You did realize that by having a different version CoreGraphics file than for what this was designed, the offset for the replaced bytes would likely be wrong and that you would destroy the file, right?
And you blame the OS for having problems after modifying this crucial system file (that is normally read-only) yourself?
Har har. I take it you never read any of my posts lambasting themes installers like this for Mac OS X.
Well, no shit, Sherlock. Tell me you never destablizied your system (in any OS) by installing and uninstalling various software? Furthermore, my reply was in response to NETROmac. If you have default install of WindowsXP and never install/uninstall any crap - guess what, it is also going to be as stable as hell!
In any case, thanks for pointing out the OBVIOUS regarding CoreGraphics file. When one breaks the system, like I did, one usually finds out AFTER the fact - not before.
Well, no shit, Sherlock. Tell me you never destablizied your system (in any OS) by installing and uninstalling various software?
ummmm... actually yeah... there are people who go years with installing "crap" from version tracker without effecting system stability...
the reason is beacaus Mac OS applications are PREDICTABLE and they (usually) only install stuff into one place so you don't have to go hunting to uninstall them... of course it goes without saying that if the developer does a bad job and puts shit all over your hard drive you will have problems... the difference is that putting shit all over your hard drive is standard procedure in XP thats why you need to have an uninstaller... Mac OS programs (good ones anyway) don't need them...
also hacking the os and installing programs are 2 different things...
I installed BeachBall - a cursor changing program is not equivalent to hacking the OS. Yes, I am now aware it did but why is cursor part of that all important CoreGraphic file?
In any case, we are diverging from the topic at hand...
Isn't that great at living up to the hype. It has its uses, for sure, but the codebase leaves something to be desired, stability and security are OK, setup and maintenance are a pain, overall polish is lacking... it's not bad, and I'd prefer it to any MS OS in a heartbeat, but that's not exactly a compliment.
This is simply techier-than-thou BS. You seem offended that Linux isn't living up to your high technical standards and yet you're happy to jump onto the bandwagon of a barely coherent troll. What's up with that?
And for future reference, being better than any MS OS *is* a complement.
Right, but BSD is more free (the license is less restrictive than GNU's), more secure, more stable, and it ran on just about everything back when Linux was still stuck on x86.
Care to explain how this extra 'freeness' provides *any* benefit.
You appear to have some disdain for open source software in general (if you pay money for it, it must be good!) and a grudge against the GPL in particular. (Later you bizarrely claim BSD licenced software is "held to a higher standard" than GPL'd software.)
You may want to meditate on what about the GPL helped Linux overtake BSD (in a variety of ways) from a standing start. Hint: why do the companies pouring billions(!) of dollars into Linux not want that extra 'freeness'?
As for 'stuck on x86', I'll come back to that later.
The administrative software runs like crap on Linux, because of limitations in Linux - in the core and elsewhere.
Now *that* is a lame cop-out. What the hell is this admin software doing that Apache, Mozilla, Emacs, many massive supercomputer clusters, Sharp Zaurus, Tivo etc. etc. aren't?
Fair enough. I'll point you to a hands-on report by a guy with years of in-depth Linux experience:
So much experience that he lost a month of data in a server crash?
Note that some of the reasons simply come down to discipline: The folks behind BSD wait two years before declaring a codebase stable, so when you install a stable build of BSD, you can be pretty damn sure that the bugs have been worked out.
Which BSD? Which Linux distribution are you comparing it with? Wouldn't you have more confidence if large numbers of people had actually been using the system for two years? You do realise that you are allowed to use other versions of Linux than the one in Linus's cvs tree?
That's a good working thesis, and it's certainly a big reason why MS is afraid of free software in general. But the same arguments apply to BSD.
No they don't, because BSD isn't threatening MSFT. You'll notice that (like yourself) MSFT reserve special malice for anything GPL'd.
But the "expensive" criticism really doesn't apply.
Yes it does. Running commodity OS's on commodity hardware (the x86's that you appear to rate as much as Linux) takes a major chunk out IT budgets and gives you leverage with suppliers, who no longer have you tied to their expensive hardware and niche OS and homebrew software tools. They are increasingly being asked to justify the cost differential and *failing*.
And we do it by paying for software that we can just drop in and expect to work.
