Jesus. I can't believe you're still talking about this.
For the vast majority of what people do -- that is, the common functions which everyone does with a computer....
Your list is woefully incomplete. Let me take the least-computer-savvy person I know: my secretary. She's even less into computers than my mother is. She owns a computer for doing basic tasks, like email and web surfing, and she also uses it to write her family's holiday letter, complete with color photos taken on her digital camera. She prints the letters on her color inkjet printer and mails 'em out. I believe she uses something like AppleWorks for this, but I couldn't say for sure, because she doesn't know either. She just talks about using the computer: downloading pictures off of her camera, cropping them and doing simple color corrections ("I made it more colorful," she likes to say), page layout, illustration, printing, it's all just "the computer" to her.
She would be unable to do these things with Linux. The tools to do them simply don't exist. If you think all the average computer user wants to do is download MP3s and play DVDs, I think you're being a bit naive. In fact, I don't actually know anybody who downloads MP3s-- that generally being illegal-- or who uses their computer for DVD playback, with the exception of watching movies on planes. My understanding is that there are no legal DVD playback applications for Linux anyway, so that point is moot, unless my knowledge is out of date.
For typical home use, the best-suited is probably GNUcash.
No online banking, no recurring payments, inadequate reporting, no financial planning features to speak of, no integration with external data sources for things like stock portfolio management... yes, Gnucash is certainly well-suited.
Don't think just because I don't like or respect Linux as a desktop operating system that I'm uninformed. The main disconnect between us seems to be that I know what you're talking about, but that you have no idea what I'm talking about.
Another fact about Linux which is very commonly overlooked is that you gets tons of software with any major distribution of Linux. A whole slew more than you'll ever need, and orders of magnitude more than you get along with Windows and MacOS.
If none of it is useful, though, then all that software only exists to confuse and annoy. Red Hat 7.3 comes with something like six different shells, and no page layout programs. Amazing.
In Linux, if one process crashes, that doesn't effect the system; it does in Windows.
Except that it doesn't in Windows. Or have you never used Windows 2000 or Windows XP? Neither of these OS's is 100% crash-proof, but crashes are incredibly unlikely, roughly on par with kernel panics in Linux. The same is true of OS X.
The other reason to be running Windows/Mac programs in an emulator when needed is to take advantage of the powerful features of Linux w/o having to reboot.
What features are those, exactly? And which of them exist in Linux, but don't exist in OS X? OS X is a full-fledged UNIX operating system, recall. At any point I can drop out of the graphical UI and work in a fully POSIX-compliant command-line UI environment. Without, as you say, having to reboot.
Tasks that most people want to accomplish an be done in all three OS'.
Please explain how my secretary can create her family's holiday letter using Linux. Please explain how I can create the marketing data sheets for my company's products using Linux. Please explain how my girlfriend can enjoy her photography hobby using Linux. You are completely unaware, evidently, of how most people use computers.
I disagree b/c Linux is used for big-time special effects
Linux is primarily used as a renderer. In other words, as a computation server. I don't know anybody-- with the exception of one guy who's a freelance animator-- who owns his own render farm. So that's kinda silly of you to say.
In those instances where Linux is being used on the desktop, it's either for running other third-party UNIX applications-- such as Maya or Softimage-- on cheap hardware, or it's for running proprietary UNIX applications on cheap hardware. ILM is currently moving all of their compositors from SGI workstations to Linux workstations, because once they got all the bugs out of the first one, they were able to mass-produce them really cheaply. The software in that case is CompTime. Again, this has nothing to do with anybody but ILM. It's not a reflection of how suitable Linux is for doing typical creative work at all.
but there's also things that one can do in Linux that one can't do elsewhere (scientific applications...
Like which scientific applications, exactly? And why do they matter to people who use personal computers?
complete customizability...
Linux advocates often fall back on this argument. What they fail to realize is that practically nobody cares about customizability. That's like saying that your computer is better because the RAM is purple. Doesn't matter. In fact, it's even worse than that, because customizability can-- not always, but sometimes-- get in the way of a usable user interface. The world is full of trade-offs, and if programmers spend more time making their applications customizable than they do making them usable, then you end up with loads and loads of highly customizable, really crappy software. Which is how I would describe software like KDE, Gnome, and Mozilla: thoroughly executed, embarrassingly designed.
games -- Linux is nowhere near MS in that category, and if your honest, neither is MacOS
That's funny, I would have sworn that Warcraft III released on both Mac OS and Windows on the same day. How odd. I must have been mistaken.
Saying Linux is at best comparable to a subset of OS X is insulting.
