "how" only for the most confusing or subtle code. Put another way, if you really have to document the "how", consider a different solution.
So the US Postal Service is in unfair competition against FedEx and UPS? Or perhaps the police are in unfair competition against security guards? How about public schools putting private schools out of business? (Note that there are examples of successful businesses in every one of the above.)
Like it or not, there will always be some overlap between the public and private sectors, and in a democracy, voters decide where that overlap ends. Perhaps you would like to rephrase your objections.
I doubt you would like to compete against a government that can simply take whatever money it needs from it's "customers".
What I don't assume to have is a God-given right to make a profit doing any particular thing. I think a pay-per-extinguish service instead of a public fire department might be profitable (especially when several neighbors bid for my services while the houses burn). That doesn't mean the government's fire department is unfairly competing against me, does it?
This has the net effect of a compatible but undetected CPU falling into the slow 'else' case. Sure, they might complain, but it doesn't do me enough good to fix that and test it on competing CPUs. I might lose some customers, but I'll lose a great deal more by actively helping my competitor.
While I sympathize with AMD, it's plainly unreasonable to force Intel to support AMD chips with the same vigor as they do their own. I think a nice compromise would be for Intel to accept input from AMD (which must do its own testing of the Intel compiler) indicating which of AMD's CPUs map exactly to which Intel CPU. AMD would then shoulder all support expenses of using Intel's compiler on AMD chips, particularly if AMD's mapping was not entirely correct. I doubt AMD will get a better deal (in practice) than that even if they win in court.
Then you are not fundamentalist. You are a person who is trying to understand the universe through religion. But just for the sake of argument:
1. The existence and complexity of the universe
You are making a leap of faith here, assuming that complexity requires design. But I don't want to argue about that.
2. The accuracy of the biblical record
The bible contains many fantasies. For example, do you believe that the children of Adam and Eve are incestuous? Also, do you believe in Noah's Ark, literally? Do you believe that Moses parted the Red Sea? These are all major events in the Bible that are unlikely to be literally accurate.
3. The evidence of the changed lives of the people who knew Jesus personally
How many people's lives changed when they encountered Buddhism, Islam, or maybe Hinduism?
4. The evidence of the way that my relationship with Jesus Christ has changed my life.
Again, nothing I wish to argue. If you're interested enough, you can see that the points I don't argue cover what science simply doesn't know (and maybe cannot know).
Naturalists and creationists share the same evidence, and we each have things that we cannot fully explain.
I wasn't talking about a naturalist or creationist, but a fundamentalist. A fundamentalist can fully explain everything. As I've said, their knowledge of the truth is both absolute and complete.
Based on our current knowledge of science, all that is needed for that fertilized egg to grow up like you or me is to have food and shelter.
Well, not quite. For an arbitrary embryo to grow up like me, he or she would have to be born into a "regular" family and be able to afford to go to college. Also, in many parts of the world, food and shelter are not available. It takes quite a bit of resources.
Is it ok to kill them for the furthering of scientific goals?
Yes, I think so. If you believe in God, then you must believe that he made one egg cell per month per woman. Roughly guessing that we have three billion women and twenty years of fertility each, we're talking 720 billion eggs (and God knows how many trillion sperms) per century or so. I don't think you can argue that most of them were not intended to die unproductively. That is, I know about "fill the earth and multiply", but 700 billion people on this earth should be ridiculous.
So if God intended for them to die, why is it immoral to "recycle" them to improve the life that had been produced?
This is really not comparable to the heliocentric issue.
I was responding to your "moral guidelines", which sought to quash scientific discovery. I can't be sure, but I'd say it's likely that they were at least as sure of their moral superiority as you are of yours when they persecuted Copernicus. Point is, morality frequently lags, because it tends to be controlled by conservative and slower-moving entities. So how do we know it's not just lagging right now?
People who believe as I do do not deny the evidence. We collect evidence and draw inferences from it to see how that fits into our view of how the universe works.