How does "paying money == better" any more than "free == better"? Why does paying for some software mean paying for all software?
Your fervent enmity for Linux in particular and open source software in general is simply incomprehensible, particularly given your praise for the BSDs. Did the Linux geeks bully you at school or what?
I think Apple has taken the best of open source, which is free (as in beer), solid and flexible code, and merged it with the focus and accountability of closed-source software. As a result you see masterpieces like Safari and the Mac OS itself.
I don't know if that will be enough to get Mac OS X noticed by IT professionals, but it has worked wonders with UNIX developers.
Yes, I am now aware it did but why is cursor part of that all important CoreGraphic file?
Because the cursor is a fundamental part of the interface. Because Apple never intended for anyone to change the cursor graphics. Therefore, you change it, and the system won't boot.
This monitors uptimes of companies servers. The first example is apple (switching over to X back in 2000) and the second example which is of more interest to this discussion shows at least one company dropping linux for HP-UX. I don't know much about the other OS's, but they do seem to be liking their update.
If you look at their current systems, it looks like through a bunch of sellings and stuff, they ran the gamut with OSes
At my previous job, our Apple gurus tried to push OS X in a big way. A single problem hindered rolling out Mac to PC for over 3000 Staff members.
You can't get your password to expire after 30 days!
I don't know what version of Mac OS X Server you were using but we're using v 10.2.6 and in Workgroup Manager under password options you can definitely have a password expire.
A lot of you people seem to be confused. You are missing the line between OS and UI.
In Windows, this is quite hard to see as the UI is welded to the OS. However in linux and MacOS X its quite easy to see the difference. Your linux kernel and support files = the OS, KDE, XFree4, Gnome etc = the UI or Darwin = the OS, and Aqua = the UI.
Linux OS is good. It has the same roots as FreeBSD, and is maintained by some of the best programmers on the planet. As is BSD, and Darwin. However some of the "bolt-on" UI's for linux are crap.
Simarly, some people forget where darwin comes from, and continually assume that MacOS X cannot do something just because there isn't a nice GUI for such a task.
If I were responsible for a network, I would use macos X purly because in the last 4 months of owning this iMac, it hasn't crashed once!!
This monitors uptimes of companies servers. The first example is apple (switching over to X back in 2000) and the second example which is of more interest to this discussion shows at least one company dropping linux for HP-UX. I don't know much about the other OS's, but they do seem to be liking their update.
If you look at their current systems, it looks like through a bunch of sellings and stuff, they ran the gamut with OSes
Code Master, it's easy to manipulate Netcraft's "what's that site running?" function by having an intermediary server running a different OS. Check www.goatse.cx's uptime if you don't believe me. (But whatever you do, don't actually visit the site. I can't stress this enough!)
Rumor has it that Microsoft still runs Hotmail off of FreeBSD because it's simply impossible to get the stability and flexibility for such a huge site to run on WinNT/2000. They just put an intermediary Win 2000 server in the way so that anyone who asks "what OS" will be fooled into thinking that MS eats its own dog food.
This is simply techier-than-thou BS. You seem offended that Linux isn't living up to your high technical standards and yet you're happy to jump onto the bandwagon of a barely coherent troll. What's up with that?
Huh?
Quote:
You appear to have some disdain for open source software in general (if you pay money for it, it must be good!) and a grudge against the GPL in particular. (Later you bizarrely claim BSD licenced software is "held to a higher standard" than GPL'd software.)
You're reading all kinds of really bizarre statements into my post. First of all, for the record, I've known of and admired and used the FSF's cause and software for about 15 years now. I do not disdain it. Furthermore, I specifically said that most GNU software is written to a much higher standard than Linux is, so your parenthetical claim is simply false.
Also, the "if you pay money for it, it must be good" is false from my post, trivially, or I would have said that Windows was preferable to Linux and BSD both. I didn't. However, when you're looking at a solution, you should look at the overall cost. A solution with an upfront cost and low support costs might be better than a solution with no upfront cost and higher support costs. It depends on what you need. But the original poster's assertion that Linux is automatically cheaper just because it has no upfront cost is false.
Quote:
You may want to meditate on what about the GPL helped Linux overtake BSD (in a variety of ways) from a standing start. Hint: why do the companies pouring billions(!) of dollars into Linux not want that extra 'freeness'?