Well, gee, dude, earlier in your message you compared Linux directly to FreeBSD. Mac OS X is a superset of FreeBSD 4.4. Ergo...?
One could say that OS X is only comparable to a subset of Linux because of all the scientific apps available in Linux with no equivalent in OS X.
Name one. I'll download and compile it in short order. If it's a commercial application-- although I'm pretty sure most of the companies who sell commercial software for Linux have gone out of business-- then it's a trivial matter for the vendor to port.
Sounds like Mac OS X is functionally a superset of Linux, to me.
You are simply trying to close off an avenue of debate by saying that Linux shouldn't even be an option for serious consideration, as it doesn't allow one to do all of the things that one can do in MacOS; but as I just demonstrated, that argument can be turned on its head.
Yes, I am most certainly saying that Linux shouldn't be an option for serious consideration, and just for the reason you give. However, I fail to see how you have turned any argument on its head, exactly. Maybe I missed it. You should go over it again.
I choose Linux because of the many quality apps that come with it for free and because of its stability & speed; as well as the fact that it doesn't require you to accept any draconian EULA's.
Ah, finally we get down to the heart of it. Inevitably, whenever people try to discuss Linux on its merits, they end up falling back to politics. Licensing agreements are simply not a factor. One either agrees to them and abides by them, or one doesn't. It's really simple. Linux advocates, on the other hand, seem to be opposed to licensing agreements in principle. Principle is all fine and good for things like voting booths and church confessionals. It has little to do with one's choice of personal computer.
Sounds to me, dh003i, like all of my original arguments stand, and all you have to fall back on is your political views. Does that pretty much sum it up?
First of all, I understand the difference between OS and application very well. But I have news for you. For practical purposes, there is no difference. This is an important fact of life that Linux advocates often fail to understand.
One chooses an OS because of what one can do with it, which is a function of the applications. If two OS's can offer the same or equivalent applications for a given task, then the OS's are equivalent for that task, and you can make your choice between them based on preference or minor features or whatever.
But if two operating systems do not offer the same or equivalent applications for a given task, then they are not equivalent for that task. You can't run an enterprise-class DBMS under Mac OS 9, and you can't play Warcraft under OS/400. So comparing Mac OS 9 to OS/400 for running a database would be pointless, just as it would for running Warcraft.
In large part, Mac OS X and Windows XP are equivalent; you can do the same things on one as you can on the other. Some things are easier on Mac OS X-- running Apache/Tomcat, for example, is trivial-- while other things are easier on Windows XP-- trying to use the Mac version of Outlook under OS X's Classic mode is painful at best. But despite the differences, the two platforms are functionally equivalent, so comparing them makes sense.
Linux is not comparable to OS X or to Windows. There are simply too many basic tasks that one might want to do with one's computer that cannot be done under Linux. You say you've never used Illustrator or InDesign; that explains, in part, why you don't understand my point. There is no Linux equivalent to either of these programs, just as there's no Linux equivalent to Quicken or Microsoft Access or After Effects or Quark XPress or Crystal Reports... the list goes on and on.
If your best suggestion is to run the Windows versions of these programs thorough an emulator, virtual machine, or other similar technique, then you really ought to be running Windows. It'll be more reliable.
I stand by my argument: OS X and Linux aren't comparable because they cannot both be used to accomplish the same set of tasks. At best, Linux is comparable to a subset of OS X, which is no meaningful comparison at all.
There are equivalent appllications in Linux to everything you've mentioned...
What's the Linux equivalent of Illustrator? Or of InDesign? There are none at all. How about the Linux equivalent of PowerPoint? Or Quicken? If you believe there are Linux equivalents of these programs, you are mistaken.
You seem to be implying that its impossible to run many applications at once under Linux.
What? No, I didn't. I implied-- hell, I flat-out said-- that the applications I use are not all available on any platform other than OS X. That's as simple as it gets.
These OS' do the same things, just in different ways.
Um... nobody really cares what an OS does, okay? People care about what they can do. Using my computer, I can accomplish everything I need to do. This would not be true if I used Linux, because the tools I need to use simply don't exist there. Ergo, OS X and Linux do not "do the same things, just in different ways." There are things you can do on OS X that simply can't be done under Linux.
A successful company-- and IBM is probably the textbook definition of a successful company-- has no desire to gouge its customers. IBM could, if they wanted, aggressively sell DB2 to small companies that can't really afford it. But those companies would end up being unsatisfied customers, which would hurt IBM more than the revenue of the sale helped them.
