Re:I don't agree with the article
on
A Better Finder?
·
· Score: 1
It's not just that people have high expectations from Apple. It's also that Apple just completely rotated its feature set while keeping the user base the same - and the expectations of those users. With OS9, users had a certain UI which was really, really strong in a lot of ways. OSX brings a number of enhancements to that UI (the dock, aqua), but also makes it a lot worse in a lot of ways (the dock, aqua - and read the article). A large subset of Apple users are going to notice these failures and complain about them. Many aspects of OSX did this, BTW. OSX is faster than OS9 (multitasking), but it's also a lot slower (actually using anything, especially graphically intensive things). OSX has much better memory management (VM doesn't suck), but it's also a little worse (VM can't be turned off and is used in situations where it shouldn't be). The system became much more open (Darwin is open source) but it also became much more closed (the amount of stuff you can customize, at least for now, is a lot less - display control, input device control, lots of functionality provided by 3rd party extensions in OS9 land).
This is not comparable to anything that happened elsewhere recently. Linux 2.4 did not come out missing a lot of what Linux 2.2 offerred (well, maybe it did in terms of stability, but I don't think the situation is comparable). Windows XP did not come out missing a lot of what Windows 95/98/ME offered. Now, if Microsoft had come out with Win98 and WinME and promoted them as replacements for NT 4, that would have been comparable. But they knew better than to do that. Apple was smart for not forking their OS into 2 products (a move Microsoft has only really recovered from with XP, but even so it's still forked much more than OSX/OSX Server are).
This *is* comparable to the 68k to PPC transition. PPC Macs were a lot faster, but they were also a lot slower since everything had to be emulated. However, that transition really wasn't that bad. And since it's just speed, it's easy enough to let Moore's law take care of that problem with time. Except for the part where Motorolla suspended Moore's law for PPCs for a year in 1999, but I digress...
Labels: Worse than useless, at least in the incarnation we know from OS9. Better systems can be devised, as the myriad workflow tools in existence have shown us.
I think you're trolling. Labels being worse than useless would imply that just having them around without using them is a detriment to your system. They have essentially no overhead to speak of so this is patently false.
On the contrary, labels are insanely useful. Even though I don't use them very often, I say they're insanely useful because there is no way to replicate a similar behavior without them (i.e. in OSX, in Windows, or in standard DOS or UNIX shells).
I want to be able to assign a personal tag to a file or a group of files in a directory. I want that tag to have its own namespace and not interfere with the other "metadata" of that file (whatever that may mean in your context). Invading the filename namespace is totally unacceptable and kludgy. The purpose of this tag will be to let me do at least the following two things. It will let me sort a directory by an arbitrary criteria, placing certain items before others, based on categories I have assigned that may have nothing to do with anything else about the file. I used to sort my System Folder this way, back when that was a reasonable thing to do (System 7 through MacOS 9). It will also let me put a long-lasting "selection" tag on some files. Lets say I have a group of files that I want to do a series of operations with. I don't want to change the directory structure of the stuff I'm working with, and I don't want to archive the files together (as in tarring them). I want to email them to someone. Then I want to move them to a different drive. Then I want to delete the originals. Then I want to edit each of the files and put a note in them saying I've sent them to their destination and archived them. I can't do this with the normal Finder selection ability since I might want to do something else in between which would cause that selection to be forgotten about. All of this can be done easily because the Finder provided me a way to mark a bunch of files for <whatever the hell i want>ing. Having a centralized and consistent way to do all this is much less kludgy than any alternative.
I see Apple's OS X as the best option for these kind of insane resolutions, with its built-in Display PostScript (a.k.a. Quartz) handling everything.
FWIW, Quartz is a PDF engine. Display PostScript is a remnant from the days of Rhapsody.
But more relevantly, MacOS X makes essentially no effort to ever scale display resolution to actual monitor resolution in dpi. I think this is a major failing. I think MacOS X basically assumes my monitor is displaying 72 dpi, even though my CRT is closer to (and would be much easier to read if it used) ~105 dpi.