You might want to meditate on what the AT&T suit did for BSD's momentum. Before Linux got anywhere, Sun, Digital and many other companies were pouring money into BSD. Just to stay on AT&T's good side, they all migrated to SVR4.
Linux' success had very little to do with the GPL, although the GPL didn't hurt.
Quote:
The administrative software runs like crap on Linux, because of limitations in Linux - in the core and elsewhere.
Now *that* is a lame cop-out. What the hell is this admin software doing that Apache, Mozilla, Emacs, many massive supercomputer clusters, Sharp Zaurus, Tivo etc. etc. aren't?
Talking to an expensive, high-bandwidth piece of hardware at the driver level?
Quote:
So much experience that he lost a month of data in a server crash?
Read the article and draw your own conclusions.
Quote:
Which BSD? Which Linux distribution are you comparing it with? Wouldn't you have more confidence if large numbers of people had actually been using the system for two years? You do realise that you are allowed to use other versions of Linux than the one in Linus's cvs tree?
That doesn't mean the older releases are as rigorously tested as FreeBSD's (the BSD the article refers to, with the two year testing period). Last I checked, Yahoo.com ran on FreeBSD, so I'd say their releases are very well tested.
You can't ignore the basic problem posed in the article: The fix-by-incremental-upgrade model doesn't work well. This isn't a free software issue, it's a discipline issue. It's the same reason AIX is far more secure than Solaris, even though they're both variants on the same codebase.
Quote:
No they don't, because BSD isn't threatening MSFT. You'll notice that (like yourself) MSFT reserve special malice for anything GPL'd.
BSD doesn't have the "buzz" or whatever that Linux has, but MS is going after open source in general, and the GPL in particular, because of the restrictions against commercial use - and, since you seem to have gotten the impression from somewhere that not being impressed by Linux equates to a raging hatred of the GPL and everything it stands for, I will emphasize that I have no problem whatsoever with the GPL. I admire its means, and its ends, and unilaterally oppose MS' attempts to attack it. They might not care so much about the BSD license (and licenses like it, such as Perl's) just because those licenses don't prevent or restrict interaction with commercial code.
Quote:
And we do it by paying for software that we can just drop in and expect to work.
How does "paying money == better" any more than "free == better"? Why does paying for some software mean paying for all software?
It doesn't. You're drawing a complete non sequitur from my argument. I'm am arguing against the two points that "free == better" (by saying only that "free doesn't necessarily mean better," and using my own shop as an example), and that Linux is some exemplar of either free software or free OS' that can run on commodity hardware.
On the other hand, I'm currently trying to leverage the Apache Foundation's XML efforts to the greatest degree possible, with an eye toward eventually getting something like Mozilla to render the front end. Because, you see, I've used and admired open source software for 15 years or so, whenever it's been the best tool for the job.
Quote:
Your fervent enmity for Linux in particular and open source software in general is simply incomprehensible, particularly given your praise for the BSDs. Did the Linux geeks bully you at school or what?
Linux didn't exist when I was in school. I don't know where you got "fervent emnity" from; it's just that I don't think Linux lives up to the hype around it, I don't think it's yet ready for some of the things it's being groomed for, and I don't think it's always chosen for the right reasons, or the tool best suited to the job. Nothing I have seen has convinced me that it's better than BSD for most server duties, or that its learning curve is lower than other alternatives, etc. That doesn't mean that it's evil or useless or any of the wild conclusions that you're trying to draw from my skepticism.
You're reading all kinds of really bizarre statements into my post.
My apologies for getting the wrong end of the stick, I was thrown off by your inital agreement with inkhead (who by the way, does not have a point).
It's easy to get lost when you have to change gears from a rant about Linux that criticises the GUI (i.e. Gnome/KDE or others) and a mish-mash of other classic trolls, to one that criticises Linux kernel development but praises the BSDs which share (or cross-pollinate) a great deal of code, philosophy, software, users etc. with the Linux distributions that inkhead was attempting to rubbish.
I think therefore the much subtler point you were making (that I missed completely) is that the Linux (kernel) development model of continual improvement from a low starting point isn't as good as the engineering excellence of the BSD and others (not-necessarily open-source).
You ask
Quote:
It's "good enough." Haven't we had enough of "good enough" at this point?