Business relationships between vendor and customer are more like symbiosis than parasitism. Keep that in mind here.
Um... dude... you're talking about ease-of-use, and I'm talking about pure functionality. There no comparison between OS X and Linux. Sorry to burst your bubble.
On my Mac, I typically run the following: OmniWeb, Mail.app, Project Builder, Interface Builder, Apache/Tomcat, Word, Excel, PowerPoint, Palm Desktop, iCal, iTunes, XEmacs, Maya, Illustrator, Photoshop, InDesign, VirtualPC (for IE and Outlook), OmniOutliner, Quicken, Acrobat, iChat, and, sometimes, Warcraft III.
This is not possible under Linux. At all, with any combination of hardware and software. Ergo, there is no comparison. It's like comparing a toothbrush to a boxing glove. They just don't do the same things, so talking about them as if they do is just a waste of time.
When it comes to web servers in mid-range companies, MySQL and DB2 are competitors.
No, they're not. Because they're not equivalent products. If you would use MySQL for a job, then you would never have chosen DB2, because it's overkill. Likewise, if you use DB2, then MySQL could never have met your needs in the first place. There's really no overlap between DB2 and MySQL at all.
Now, Oracle versus DB2, or MySQL versus Microsoft Access, those are reasonable comparisons.
One of the posters below made the comparisons between a landrover and 747. You don't need a 747 for all tasks and there is nothing wrong with making cars and not planes.
That was precisely my point, and I'm glad somebody got it.
No, no, see, I deliberately avoided comparing two similar things, because I don't want to get into a conversation about whether MySQL is better or worse than DB2. That conversation is so fucking loaded. For instance, if you need to whip up a really quick database on your personal time to do something simple, MySQL's simplicity beats DB2's robustness hands down. So that whole conversation is pointless.
That's why I compared a car to a plane. They're both transportation machines (i.e., databases), but they're designed to do radically different jobs. They're just not comparable.
I'm disappointed that you were moderated as a troll. I think the way you said it was a little off the mark, but your ultimate point is correct. The major disconnect between the "free" software guys and the commercial software guys is that they're not on the same playing field.
The "free" software guys say things like, "Our software is politically and morally superior because it's free," and then they launch into a big discussion of liberty and rights.
The commercial software guys look at the bottom line. Take SGI for example. When they decided to build a big, scalable server system designed around the IA-64 chip family, they were faced with the prospect of doing a lot of work to port IRIX from the MIPS architecture. On the other hand, there was Linux, which needed a lot of work to be scalable and reliable, but was easier to use for this purpose than IRIX was. So they're running Linux. Do you think it's for political reasons? Shit, no. It's about costs and profitability.
Are you talking about DB2? You don't seriously consider MySQL to be a competing product to DB2, do you? That's kind of like saying Land Rover competes with Boeing. They're just in different classes altogether.
Geez, are you trying deliberately to be a twat, or is it just coming to you naturally?
If Apache does what you want, use it. If IIS or iPlanet or something does it better, use that. Do not simply use Apache because it's free. Use Apache because it's the right tool for the job.
Same thing goes for sendmail. If it works for you, use it. If it doesn't-- if you're somebody like AOL, say, who handles tens of millions of email messages a day-- then use something better. Whether the software is commercial or free only matters when you're planning your budgets.
Anybody who would choose an inferior free program over a superior commercial program simply because the free program is free is either a cheapskate or a fuckwit.
If I haven't convinced you by now, I never will. Which is fine. Again I say, it's not skin of my back if you want to be a moron.
But if you want to compare, compare fairly. Take the very best possible OS that the free software folks can come up with-- I have no opinion about what that is, but you could pick FreeBSD or Linux or whatever you like-- and compare it to the best possible OS that the commercial software folks can come up with-- OS/400, say. How does the free software stand up?
If you pick a piece of commercial software at random and a piece of similar free software, also at random, the commercial software will most likely be superior. That's all I'm sayin'.
Free software (by that I mean "open source", preferrably GPL if that's where you're getting confused) is fundamentally different than commercial software.
Wrong. Zealots often try to claim that this is so. Are you a zealot?
Free software-- which is neither free nor free, but that's an argument for another day-- is not fundamentally different from commercial software. It is made up of source code in one or more of a number of languages that gets run through a compiler to produce executable objects. These objects do various things, some autonomously, some interactively.
Free software is fundamentally exactly the same as commercial software. It is different only in execution. Commercial software must compete in the marketplace to survive. In some cases, it must compete against one or more other pieces of software that does roughly the same thing. In some cases, it must compete against the absence of itself; in other words, it doesn't matter how great Photoshop is, if it cost $10,000 a license, many people would probably choose not to buy it.