Of course, this completely ignores the relative operational overhead of each protocol. Which is irrelevant since I can almost guarentee your firewire is going to be running over PCI or some PCI-like bus anyway.
The original article was in the context of courts of law. A court has to judge the truth of an accusation, even if it can't be scientifically proved. It would be nice if every case could be proved scientifically, but that's not reality. Society can't afford to ignore all questions that are outside of the limits of science.
Your mistake is in defining science and what it can do too narrowly. A judge has a number of scientific tools available. Mainly forensic science, of course, but there's also a lot of other social and criminal science which is totally applicable in the court room: psychology, sociology, criminology, political science, etc. A judge's job gets easier because judges don't have to prove anything - they have to come to a decision based on evidence and law (and a jury). This can be a scientific process or not. To discount all social phenomenon as non-scientific because they aren't based in physical science is unreasonable.
And for the record, I think society would do a lot better to reject the romantic idea that there are realities "outside the limits of science", which is not to say we should reject ideas we don't have the means to understand scientifically, but to reject the notion of transcendence.
That's nice, but unfortunately you didn't actually justify a need for tabs at all.
In any decent browser, there's a shortcut to open a link in a new window, but in the background, or behind the current window. This is very useful. And - unlike with tabbed browsing - you CAN organize your windows. As opposed to having one page which shares the same position and size as all of your other pages (which are of course not simultaneously viewable either).
Market. Share. Is. Not. Necessarily. An. Indication. Of. A. Company's. Success.
Why can't people understand this? Why do they keep clinging to notions that have been disproved time and time again, are intuitively wrong, and yet people still believe them?
Because in a market of commodities, market share is a pretty decent indication of future success. Time and time again, personal computers haven't been commodities. Today, they're a lot closer. In 5 years, they'll be even closer. Apple has not recognized this and since 1998 has tried nothing but to give itself and its products distinguishing characteristics and brand identity. In effect, they're trying to de-commoditize their market. Unfortunately they're up against a tidal wave and they're going to fail (which is not to say they're going out of business) unless they remember to compete on the same terms as everyone else in the personal computer industry: speed, price, quality, value, etc.
For example, you're claiming that apps should establish locks on network-resident files. That's fine and dandy, until some app goes to the great computer in the sky with a file locked.
WTF? The operating system has no ability to handle this condition and expire locks from dead processes? What year is this?
Finally, I hate to say it, but if you're trying to set up a collaboration system (as it sounds like you are), some sort of dedicated checkout system is probably better. For text files on UNIX, this usually means cvs, but other platforms have their own favored revision control/collaboration systems.
While CVS has its uses, on the operating system level, this is not a solution. Running a CVS server on every system for single-user application work is not a solution, it's just ignoring the problem.
but I really think that UNIX's approach is pretty good. You have the ability to use *both* types of locking, and the default is the one that lets the user do the most -- so dead apps don't tie up files.
Encouraging users to haphazardly write to files and corrupt them does not count as "letting them do the most". Dead apps are no excuse for anything, see above.
Yes, although it's only slightly harder when you do. Consider that the mouse is just about as well handled in MacOS X as it was in System 1 through MacOS 9.
(is there a single application that uses all 15 F-keys on a Mac keyboard??)
Some of the default shortcuts in WordPerfect 3.5 utilized F13, F14, and F15. This was most inconvenient for myself since with the advent of the iMac keyboard (which Apple sold exclusively during the reign of the Blue G3), Mac Keyboards went straight from F12 to 'help' 'home' 'pg up' 'pg down'.
BTW, the iMac keyboard is a highly underrated keyboard, and other than the lack of keys and trendy caps lock (I'm talking about delete and end here, not the missing F-keys), it's really excellent. Which is to say it had really good key action and layout.
And it goes on to mention issues with Solaris. Nothing about Java itself being inherently problematic, just issues with certain implementation.