And the answer is no. We've had enough of "good enough" software that puts its competitors out of business with shady practices, of "good enough" software that gets sold on the basis of powerpoint presentations and locks you into a becoming the cash cow of a lazy company with no motivation to innovate or improve.
If your future choices aren't going to be impeded by the current choice then good enough software is exactly what you want.
Also, you claim that the fix-by-incremental-upgrades model doesn't work but sometimes worse *is* better
My apologies for getting the wrong end of the stick, I was thrown off by your inital agreement with inkhead (who by the way, does not have a point).
I was implicitly ratcheting down his language, which was probably the source of your confusion. Insofar as he states that Linux is not a great choice for a lot of enterprise work, I have to agree. You can massage it up to the point where it is, sometimes, but that's not really a recommendation.
Quote:
It's easy to get lost when you have to change gears from a rant about Linux that criticises the GUI (i.e. Gnome/KDE or others) and a mish-mash of other classic trolls, to one that criticises Linux kernel development but praises the BSDs which share (or cross-pollinate) a great deal of code, philosophy, software, users etc. with the Linux distributions that inkhead was attempting to rubbish.
I don't know how much philosophy BSD shares with Linux, or for that matter how much Linux shares with much other GNU code. There are choices beside the cathedral and the bazaar, which both the BSDs and the FSF itself have taken (Apple, of course, is erecting a cathedral with its doors flung open to the bazaar outside). I am unimpressed with GUI efforts like Gnome for many of the same reason that I'm unimpressed with kernel efforts like Linux. (Which, again, is not to say that Gnome is worthless, or should be discarded, or anything similarly absolutist. Lots of neat tools and libraries have come out of that effort, just as many cool and useful things have come from the Linux development effort. But the "bazaar" development model is better suited to tools and libraries than to large-scale projects like kernels and GUIs.) When Linux came out, and I saw the development model, I wondered what happened to the GNU Project's once-reknowned coding standards. I wonder how many recent converts to open source even know that the GNU Project has coding standards.
Quote:
You ask ["Haven't we had enough of 'good enough'?"]
And the answer is no. We've had enough of "good enough" software that puts its competitors out of business with shady practices, of "good enough" software that gets sold on the basis of powerpoint presentations and locks you into a becoming the cash cow of a lazy company with no motivation to innovate or improve.
If your future choices aren't going to be impeded by the current choice then good enough software is exactly what you want.
This smells of fatalism, which is basically just the passive tense blown up into a worldview. It also seems to miss one of the whole advantages of open source, which is that it simply doesn't go away. The BSD community doesn't have to be immense, it simply has to be big enough to keep the codebase up. There's really not that much development left to do on it. It's been mature for over a decade. So I wouldn't hesitate to adopt it, even if Linux runs away with mindshare. I certainly didn't abandon the Macintosh just because Windows ran away with all the market share, and the Mac is tied to one company's fortunes.
Quote:
Also, you claim that the fix-by-incremental-upgrades model doesn't work but sometimes worse *is* better
Certainly, advantages and disadvantages are situational. However, I don't believe that worse is better in the case of Linux, and in particular in the case of Linux as a useful alternative for enterprise. I doubt the platform would suffer if someone decided to take a given branch, battle-test it for two years, and then release a "stable" distro with the result. Many of BSD's advantages are simply the result of discipline over time. There's nothing preventing their best practices from being applied to Linux or Gnome or what-have-you. You wouldn't even have to change the way the main trunk was managed (although, obviously, that would help).
[edit: Having read the article above, I don't see the relevance. That's less about quality of implementation and more about philosophy of design. For the record, I have always fallen firmly into the New Jersey camp. But if you do implement something, you do it right. The original UNIX source was compared to poetry and assigned literary value - that quality, in fact, is what spawned the open source movement in the first place. It is also not a common trait within the Linux codebase, even allowing for Dennis Ritchie's NJ-ish maxim that any nontrivial program contains at least one hack.]
Now I want to argue that worse-is-better is better. C is a programming language designed for writing Unix, and it was designed using the New Jersey approach. C is therefore a language for which it is easy to write a decent compiler, and it requires the programmer to write text that is easy for the compiler to interpret. Some have called C a fancy assembly language. Both early Unix and C compilers had simple structures, are easy to port, require few machine resources to run, and provide about 50%--80% of what you want from an operating system and programming language.