Commercial software that can't compete dies. Commercial software costs money to develop and maintain, so if it can't be sold for a profit, that money dries up and the software disappears. That fact creates an inherent lower bound for the quality of commercial software. If it's not good enough so that at least some people will buy it, it simply evaporates.
At the other end of the spectrum, software quality becomes asymptotic in nature; i.e., there's always one more bug. You can get software to what I call beta quality without an inordinate amount of work: simply implement all the features and let the bugs fall where they may. This is where a lot of free software is on the quality spectrum. The stuff that's not completely useless-- won't compile properly, or doesn't implement all of the necessary functions-- is beta-quality.
Improving software quality gets harder and harder as your standards rise. There's the test-and-fix cycle, of course, and that's where most hobbyist or free software projects stop. But many bugs will never be found through test-and-fix alone. Many bugs, like memory leaks, synchronization problems such as deadlocks or race conditions, or stack or buffer overflows, can only be found through serious looking. Regression testing is a decent technique, but it only helps you guarantee that your software doesn't get any worse; if you start out with buggy software, regression testing won't necessarily help you make it better. Unit testing is another decent technique, but you have to take the time and effort to write good tests for this technique to be helpful. Often, finding bugs happens through simple, rigorous, boring code review processes. These processes simply don't exist in the free software world, because there's no financial motivation to implement and enact them.
This area-- software quality-- is where free software completely drops the ball. Commercial software is subjected to this kind of testing because the stakes are higher; if you have a $10 million software company, you're not going to take a chance on releasing a piece of software until it's been thoroughly tested. Hobbyists, on the other hand, release whenever they feel like it, which often means before any rigorous testing has been done.
So if you say to yourself, "I will only use free software," you are guaranteeing that the highest-quality software will be unavailable to you, and you are taking on the harder job of finding acceptable software among all of the completely unacceptable software out there. Dumb.
1. The G4 is up to 1.25 Ghz and only comes in dual configurations.
Mostly true, but technically false. The Apple Store for educators will still sell you a 900 MHz single-processor Quicksilver system, if you're a teacher or a student. If I recall, the price is about $1,200, but that's totally from memory, so don't bitch at me if I got it wrong.
That's really, really impressive. Almost every sentence in your post was wrong! The only one that you got right was, "Yes, it makes a good desktop." You obviously put a lot of effort into this post, and I respect that. I laugh at it, but at the same time I respect it.
Please don't use Mac OS X. Sorry, dude, but you're just not cut out for it. Everything you said here is just a Mac OS 9 feature that you wish you had in Mac OS X. I have a great solution for you: use Mac OS 9.
Furthermore, your list of suggestions on the web site you mentioned are really uninformed. You say things like "Idealy, right clicking should bring up options menus, while middle clicking would bring up program menus. Why make the user go to the menu, when the menu can come to the user?" that make it clear that you've never watched a novice use a computer before. If anybody were to take your suggestions, they'd end up with a desktop environment that was only usable to perhaps 1% of the population. Which is pretty pointless.
It is quite schizophrenic about APIs: the BSD, Carbon, and Cocoa APIs really aren't all that well integrated.
You say schizophrenic, I say rich. All of the BSD, Carbon, and Cocoa APIs can be called from the same Objective C program; BSD and Carbon are C-language APIs, while Cocoa is Objective C. You can mix-and-match calls to your heart's content.
There are half a dozen different kinds of executables, with entirely different behaviors.
Half a dozen? Don't you mean two? There's CFM and dlyd, or PEF and Mach-O if you prefer those names.
But they are not replacements for UNIX workstations or Linux machines--they are replacements for Windows desktop machines.
Well, since my company replaced a bunch of SGI O2 workstations (UNIX workstations) with Power Macs running OS X, I'd have to say I don't agree with your assessment. Not that that's all I disagree with; your comments on Quartz are just way off the mark.
Slashdot Mirror
Because the world is a better place thanks to BSD, maybe?
Jesus. I can't believe you're still talking about this.
For the vast majority of what people do -- that is, the common functions which everyone does with a computer....
Your list is woefully incomplete. Let me take the least-computer-savvy person I know: my secretary. She's even less into computers than my mother is. She owns a computer for doing basic tasks, like email and web surfing, and she also uses it to write her family's holiday letter, complete with color photos taken on her digital camera. She prints the letters on her color inkjet printer and mails 'em out. I believe she uses something like AppleWorks for this, but I couldn't say for sure, because she doesn't know either. She just talks about using the computer: downloading pictures off of her camera, cropping them and doing simple color corrections ("I made it more colorful," she likes to say), page layout, illustration, printing, it's all just "the computer" to her.