Well, actually, you're wrong, and it does. Specifically, points 3 and 4 of the memo relate to problems in the way Sun handles and supports new versions of the Java platform: extensions/modules and backwards compatibility in new releases. Of course, cowmix's summary was totally inaccurate and trollful, but you're not any better.
How did it come about that you did the movie Free Enterprise? How did the filmmakers approach you about doing the project? Were you enthusiastic about it? How much of your 'character' was you and how much of it was parody?
I read his criticism of the "Quit" function several times and still don't understand his gripe. Yes, our computers can multi-thread so we can run multiple applications, but they have limited memory so we can't run EVERYTHING at once. And I for one would rather control which ones are running rather than wondering what my computer chose to quit.
Reality is moving in the direction of not having limited memory, and MacOS X acts as such. Memory is cheap. Disk is cheaper. Having the right combination of RAM and Disk available at the right time is still not so simple. But it's no longer nearly as necessary as it used to be to expose the implementation and assign the user the task of choosing which pieces of code need to be in memory. MacOSX does this by having single-windowed applications NOT quit when their window is closed (there are exceptions...). With an ideal OS(1), launching an application for the first time is very much like re-activating a swapped-out application. With an ideal VM system(2), swapping in and out memory is clean and efficient. You don't need to care what it's actually doing. But there are other reasons to keep 'quit' around. For one, having an application loaded is not necessarily a zero-impact circumstance for the rest of the system even if it's not in use. Also, dock space is limited, and currently an application is either shown in the dock or it's not loaded.
(1)MacOS X isn't perfect, although 10.2 is a lot better.
(2)MacOS X's VM system is far from ideal.
Of course Apple is a monopoly under any sensible definition of the word. If the only requirement for not being a monopoly in the OS vending market was to have at least 2 other people willing to sell you an OS, Microsoft would not be a monopoly - but the DOJ has found that they are, and rightfully so. Ford is not a monopoly, but not because there are other car vendors. Ford is not a monopoly because you can buy a car from another vendor which will run on all the same roads, on the same unleaded fuel, etc., and Ford can't say anything about it.
In a lot of ways Apple's monopoly is worse than Microsoft's. If the DOJ case against Microsoft is successful (i.e., MS is punished with something more severe than "fines"), and the politics of the DOJ move in the direction of less conservative (note: I don't predict this will happen in the near future), Apple will be up next for investigation and prosecution for its monopolistic practices. However, it's certainly arguable that Apple is not nearly as malicious with it's monopoly as Microsoft has been, and therefore does not need as strict a punishment.
Let's say that a monopoly is "exclusive or near-exclusive control of a market, service, or commodity by a single group". This seems like a reasonable definition to me. Microsoft has a monopoly on OSes that run on x86-based personal computer systems. They use this monopoly to leverage to their advantage:
Pricing and availability of x86 OSes
Selection and licensing of OSes shipped on new x86 PCs
Selection and licensing of applications shipped on new x86s PCs
Just as certainly as Microsoft, Apple has a monopoly on:
PowerPC-based personal computer systems
OSes that run on PowerPCs
Computers that run MacOS and MacOS X
Apple uses this monopoly to leverage to their advantage:
Availability and pricing of PowerPC hardware
Availability and pricing of MacOS
Selection and licensing of applications shipped on new Macs
Still don't believe it? Microsoft abused their monopoly and as a result it became very difficult to compete in the web browser market. It wasn't enough to beat them on quality or value, because Microsoft controlled the market, and they could throw their weight around and crush you. Same with office applications and x86 desktop operating systems. Imagine the fun you'd have if you decided you wanted to make a mp3 player app for MacOS. How could you compete with iTunes? Or if you wanted to build and sell PPC-7400 (G4) systems. Apple could make sure you can't get any if they perceived you as a threat to their market. Or if you wanted to make a desktop OS which would compete directly with MacOS X. Do you think Apple would let you bundle it on new Macs? Do you think you'd have any recourse if they didn't?