Half the computers that exist at any point are worse than median (smaller or slower). Unix and C work fine on them. The worse-is-better philosophy means that implementation simplicity has highest priority, which means Unix and C are easy to port on such machines. Therefore, one expects that if the 50% functionality Unix and C support is satisfactory, they will start to appear everywhere. And they have, haven't they?
Unix and C are the ultimate computer viruses.
A further benefit of the worse-is-better philosophy is that the programmer is conditioned to sacrifice some safety, convenience, and hassle to get good performance and modest resource use. Programs written using the New Jersey approach will work well both in small machines and large ones, and the code will be portable because it is written on top of a virus.
It is important to remember that the initial virus has to be basically good. If so, the viral spread is assured as long as it is portable. Once the virus has spread, there will be pressure to improve it, possibly by increasing its functionality closer to 90%, but users have already been conditioned to accept worse than the right thing. Therefore, the worse-is-better software first will gain acceptance, second will condition its users to expect less, and third will be improved to a point that is almost the right thing. In concrete terms, even though Lisp compilers in 1987 were about as good as C compilers, there are many more compiler experts who want to make C compilers better than want to make Lisp compilers better.
The good news is that in 1995 we will have a good operating system and programming language; the bad news is that they will be Unix and C++.
Re-read this bit and think Linux, remembering that this guy is insulting (sorta) the Unix and C approach as he comes from a Lisp background.
You think Linux isn't very good. This guy is explaining why it will succeed because it isn't very good, though he's not talking about Linux in particular and it's probably best to substitute "run on cheap commodity hardware" for "portable".
Quote:
[Linux] provide[s] about 50%--80% of what you want from an operating system
Quote:
Therefore, the worse-is-better software [Linux] first will gain acceptance, second will condition its users to expect less, and third will be improved to a point that is almost the right thing.
Quote:
Half the computers that exist at any point are worse than median (smaller or slower). [Linux runs] fine on them.
Quote:
A further benefit of the worse-is-better philosophy is that the [Linux] programmer is conditioned to sacrifice some safety, convenience, and hassle to get good performance and modest resource use.
Quote:
... there are many more [kernal hackers] who want to make [Linux] better than want to make [VMS] better.
First, that this approach has limits. His assertion that the "worse-is-better" approach eventually runs as well on big hardware as on small is false. This approach scales down nicely, and scales up indifferently. Conversely, the "better is better" approach (which VMS nearly exemplifies) scales down poorly. VMS' filesystem is unbelievably robust, but the tradeoff is that it's complex and difficult to write for. When you have a big box running a mission-critical database under heavy load, that complexity and difficulty is a worthwhile tradeoff. But for any other purpose I would much rather deal with UNIX' brilliantly simple "everything is a stream" paradigm. Apparently, I'm not alone in this assessment, because VMS has a significant share of the uptime-critical market (and a good chunk of the high-performance computing market, because its clustering technology has been substantially better than anyone else's for over a decade now), and UNIX and UNIX-like OS' dominate markets where reliability and efficiency are desirable, but not critical.
I would also argue against the idea that C++ evolved C closer to a "perfect" solution. Objective-C did that, New Jersey-style, and look where it went. C++ exemplifies neither the MIT approach nor the New Jersey approach; it exemplifies design by committee and nothing else. It certainly isn't easy to write a compiler for!
Finally, again, there is the question of elegance of implementation. Again, what he's talking about is design. C doesn't do everything, and it doesn't try to, but it's clean and simple. UNIX doesn't do everything, and it doesn't try to (although System V made a credible lunge in that direction) but it was beautifully written. Ditto BSD (and its associated technologies, such as threading and TCP/IP). But the New Jersey style emphasizes elegance of implementation over completeness. Linux emphasizes neither - if anything, it's concentrated on speed optimization in a few areas, on portability (which does sort of dovetail with the NJ style) and on a wealth of user-level tools. But it's fundamentally different in approach and in philosophy from the other examples cited - not to mention in quality - and so I don't think the analogy holds directly. Linux is a different kind of worse, and to the extent that it actually succeeds in getting to the next level (as BSD had, before the lawsuit - it's hard to remember now, but in the late 1980s BSD was a major force with a very high rate of acceptance) it will have to realign itself to the NJ style. Or, it'll succeed by sheer momentum, and we'll have another Windows on our hands - but nothing of the quality produced by the likes of Thompson and Ritchie and Jordan Hubbard. And that's what I meant by "haven't we had enough of 'good enough'?" I consider the NJ style to be excellent, not 'good enough'. The MIT style still has to answer for MULTICS.