She would be unable to do these things with Linux. The tools to do them simply don't exist. If you think all the average computer user wants to do is download MP3s and play DVDs, I think you're being a bit naive. In fact, I don't actually know anybody who downloads MP3s-- that generally being illegal-- or who uses their computer for DVD playback, with the exception of watching movies on planes. My understanding is that there are no legal DVD playback applications for Linux anyway, so that point is moot, unless my knowledge is out of date.
For typical home use, the best-suited is probably GNUcash.
No online banking, no recurring payments, inadequate reporting, no financial planning features to speak of, no integration with external data sources for things like stock portfolio management... yes, Gnucash is certainly well-suited.
Don't think just because I don't like or respect Linux as a desktop operating system that I'm uninformed. The main disconnect between us seems to be that I know what you're talking about, but that you have no idea what I'm talking about.
Another fact about Linux which is very commonly overlooked is that you gets tons of software with any major distribution of Linux. A whole slew more than you'll ever need, and orders of magnitude more than you get along with Windows and MacOS.
If none of it is useful, though, then all that software only exists to confuse and annoy. Red Hat 7.3 comes with something like six different shells, and no page layout programs. Amazing.
In Linux, if one process crashes, that doesn't effect the system; it does in Windows.
Except that it doesn't in Windows. Or have you never used Windows 2000 or Windows XP? Neither of these OS's is 100% crash-proof, but crashes are incredibly unlikely, roughly on par with kernel panics in Linux. The same is true of OS X.
The other reason to be running Windows/Mac programs in an emulator when needed is to take advantage of the powerful features of Linux w/o having to reboot.
What features are those, exactly? And which of them exist in Linux, but don't exist in OS X? OS X is a full-fledged UNIX operating system, recall. At any point I can drop out of the graphical UI and work in a fully POSIX-compliant command-line UI environment. Without, as you say, having to reboot.
Tasks that most people want to accomplish an be done in all three OS'.
Please explain how my secretary can create her family's holiday letter using Linux. Please explain how I can create the marketing data sheets for my company's products using Linux. Please explain how my girlfriend can enjoy her photography hobby using Linux. You are completely unaware, evidently, of how most people use computers.
I disagree b/c Linux is used for big-time special effects
Linux is primarily used as a renderer. In other words, as a computation server. I don't know anybody-- with the exception of one guy who's a freelance animator-- who owns his own render farm. So that's kinda silly of you to say.
In those instances where Linux is being used on the desktop, it's either for running other third-party UNIX applications-- such as Maya or Softimage-- on cheap hardware, or it's for running proprietary UNIX applications on cheap hardware. ILM is currently moving all of their compositors from SGI workstations to Linux workstations, because once they got all the bugs out of the first one, they were able to mass-produce them really cheaply. The software in that case is CompTime. Again, this has nothing to do with anybody but ILM. It's not a reflection of how suitable Linux is for doing typical creative work at all.
but there's also things that one can do in Linux that one can't do elsewhere (scientific applications...
Like which scientific applications, exactly? And why do they matter to people who use personal computers?
complete customizability...
Linux advocates often fall back on this argument. What they fail to realize is that practically nobody cares about customizability. That's like saying that your computer is better because the RAM is purple. Doesn't matter. In fact, it's even worse than that, because customizability can-- not always, but sometimes-- get in the way of a usable user interface. The world is full of trade-offs, and if programmers spend more time making their applications customizable than they do making them usable, then you end up with loads and loads of highly customizable, really crappy software. Which is how I would describe software like KDE, Gnome, and Mozilla: thoroughly executed, embarrassingly designed.
games -- Linux is nowhere near MS in that category, and if your honest, neither is MacOS
That's funny, I would have sworn that Warcraft III released on both Mac OS and Windows on the same day. How odd. I must have been mistaken.
Saying Linux is at best comparable to a subset of OS X is insulting.
Well, gee, dude, earlier in your message you compared Linux directly to FreeBSD. Mac OS X is a superset of FreeBSD 4.4. Ergo...?
One could say that OS X is only comparable to a subset of Linux because of all the scientific apps available in Linux with no equivalent in OS X.
Name one. I'll download and compile it in short order. If it's a commercial application-- although I'm pretty sure most of the companies who sell commercial software for Linux have gone out of business-- then it's a trivial matter for the vendor to port.