Undergraduate entrance exams make sense; they verify that you have the prerequisite knowledge to take (pretty standard) undergraduate courses.
On the contrary, undergraduate entrance exams do not make sense because (okay, I'm only talking about the SAT here - limited experience), because they are so easy that all you need to do is be a good "test taker" and you can own them. Being a good test taker basically means approaching every question with the attitude that you can get the right answer because one of the choices is right and you've been exposed to the information needed to find it at some point in the past (by 9th grade, 10th grade at the latest). Little actual thinking and knowledge are required to get any question on the SAT correct.
As a result, anyone can do well on an undergraduate entrance exam, not necessarily people who are either smart or a good student. GREs are a little different in that the material is "hard", but they're still multiple guess. The requirement of having a hard multiple choice problem means that most problems have to fall into one of a few patterns of "ways you can make a multiple choice question hard." So again, good test takers can own the tests. Except this time, the amount of knowledge required is higher, indicating that the taker may have had to learn something as an undergrad. This means the test does a better job of measuring if the taker is a good student: a person who can readily learn what they do not already know (and work hard). Remember, "smart" people are not necessarily what a school desires. Graduate schools are a little more keen to this and therefore don't require the GRE or subject GREs as often as they require SATs/ACTs.
For those of you who use VC++, I would encourage you to set aside perhaps 2 weeks where you compile both on VC++ and gcc. You'll be stunned at the number of errors that gcc will catch but VC++ will let slip through. Lord only knows what the VC++ compiled code is actually doing...
And I challenge you to set aside 2 weeks and find developers of commercial software applications who use VC++ and who care. Not agreeing or disagreeing with you here.
You can read the original piece by the brilliant Steve Albini here, and probably lots of other places. Thanks to some slashdot comment I read last week but have since lost.
Bonker's point (other than being a troll) was that Apple had to write an interpreter/emulator for 68k apps. This is true. They did. It was not provided by IBM/Motorola.
I agree (regarding Apple not taking any extra effor to support OS9 for future hardware tech), but it seems likely that the 1/2003 date was chosen due to Apple's desire to kill off OS9 sooner rather than later, not because upcoming hardware has a technical reason for not working.
Apple is not explicitly doing anything to the hardware to prevent OS 9 booting that could otherwise occur. They are simply not bothering to update OS 9 to boot on the new hardware, since it is legacy code.
Do you have any evidence to support your assertion that a) Apple isn't intentionally crippling their future hardware so as to not run OS9 and b) Apple's next model, circa 2003, includes significant architecture changes such that OS9 will not boot it? There is no reason to believe that OS9 would have to be updated to support the 2003 Macs. There are many examples of new systems Apple has released that did not require new versions of MacOS to run. For example, the latest Dual G4s.
Apple even watched it happen once before, when they switched from 68k architecture, to PowerPC. They ended up having to include an interperater/emulator for 68k apps in later versions of Mac0S.
To which scout.finch wrote:
This is completely untrue.. PowerPC machines emulated 68K code from the start, which is often lauded as one of the most graceful computer transition in the industry's short history. Completely transparent, totally useable.
Completely untrue is your implication that PowerPCs emulate 68K code. Apple wrote an emulator - a GOOD one - and included it in System 7 (7.5, I believe). By 7.5.3 they basically had it working perfectly, and quickly. So yes, they DID have to write an emulator. Fortunately, they were migrating between two well-designed architectures, and they did a good job. The emulator persists in some form today in 9.2.2. As a result they lost essentially no users or applications in the 68k Mac -> PowerMac transition. The forced-OSX transition (for new hardware) is not quite comparable, since there are a number of applications which have no equivalents which will run in OSX or acceptably under Classic [but I will leave that discussion to someone else].