Either you ignore contrarian evidence, or you are not fundamentalist. A fundamentalist is not open to new evidence or ideas, because he or she is already in possession of absolute and complete truth. You cannot reason with a fundamentalist, by definition, and they cannot lose an argument because there isn't an argument.
Anybody who is willing to accept that he or she may be wrong about God is not fundamentalist. (There's a big difference between thinking you are right and not accepting that you may be wrong.)
Embryonic stem cell research is different.
Not any different from letting the embryo die any other way. A fertilized egg that dies because of birth control pills is literally flushed down the toilet, and "leftover" embryos from artificial insemination are not exactly killed with any more ceremony. Would your opinion change if there was a way to take such an embryo from certain death into a lab, instead of fertilizing an egg specifically for research? The morality of such an act seems akin to harvesting body parts from the braindead, as an embryo has no discernible brain.
I guess the question is, is experimenting on embryos absolutely immoral, or only when embryos are created for experimentation?
Science has long existed in a realm where there were moral guidelines on appropriate research.
You mean like how the first folks who figured out that the sun didn't revolve around the earth were accused of heresy? Where would we be if science always stayed within moral guidelines?
You did not understand what I wrote. Getting Photoshop to compile on a PPC is trivial, but getting a PPC Photoshop to be ready to put in a box for sale is not.
But it's not such a big difference from the current situation where they are not able to test their programs in different PPC Mac models.
Even if you stick with gcc on both CPUs, you'll be using an entirely different code generator backend, which will have different bugs. If you use Intel's excellent compiler instead, then it's a different compiler altogether. Changing compilers is not usually a negligible risk.
Apple is pushing developers a lot to build universal binaries, so I really think that they must have worked a lot to ensure this goal.
I'm sure Apple will work very hard to make the transition as smooth as possible. I'm just pointing out to you that simply getting the code to compile for both CPUs is trivial next to getting it to ship on both.
It's not compiling for PPC that will be difficult. It's testing it sufficiently to achieve release quality that is expensive. Very small vendors might just make a build and have users test it for them, but vendors with reputations usually need more assurance than "it compiles". I don't think Photoshop or Word will ever ship with "use at your own risk" PPC binaries.
But as you said, it's not at all stupid to support the PPC. This is because even when the number of PPC Macs sold drops to zero, there is still a big installed base of PPC Macs to sell to. If you look at the MacOS 8/9 section of versiontracker.com, you'll see that there are at least five (on a Sunday!) software updates a day for the past five days, when MacOS 9 was discontinued in 2002 and and the last Mac that could boot MacOS 9 was discontinued in June 2004.
Secondly, just because GNUstep apps run in OS X doesn't mean that the reverse is true. From the GNUstep developer FAQs:
MacOS X also has various frameworks for audio, video, disc burning, etc. So no, it will definitely be a non-trivial task, along the lines of WINE.
Yet one key aspect of that "interface" is the scroll wheel, which is hardware, just as one key aspect of the original Macintosh GUI is the hardware mouse. I find these hardware versus software arguments silly, because to me Apple is a company that is able to solve problems either in hardware or software. Therefore they are both.
Because of the difference in definitions of the word "supported". In MacOS, that word usually means "auto-detected, driver already present or on companion CD-ROM, plug-and-play". In Linux, it can mean exactly the same, or it can mean "look online, read config file comments, experiment, deal with lack of meaningful error messages" and more.
In the end, whether you value time or money more is entirely your own decision, and the people who find the Mac "better" probably value their time more. You don't have to agree, but it probably helps to understand why.
I've been to Fallingwater, and it's an awesome sight. But just like computer science, architecture is still learning all the time.
You are entirely wrong here, as I made no conclusion. In fact, I specifically wrote: "note that I'm not making the opposite assertion. I'm pointing out that we lack data", which you apparently missed.
Are you telling me that a switch to Intel is suddenly going to cause a massive influx of new users?
No, I'm saying we don't know yet, and that Solaris is not much of a precedent guide.