Sounds like Mac OS X is functionally a superset of Linux, to me.
You are simply trying to close off an avenue of debate by saying that Linux shouldn't even be an option for serious consideration, as it doesn't allow one to do all of the things that one can do in MacOS; but as I just demonstrated, that argument can be turned on its head.
Yes, I am most certainly saying that Linux shouldn't be an option for serious consideration, and just for the reason you give. However, I fail to see how you have turned any argument on its head, exactly. Maybe I missed it. You should go over it again.
I choose Linux because of the many quality apps that come with it for free and because of its stability & speed; as well as the fact that it doesn't require you to accept any draconian EULA's.
Ah, finally we get down to the heart of it. Inevitably, whenever people try to discuss Linux on its merits, they end up falling back to politics. Licensing agreements are simply not a factor. One either agrees to them and abides by them, or one doesn't. It's really simple. Linux advocates, on the other hand, seem to be opposed to licensing agreements in principle. Principle is all fine and good for things like voting booths and church confessionals. It has little to do with one's choice of personal computer.
Sounds to me, dh003i, like all of my original arguments stand, and all you have to fall back on is your political views. Does that pretty much sum it up?
First of all, I understand the difference between OS and application very well. But I have news for you. For practical purposes, there is no difference. This is an important fact of life that Linux advocates often fail to understand.
One chooses an OS because of what one can do with it, which is a function of the applications. If two OS's can offer the same or equivalent applications for a given task, then the OS's are equivalent for that task, and you can make your choice between them based on preference or minor features or whatever.
But if two operating systems do not offer the same or equivalent applications for a given task, then they are not equivalent for that task. You can't run an enterprise-class DBMS under Mac OS 9, and you can't play Warcraft under OS/400. So comparing Mac OS 9 to OS/400 for running a database would be pointless, just as it would for running Warcraft.
In large part, Mac OS X and Windows XP are equivalent; you can do the same things on one as you can on the other. Some things are easier on Mac OS X-- running Apache/Tomcat, for example, is trivial-- while other things are easier on Windows XP-- trying to use the Mac version of Outlook under OS X's Classic mode is painful at best. But despite the differences, the two platforms are functionally equivalent, so comparing them makes sense.
Linux is not comparable to OS X or to Windows. There are simply too many basic tasks that one might want to do with one's computer that cannot be done under Linux. You say you've never used Illustrator or InDesign; that explains, in part, why you don't understand my point. There is no Linux equivalent to either of these programs, just as there's no Linux equivalent to Quicken or Microsoft Access or After Effects or Quark XPress or Crystal Reports... the list goes on and on.
If your best suggestion is to run the Windows versions of these programs thorough an emulator, virtual machine, or other similar technique, then you really ought to be running Windows. It'll be more reliable.
I stand by my argument: OS X and Linux aren't comparable because they cannot both be used to accomplish the same set of tasks. At best, Linux is comparable to a subset of OS X, which is no meaningful comparison at all.
If it was fine on MySQL, then those people shouldn't have chosen Oracle without thinking about it. Oracle would have been overkill for that job.
Which is my point. MySQL and Oracle are not equivalent. They're not interchangeable.
There are equivalent appllications in Linux to everything you've mentioned...
What's the Linux equivalent of Illustrator? Or of InDesign? There are none at all. How about the Linux equivalent of PowerPoint? Or Quicken? If you believe there are Linux equivalents of these programs, you are mistaken.
You seem to be implying that its impossible to run many applications at once under Linux.
What? No, I didn't. I implied-- hell, I flat-out said-- that the applications I use are not all available on any platform other than OS X. That's as simple as it gets.
These OS' do the same things, just in different ways.
Um... nobody really cares what an OS does, okay? People care about what they can do. Using my computer, I can accomplish everything I need to do. This would not be true if I used Linux, because the tools I need to use simply don't exist there. Ergo, OS X and Linux do not "do the same things, just in different ways." There are things you can do on OS X that simply can't be done under Linux.
This is why there's simply no comparing the two.
Aww, you don't understand business.
A successful company-- and IBM is probably the textbook definition of a successful company-- has no desire to gouge its customers. IBM could, if they wanted, aggressively sell DB2 to small companies that can't really afford it. But those companies would end up being unsatisfied customers, which would hurt IBM more than the revenue of the sale helped them.
Business relationships between vendor and customer are more like symbiosis than parasitism. Keep that in mind here.
Um... dude... you're talking about ease-of-use, and I'm talking about pure functionality. There no comparison between OS X and Linux. Sorry to burst your bubble.