Slashdot Mirror
It's not just that people have high expectations from Apple. It's also that Apple just completely rotated its feature set while keeping the user base the same - and the expectations of those users. With OS9, users had a certain UI which was really, really strong in a lot of ways. OSX brings a number of enhancements to that UI (the dock, aqua), but also makes it a lot worse in a lot of ways (the dock, aqua - and read the article). A large subset of Apple users are going to notice these failures and complain about them. Many aspects of OSX did this, BTW. OSX is faster than OS9 (multitasking), but it's also a lot slower (actually using anything, especially graphically intensive things). OSX has much better memory management (VM doesn't suck), but it's also a little worse (VM can't be turned off and is used in situations where it shouldn't be). The system became much more open (Darwin is open source) but it also became much more closed (the amount of stuff you can customize, at least for now, is a lot less - display control, input device control, lots of functionality provided by 3rd party extensions in OS9 land).
This is not comparable to anything that happened elsewhere recently. Linux 2.4 did not come out missing a lot of what Linux 2.2 offerred (well, maybe it did in terms of stability, but I don't think the situation is comparable). Windows XP did not come out missing a lot of what Windows 95/98/ME offered. Now, if Microsoft had come out with Win98 and WinME and promoted them as replacements for NT 4, that would have been comparable. But they knew better than to do that. Apple was smart for not forking their OS into 2 products (a move Microsoft has only really recovered from with XP, but even so it's still forked much more than OSX/OSX Server are).
This *is* comparable to the 68k to PPC transition. PPC Macs were a lot faster, but they were also a lot slower since everything had to be emulated. However, that transition really wasn't that bad. And since it's just speed, it's easy enough to let Moore's law take care of that problem with time. Except for the part where Motorolla suspended Moore's law for PPCs for a year in 1999, but I digress...Labels: Worse than useless, at least in the incarnation we know from OS9. Better systems can be devised, as the myriad workflow tools in existence have shown us.
I think you're trolling. Labels being worse than useless would imply that just having them around without using them is a detriment to your system. They have essentially no overhead to speak of so this is patently false.
On the contrary, labels are insanely useful. Even though I don't use them very often, I say they're insanely useful because there is no way to replicate a similar behavior without them (i.e. in OSX, in Windows, or in standard DOS or UNIX shells).
I want to be able to assign a personal tag to a file or a group of files in a directory. I want that tag to have its own namespace and not interfere with the other "metadata" of that file (whatever that may mean in your context). Invading the filename namespace is totally unacceptable and kludgy. The purpose of this tag will be to let me do at least the following two things. It will let me sort a directory by an arbitrary criteria, placing certain items before others, based on categories I have assigned that may have nothing to do with anything else about the file. I used to sort my System Folder this way, back when that was a reasonable thing to do (System 7 through MacOS 9). It will also let me put a long-lasting "selection" tag on some files. Lets say I have a group of files that I want to do a series of operations with. I don't want to change the directory structure of the stuff I'm working with, and I don't want to archive the files together (as in tarring them). I want to email them to someone. Then I want to move them to a different drive. Then I want to delete the originals. Then I want to edit each of the files and put a note in them saying I've sent them to their destination and archived them. I can't do this with the normal Finder selection ability since I might want to do something else in between which would cause that selection to be forgotten about. All of this can be done easily because the Finder provided me a way to mark a bunch of files for <whatever the hell i want>ing. Having a centralized and consistent way to do all this is much less kludgy than any alternative.
I see Apple's OS X as the best option for these kind of insane resolutions, with its built-in Display PostScript (a.k.a. Quartz) handling everything.
FWIW, Quartz is a PDF engine. Display PostScript is a remnant from the days of Rhapsody.
But more relevantly, MacOS X makes essentially no effort to ever scale display resolution to actual monitor resolution in dpi. I think this is a major failing. I think MacOS X basically assumes my monitor is displaying 72 dpi, even though my CRT is closer to (and would be much easier to read if it used) ~105 dpi.
I'm almost certain that it is faster. [Comparing Firewire To PCI]
Um, no. Not even close.