Inkhead was playing Devil's Advocate I admire him. No one ever gets to say anything bad about Linux. Which is "Line Ux" because it's Linux like Linux and Lucy, not Linn Ux.
I have Redhat 8. I can't see anywhere Linux would be better than BSD since that is all Linux is good for anyway, command line apps. The GUI is much worse than Windows 95. In fact it is a poor copy of Windows. The Virtual Desktops are nice but hey OS X has them. MacOS X is direct competition with Linux on the desktop and it wins. BSD wins on the server side. Why don't the Linux people start writing for Darwin/FreeBSD and help Apple fight M$? I don't see how *NIX divisions help anyone. The BSDs should unite, Linux should die, and everyone should help Apple, at least until M$ is no longer a monopoly. Apple should also return the favor: if Safari is free why isn't it open source, for example? iCal? iChat? iPhoto? Address Book? TextEdit? Hell Stickies? There are many small things that could be enhanced added or fixed that Apple might see in a submitted version and like and add to the apps in an update. As long as there is an agreement not to use the Apple code for commercial gain I don't see how it would hurt Apple, it could only help them and the open source community.
Inkhead was playing Devil's Advocate I admire him. No one ever gets to say anything bad about Linux.
There are plenty of uninformed Linux cheerleaders about. That fact and claims of playing devil's advocate don't give you carte blanche to be wrong about things e.g. the fact that no-one ever gets to criticize Linux.
I see it all the time, just like I see Mac users criticize Apple all the time. However, no-one likes their dirty laundry done in public though, *especially* by people with no clue what they're talking about (see almost any article on Apple by John C Dvorak).
Which is "Line Ux" because it's Linux like Linux and Lucy, not Linn Ux.
That doesn't really make any sense, but if it did it would still be wrong. We covered this recently try the search function.
I have Redhat 8. I can't see anywhere Linux would be better than BSD since that is all Linux is good for anyway, command line apps.
What about if you were creating a platform for home entertainment system that you wanted to share with other big businesses with no fear that one would fork it and take it closed source? Like Sony and Panasonic have. Or does your entire conception of computer OS's and what they are used for stretch from Windows 95 to Mac OS X, with a vague idea about "servers"?
The GUI is much worse than Windows 95.
Which GUI? The one on your phone, on your tivo, on your PDA, on your network appliances, on your home multimedia system, or what? Even if your only talking about Red Hat 8 there are many different GUIs: Gnome, KDE, Ximian etc.
Looking at the beatufilly anti-aliased fonts on my screen I'd say that they are better than Windows 95, but I'm a techy, so I suppose you'd better clarify for who and for what it is so much worse.
MacOS X is direct competition with Linux on the desktop and it wins. BSD wins on the server side.
Did you miss the discussion of how being "the best" doesn't help you "win"?. Reminder: 95% Windows: 5% Mac.
Why don't the Linux people start writing for Darwin/FreeBSD and help Apple fight M$? I don't see how *NIX divisions help anyone. The BSDs should unite, Linux should die, and everyone should help Apple, at least until M$ is no longer a monopoly.
You may have missed it, but basically they are. The vast majority of code above the kernel is not technically Linux and is used in the BSDs and many other crazy operating systems you've never heard of. More importantly they have no reason to not interoperate so I can, for example control several X11 or command line programs running on several *nix boxes through my Mac.
Apple should also return the favor: if Safari is free why isn't it open source, for example? iCal? iChat? iPhoto? Address Book? TextEdit? Hell Stickies? There are many small things that could be enhanced added or fixed that Apple might see in a submitted version and like and add to the apps in an update.
This would mostly be stupid for Apple. As long as they keep using open standards (which almost everything you mention does) and using open source effectively to advance their own economic interests (which they are, see Safari WebCore and Rendezvous for examples) then you can't ask for much more.