On my Mac, I typically run the following: OmniWeb, Mail.app, Project Builder, Interface Builder, Apache/Tomcat, Word, Excel, PowerPoint, Palm Desktop, iCal, iTunes, XEmacs, Maya, Illustrator, Photoshop, InDesign, VirtualPC (for IE and Outlook), OmniOutliner, Quicken, Acrobat, iChat, and, sometimes, Warcraft III.
This is not possible under Linux. At all, with any combination of hardware and software. Ergo, there is no comparison. It's like comparing a toothbrush to a boxing glove. They just don't do the same things, so talking about them as if they do is just a waste of time.
When it comes to web servers in mid-range companies, MySQL and DB2 are competitors.
No, they're not. Because they're not equivalent products. If you would use MySQL for a job, then you would never have chosen DB2, because it's overkill. Likewise, if you use DB2, then MySQL could never have met your needs in the first place. There's really no overlap between DB2 and MySQL at all.
Now, Oracle versus DB2, or MySQL versus Microsoft Access, those are reasonable comparisons.
One of the posters below made the comparisons between a landrover and 747. You don't need a 747 for all tasks and there is nothing wrong with making cars and not planes.
That was precisely my point, and I'm glad somebody got it.
Not the greatest endorsement in the world. The database behind Slashdot falls over pretty frequently.
No, no, see, I deliberately avoided comparing two similar things, because I don't want to get into a conversation about whether MySQL is better or worse than DB2. That conversation is so fucking loaded. For instance, if you need to whip up a really quick database on your personal time to do something simple, MySQL's simplicity beats DB2's robustness hands down. So that whole conversation is pointless.
That's why I compared a car to a plane. They're both transportation machines (i.e., databases), but they're designed to do radically different jobs. They're just not comparable.
Forget transactions -- you can fake that with LOCKs.
How do you fake a roll-back with LOCK?
I'm disappointed that you were moderated as a troll. I think the way you said it was a little off the mark, but your ultimate point is correct. The major disconnect between the "free" software guys and the commercial software guys is that they're not on the same playing field.
The "free" software guys say things like, "Our software is politically and morally superior because it's free," and then they launch into a big discussion of liberty and rights.
The commercial software guys look at the bottom line. Take SGI for example. When they decided to build a big, scalable server system designed around the IA-64 chip family, they were faced with the prospect of doing a lot of work to port IRIX from the MIPS architecture. On the other hand, there was Linux, which needed a lot of work to be scalable and reliable, but was easier to use for this purpose than IRIX was. So they're running Linux. Do you think it's for political reasons? Shit, no. It's about costs and profitability.
This is, incidentally, exactly as it should be.
Are you talking about DB2? You don't seriously consider MySQL to be a competing product to DB2, do you? That's kind of like saying Land Rover competes with Boeing. They're just in different classes altogether.
I moderate this article -1, Flamebait.
Geez, are you trying deliberately to be a twat, or is it just coming to you naturally?
If Apache does what you want, use it. If IIS or iPlanet or something does it better, use that. Do not simply use Apache because it's free. Use Apache because it's the right tool for the job.
Same thing goes for sendmail. If it works for you, use it. If it doesn't-- if you're somebody like AOL, say, who handles tens of millions of email messages a day-- then use something better. Whether the software is commercial or free only matters when you're planning your budgets.
Anybody who would choose an inferior free program over a superior commercial program simply because the free program is free is either a cheapskate or a fuckwit.
If I haven't convinced you by now, I never will. Which is fine. Again I say, it's not skin of my back if you want to be a moron.
But if you want to compare, compare fairly. Take the very best possible OS that the free software folks can come up with-- I have no opinion about what that is, but you could pick FreeBSD or Linux or whatever you like-- and compare it to the best possible OS that the commercial software folks can come up with-- OS/400, say. How does the free software stand up?
If you pick a piece of commercial software at random and a piece of similar free software, also at random, the commercial software will most likely be superior. That's all I'm sayin'.
I don't know what your deal is, but you've got deeper issues here than peoples' software preferences.
My deal is that your preferences are based on dumb criteria. Demonstrably dumb. But hey, it's your life. Doesn't bother me if you wanna be an idiot.
What a fantastic straw man. Use logic much?
Actually, it's a metaphor.
Free software (by that I mean "open source", preferrably GPL if that's where you're getting confused) is fundamentally different than commercial software.
Wrong. Zealots often try to claim that this is so. Are you a zealot?
Free software-- which is neither free nor free, but that's an argument for another day-- is not fundamentally different from commercial software. It is made up of source code in one or more of a number of languages that gets run through a compiler to produce executable objects. These objects do various things, some autonomously, some interactively.