Firewire = 1-bit * 400MHz (or 800MHz now) = 400-800 MbpsPCI = 32-bit (or 64-bit) * 33MHz (or 66MHz) = 1056-4224Mbps
Of course, this completely ignores the relative operational overhead of each protocol. Which is irrelevant since I can almost guarentee your firewire is going to be running over PCI or some PCI-like bus anyway.
The original article was in the context of courts of law. A court has to judge the truth of an accusation, even if it can't be scientifically proved. It would be nice if every case could be proved scientifically, but that's not reality. Society can't afford to ignore all questions that are outside of the limits of science.
Your mistake is in defining science and what it can do too narrowly. A judge has a number of scientific tools available. Mainly forensic science, of course, but there's also a lot of other social and criminal science which is totally applicable in the court room: psychology, sociology, criminology, political science, etc. A judge's job gets easier because judges don't have to prove anything - they have to come to a decision based on evidence and law (and a jury). This can be a scientific process or not. To discount all social phenomenon as non-scientific because they aren't based in physical science is unreasonable.
And for the record, I think society would do a lot better to reject the romantic idea that there are realities "outside the limits of science", which is not to say we should reject ideas we don't have the means to understand scientifically, but to reject the notion of transcendence.
I have yet to find any use for truth (or especially big T Truth) in my life.
Fact (observations) and belief have done me just fine, so far. Science can't do anything with unobservable reality, but neither can I.
Thus science does pretty good for me.
That's nice, but unfortunately you didn't actually justify a need for tabs at all.
In any decent browser, there's a shortcut to open a link in a new window, but in the background, or behind the current window. This is very useful. And - unlike with tabbed browsing - you CAN organize your windows. As opposed to having one page which shares the same position and size as all of your other pages (which are of course not simultaneously viewable either).
Why can't people understand this? Why do they keep clinging to notions that have been disproved time and time again, are intuitively wrong, and yet people still believe them?
Because in a market of commodities, market share is a pretty decent indication of future success. Time and time again, personal computers haven't been commodities. Today, they're a lot closer. In 5 years, they'll be even closer. Apple has not recognized this and since 1998 has tried nothing but to give itself and its products distinguishing characteristics and brand identity. In effect, they're trying to de-commoditize their market. Unfortunately they're up against a tidal wave and they're going to fail (which is not to say they're going out of business) unless they remember to compete on the same terms as everyone else in the personal computer industry: speed, price, quality, value, etc.
For example, you're claiming that apps should establish locks on network-resident files. That's fine and dandy, until some app goes to the great computer in the sky with a file locked.
WTF? The operating system has no ability to handle this condition and expire locks from dead processes? What year is this?
Finally, I hate to say it, but if you're trying to set up a collaboration system (as it sounds like you are), some sort of dedicated checkout system is probably better. For text files on UNIX, this usually means cvs, but other platforms have their own favored revision control/collaboration systems.
While CVS has its uses, on the operating system level, this is not a solution. Running a CVS server on every system for single-user application work is not a solution, it's just ignoring the problem.
but I really think that UNIX's approach is pretty good. You have the ability to use *both* types of locking, and the default is the one that lets the user do the most -- so dead apps don't tie up files.
Encouraging users to haphazardly write to files and corrupt them does not count as "letting them do the most". Dead apps are no excuse for anything, see above.
Yes, although it's only slightly harder when you do. Consider that the mouse is just about as well handled in MacOS X as it was in System 1 through MacOS 9.
And yet, it's still nowhere close to as fast or smooth cursor movement as a 10 year old Macintosh.
(is there a single application that uses all 15 F-keys on a Mac keyboard??)
Some of the default shortcuts in WordPerfect 3.5 utilized F13, F14, and F15. This was most inconvenient for myself since with the advent of the iMac keyboard (which Apple sold exclusively during the reign of the Blue G3), Mac Keyboards went straight from F12 to 'help' 'home' 'pg up' 'pg down'.
BTW, the iMac keyboard is a highly underrated keyboard, and other than the lack of keys and trendy caps lock (I'm talking about delete and end here, not the missing F-keys), it's really excellent. Which is to say it had really good key action and layout.