What advantage does Intel have over the other supposed advantages mentioned above is is going to cause this jump?
- Potential ability to dual-boot Windows, or run Windows in VM at an acceptable speed.
- Potentially cheaper boxes.
- Potentially faster boxes (important for high-end apps).
Will this add up to a "massive influx of new users"? I have no idea. I'm just pointing out your conclusions seem to be based on the assumption that these are non-factors, rather than facts.Solaris, free or not, doesn't add much value over what Linux offers in terms of applications. That may be why few Linux users bothered switching to Solaris x86. MacOS X x86 offers Word, Photoshop, etc., which is more than what Solaris offers.
Now, it is entirely possible that you are right that very few Linux users would prefer to use Word or Photoshop or some other MacOS X application. What I don't know is where you get the data to make that assumption so early on.
Note that I'm not making the opposite assertion. I'm pointing out that we lack data.
If someone wants to run those applications, they are already on the Mac (or Windows
Similarly, this is a lapse in logic. If they move to a PPC Mac, they lose certain advantages of the x86 (such as dual booting). If they dual boot Windows, and MacOS allows them to not dual boot (or at least not as often), then there's considerable value in time saved. MacOS X x86 allows the former to retain whatever advantage they wanted with the x86, and the latter to avoid or decrease the need to dual boot. Why would you assume that this means nothing to users?
Either way, my entire point is that the success or failure of MacOS X cannot be predicted from that of Solaris. Your conclusions depends on unproven assumptions.
Solaris doesn't run Word, Photoshop, iLife, Final Cut Pro and Apple's other pro apps.
I'm not arguing that MacOS X is going to be a real threat to Linux on the desktop, but this point is probably wrong. If the Mac will dual-boot Windows, as reports have said it will, then the Mac is equalizing what some (think games, for one) would feel is an advantage of x86 Linux over PPC MacOS X. It's hard to say how big that market is at this point, but it's almost certainly not zero. Another sub-market that may be interested are those who just can't make do with VirtualPC, assuming VMWare begins to support the Mac.
Your world view may be entirely valid, but it is not consistent with many companies I know of. Many people who are put in charge of information systems feel more confident with vendor support.
I still don't get the hooplah over this in terms of Linux usage. I see the two as being very distinct.
The unanswered question right now is how many people run Linux on the desktop because they are somehow tied to the x86 (Windows games, etc). This is the portion of the market that might decide that MacOS X is more comfortable (which includes all sorts of reasons, including the availability of Office and Photoshop) than Linux. Whatever Slashdot likes to think, a good number of people do prefer those commercial apps over free alternatives, and are willing to pay for them (or would rather violate copyright).
Similarly, a (likely very small) portion of MacOS X users are there because of the PowerPC CPU, and they might switch to Linux if that is no longer the case.
People who use Linux for philosophical reasons will obviously not switch just because Apple changed CPUs.
Assuming we're talking about the same thing, bundles are a compromise between the need for easy installation of applications and "flat" file systems. What would you have preferred under those requirements?
Also, do you have a more objective criticism than "ugly"? I'm not saying you don't, but many times it's just another word for "unfamiliar to me".
No, it would be wrong to conclude criminal intent based on the fact that his drive is encrypted. The prosecution should have to establish that he otherwise would not have encrypted his drive, because lots of people encrypt their drives for little or no reason. It's a few clicks away in MacOS X, for example.
Consider two MacOS X users who both accidentally download the same illegal pornographic picture. One of them has File Vault enabled. Should he therefore be treated as having criminal intent by virtue of that single fact?
Trucks also carry food to starving refugees. Needles are also used to inject lifesaving drugs. Cameras are aos used to record vacations, prevent cheating in casinos, and deter robbery of convenience stores. Voices are also used to sing and say pleasant things.
So to complete your argument, the overwhelming positive and pre-dominant use of BitTorrent (which I am by no means suggesting does not exist) would be...?
No, it's not a good thing, but the US does have significant weapons on earth that can deter an enemy from attacking its assets in space.