Free software is fundamentally exactly the same as commercial software. It is different only in execution. Commercial software must compete in the marketplace to survive. In some cases, it must compete against one or more other pieces of software that does roughly the same thing. In some cases, it must compete against the absence of itself; in other words, it doesn't matter how great Photoshop is, if it cost $10,000 a license, many people would probably choose not to buy it.
Commercial software that can't compete dies. Commercial software costs money to develop and maintain, so if it can't be sold for a profit, that money dries up and the software disappears. That fact creates an inherent lower bound for the quality of commercial software. If it's not good enough so that at least some people will buy it, it simply evaporates.
At the other end of the spectrum, software quality becomes asymptotic in nature; i.e., there's always one more bug. You can get software to what I call beta quality without an inordinate amount of work: simply implement all the features and let the bugs fall where they may. This is where a lot of free software is on the quality spectrum. The stuff that's not completely useless-- won't compile properly, or doesn't implement all of the necessary functions-- is beta-quality.
Improving software quality gets harder and harder as your standards rise. There's the test-and-fix cycle, of course, and that's where most hobbyist or free software projects stop. But many bugs will never be found through test-and-fix alone. Many bugs, like memory leaks, synchronization problems such as deadlocks or race conditions, or stack or buffer overflows, can only be found through serious looking. Regression testing is a decent technique, but it only helps you guarantee that your software doesn't get any worse; if you start out with buggy software, regression testing won't necessarily help you make it better. Unit testing is another decent technique, but you have to take the time and effort to write good tests for this technique to be helpful. Often, finding bugs happens through simple, rigorous, boring code review processes. These processes simply don't exist in the free software world, because there's no financial motivation to implement and enact them.
This area-- software quality-- is where free software completely drops the ball. Commercial software is subjected to this kind of testing because the stakes are higher; if you have a $10 million software company, you're not going to take a chance on releasing a piece of software until it's been thoroughly tested. Hobbyists, on the other hand, release whenever they feel like it, which often means before any rigorous testing has been done.
So if you say to yourself, "I will only use free software," you are guaranteeing that the highest-quality software will be unavailable to you, and you are taking on the harder job of finding acceptable software among all of the completely unacceptable software out there. Dumb.
Yes. I was goofing, as evidenced by the ";-)" trigraph in my message. Seemed clear enough to me....
1. The G4 is up to 1.25 Ghz and only comes in dual configurations.
Mostly true, but technically false. The Apple Store for educators will still sell you a 900 MHz single-processor Quicksilver system, if you're a teacher or a student. If I recall, the price is about $1,200, but that's totally from memory, so don't bitch at me if I got it wrong.
If I can keep a 24 processor Sun busy for an hour, I can probably figure out how to keep a PC busy, eh?
Dude,
while (1) {
fork();
}
doesn't count.
That's really, really impressive. Almost every sentence in your post was wrong! The only one that you got right was, "Yes, it makes a good desktop." You obviously put a lot of effort into this post, and I respect that. I laugh at it, but at the same time I respect it.
Please don't use Mac OS X. Sorry, dude, but you're just not cut out for it. Everything you said here is just a Mac OS 9 feature that you wish you had in Mac OS X. I have a great solution for you: use Mac OS 9.
Furthermore, your list of suggestions on the web site you mentioned are really uninformed. You say things like "Idealy, right clicking should bring up options menus, while middle clicking would bring up program menus. Why make the user go to the menu, when the menu can come to the user?" that make it clear that you've never watched a novice use a computer before. If anybody were to take your suggestions, they'd end up with a desktop environment that was only usable to perhaps 1% of the population. Which is pretty pointless.
It is quite schizophrenic about APIs: the BSD, Carbon, and Cocoa APIs really aren't all that well integrated.
You say schizophrenic, I say rich. All of the BSD, Carbon, and Cocoa APIs can be called from the same Objective C program; BSD and Carbon are C-language APIs, while Cocoa is Objective C. You can mix-and-match calls to your heart's content.
There are half a dozen different kinds of executables, with entirely different behaviors.
Half a dozen? Don't you mean two? There's CFM and dlyd, or PEF and Mach-O if you prefer those names.
But they are not replacements for UNIX workstations or Linux machines--they are replacements for Windows desktop machines.
Well, since my company replaced a bunch of SGI O2 workstations (UNIX workstations) with Power Macs running OS X, I'd have to say I don't agree with your assessment. Not that that's all I disagree with; your comments on Quartz are just way off the mark.