And it goes on to mention issues with Solaris. Nothing about Java itself being inherently problematic, just issues with certain implementation.
Well, actually, you're wrong, and it does. Specifically, points 3 and 4 of the memo relate to problems in the way Sun handles and supports new versions of the Java platform: extensions/modules and backwards compatibility in new releases. Of course, cowmix's summary was totally inaccurate and trollful, but you're not any better.
Whoa! It's like reading MacOSRumors!
How did it come about that you did the movie Free Enterprise? How did the filmmakers approach you about doing the project? Were you enthusiastic about it? How much of your 'character' was you and how much of it was parody?
I read his criticism of the "Quit" function several times and still don't understand his gripe. Yes, our computers can multi-thread so we can run multiple applications, but they have limited memory so we can't run EVERYTHING at once. And I for one would rather control which ones are running rather than wondering what my computer chose to quit.
Reality is moving in the direction of not having limited memory, and MacOS X acts as such. Memory is cheap. Disk is cheaper. Having the right combination of RAM and Disk available at the right time is still not so simple. But it's no longer nearly as necessary as it used to be to expose the implementation and assign the user the task of choosing which pieces of code need to be in memory. MacOSX does this by having single-windowed applications NOT quit when their window is closed (there are exceptions...). With an ideal OS(1), launching an application for the first time is very much like re-activating a swapped-out application. With an ideal VM system(2), swapping in and out memory is clean and efficient. You don't need to care what it's actually doing. But there are other reasons to keep 'quit' around. For one, having an application loaded is not necessarily a zero-impact circumstance for the rest of the system even if it's not in use. Also, dock space is limited, and currently an application is either shown in the dock or it's not loaded.
(1)MacOS X isn't perfect, although 10.2 is a lot better.
(2)MacOS X's VM system is far from ideal.
...Or they could just sell themselves their own unreliable hard drives, and the books practically cook themselves!
Of course Apple is a monopoly under any sensible definition of the word. If the only requirement for not being a monopoly in the OS vending market was to have at least 2 other people willing to sell you an OS, Microsoft would not be a monopoly - but the DOJ has found that they are, and rightfully so. Ford is not a monopoly, but not because there are other car vendors. Ford is not a monopoly because you can buy a car from another vendor which will run on all the same roads, on the same unleaded fuel, etc., and Ford can't say anything about it.
In a lot of ways Apple's monopoly is worse than Microsoft's. If the DOJ case against Microsoft is successful (i.e., MS is punished with something more severe than "fines"), and the politics of the DOJ move in the direction of less conservative (note: I don't predict this will happen in the near future), Apple will be up next for investigation and prosecution for its monopolistic practices. However, it's certainly arguable that Apple is not nearly as malicious with it's monopoly as Microsoft has been, and therefore does not need as strict a punishment.
Let's say that a monopoly is "exclusive or near-exclusive control of a market, service, or commodity by a single group". This seems like a reasonable definition to me. Microsoft has a monopoly on OSes that run on x86-based personal computer systems. They use this monopoly to leverage to their advantage:
- Pricing and availability of x86 OSes
- Selection and licensing of OSes shipped on new x86 PCs
- Selection and licensing of applications shipped on new x86s PCs
Just as certainly as Microsoft, Apple has a monopoly on:- PowerPC-based personal computer systems
- OSes that run on PowerPCs
- Computers that run MacOS and MacOS X
Apple uses this monopoly to leverage to their advantage:Still don't believe it? Microsoft abused their monopoly and as a result it became very difficult to compete in the web browser market. It wasn't enough to beat them on quality or value, because Microsoft controlled the market, and they could throw their weight around and crush you. Same with office applications and x86 desktop operating systems. Imagine the fun you'd have if you decided you wanted to make a mp3 player app for MacOS. How could you compete with iTunes? Or if you wanted to build and sell PPC-7400 (G4) systems. Apple could make sure you can't get any if they perceived you as a threat to their market. Or if you wanted to make a desktop OS which would compete directly with MacOS X. Do you think Apple would let you bundle it on new Macs? Do you think you'd have any recourse if they didn't?
Undergraduate entrance exams make sense; they verify that you have the prerequisite knowledge to take (pretty standard) undergraduate courses.
On the contrary, undergraduate entrance exams do not make sense because (okay, I'm only talking about the SAT here - limited experience), because they are so easy that all you need to do is be a good "test taker" and you can own them. Being a good test taker basically means approaching every question with the attitude that you can get the right answer because one of the choices is right and you've been exposed to the information needed to find it at some point in the past (by 9th grade, 10th grade at the latest). Little actual thinking and knowledge are required to get any question on the SAT correct.
As a result, anyone can do well on an undergraduate entrance exam, not necessarily people who are either smart or a good student. GREs are a little different in that the material is "hard", but they're still multiple guess. The requirement of having a hard multiple choice problem means that most problems have to fall into one of a few patterns of "ways you can make a multiple choice question hard." So again, good test takers can own the tests. Except this time, the amount of knowledge required is higher, indicating that the taker may have had to learn something as an undergrad. This means the test does a better job of measuring if the taker is a good student: a person who can readily learn what they do not already know (and work hard). Remember, "smart" people are not necessarily what a school desires. Graduate schools are a little more keen to this and therefore don't require the GRE or subject GREs as often as they require SATs/ACTs.
For those of you who use VC++, I would encourage you to set aside perhaps 2 weeks where you compile both on VC++ and gcc. You'll be stunned at the number of errors that gcc will catch but VC++ will let slip through. Lord only knows what the VC++ compiled code is actually doing...
And I challenge you to set aside 2 weeks and find developers of commercial software applications who use VC++ and who care. Not agreeing or disagreeing with you here.
You can read the original piece by the brilliant Steve Albini here, and probably lots of other places. Thanks to some slashdot comment I read last week but have since lost.
Bonker's point (other than being a troll) was that Apple had to write an interpreter/emulator for 68k apps. This is true. They did. It was not provided by IBM/Motorola.
I agree (regarding Apple not taking any extra effor to support OS9 for future hardware tech), but it seems likely that the 1/2003 date was chosen due to Apple's desire to kill off OS9 sooner rather than later, not because upcoming hardware has a technical reason for not working.
Apple is not explicitly doing anything to the hardware to prevent OS 9 booting that could otherwise occur. They are simply not bothering to update OS 9 to boot on the new hardware, since it is legacy code.
Do you have any evidence to support your assertion that a) Apple isn't intentionally crippling their future hardware so as to not run OS9 and b) Apple's next model, circa 2003, includes significant architecture changes such that OS9 will not boot it? There is no reason to believe that OS9 would have to be updated to support the 2003 Macs. There are many examples of new systems Apple has released that did not require new versions of MacOS to run. For example, the latest Dual G4s.
Apple even watched it happen once before, when they switched from 68k architecture, to PowerPC. They ended up having to include an interperater/emulator for 68k apps in later versions of Mac0S.
To which scout.finch wrote:This is completely untrue.. PowerPC machines emulated 68K code from the start, which is often lauded as one of the most graceful computer transition in the industry's short history. Completely transparent, totally useable.
Completely untrue is your implication that PowerPCs emulate 68K code. Apple wrote an emulator - a GOOD one - and included it in System 7 (7.5, I believe). By 7.5.3 they basically had it working perfectly, and quickly. So yes, they DID have to write an emulator. Fortunately, they were migrating between two well-designed architectures, and they did a good job. The emulator persists in some form today in 9.2.2. As a result they lost essentially no users or applications in the 68k Mac -> PowerMac transition. The forced-OSX transition (for new hardware) is not quite comparable, since there are a number of applications which have no equivalents which will run in OSX or acceptably under Classic [but I will leave that discussion to someone else].