As long as there is an agreement not to use the Apple code for commercial gain I don't see how it would hurt Apple, it could only help them and the open source community.
If you can't use it for commercial gain it's not open source; it's part of the definition. #6 to be precise.
Apparently, I'm not alone in this assessment, because VMS has a significant share of the uptime-critical market (and a good chunk of the high-performance computing market, because its clustering technology has been substantially better than anyone else's for over a decade now), and UNIX and UNIX-like OS' dominate markets where reliability and efficiency are desirable, but not critical.
Here you think they're doing well because they're hanging onto their most valuable customers: a classic mistake.
Linux is a disruptive technology (I recommend the book of the same name, which doesn't directly address Linux) that steals the customers you least want first. However, in time it grows and steals so many of your customers that those that remain start to switch from the "better" product simply because of the costs of staying in a niche (assuming that the "better" product isn't simply overtaken in quality as the market focus shifts).
I would also argue against the idea that C++ evolved C closer to a "perfect" solution. Objective-C did that, New Jersey-style, and look where it went. C++ exemplifies neither the MIT approach nor the New Jersey approach; it exemplifies design by committee and nothing else. It certainly isn't easy to write a compiler for!
C++ is not "perfect", it's "worse". But then if it isn't "better" than Obj-C then why is Obj-C such a niche technology today, even with NeXT's, then Apple's championing (without which it would have died a lonely death)?
You're right that Obj-C is much, much "better". That is why it is "worse".
The design-by-commitee stuff is by-the-by, it was designed to ride upon the virus of C just like Unix was. The crux however is not technological in this case but psychological as C programmers wanted to keep on writing C while claiming to be doing OOP because they used a C++ compiler. No one wanted to really embrace OO, but no-one wanted to admit that either. (I am aware that plain C compiles as Obj-C btw)
Which GUI? The one on your phone, on your tivo, on your PDA, on your network appliances, on your home multimedia system, or what? Even if your only talking about Red Hat 8 there are many different GUIs: Gnome, KDE, Ximian etc.
My cell doesn't use Linux. Linux PDAs are not as good as PocketPCs, Symbians, or even (blech) Palms.
Quote:
Did you miss the discussion of how being "the best" doesn't help you "win"?. Reminder: 95% Windows: 5% Mac.
Exactly that is Amorph's point. Linux is "good enough." Just like Windows. Do you want to put up with that? I don't and I don't want to see one mediocre product replace another even if it's not controlled by an Evil Company. I like quality that's why I'm typing on a 12" PowerBook.
Quote:
You may have missed it, but basically they are.
No they aren't. The average end user doesn't even know what compiling is let alone how to do it. Thus you need to find binaries for each NIX flavor. NIX is very fractured and Linux just makes it worse.
Quote:
If you can't use it for commercial gain it's not open source; it's part of the definition. #6 to be precise.
Webcore is opensource but it is being put in a commercial product, OmniWeb.
I have not read a single article that calls Linux what it is: mediocre. Slashdot doesn't diss Linux like we diss Apple. And anyway Apple's software quality is lightyears ahead of anything on Linux. They just don't get it. It's about the interface.
klinux the only thing that has ever crashed OS X is SMB (smbfs or some component that Programmer clarified IIRC) and that was connecting to Windoze. I can't help but wonder if it is a coincidence that Windoze SFM and Windoze SMB crash Macs.
Comments
Originally posted by Brad
You did realize that by having a different version CoreGraphics file than for what this was designed, the offset for the replaced bytes would likely be wrong and that you would destroy the file, right?
And you blame the OS for having problems after modifying this crucial system file (that is normally read-only) yourself?
Har har.
Well, no shit, Sherlock. Tell me you never destablizied your system (in any OS) by installing and uninstalling various software? Furthermore, my reply was in response to NETROmac. If you have default install of WindowsXP and never install/uninstall any crap - guess what, it is also going to be as stable as hell!
In any case, thanks for pointing out the OBVIOUS regarding CoreGraphics file. When one breaks the system, like I did, one usually finds out AFTER the fact - not before.
Originally posted by klinux
Well, no shit, Sherlock. Tell me you never destablizied your system (in any OS) by installing and uninstalling various software?
ummmm... actually yeah... there are people who go years with installing "crap" from version tracker without effecting system stability...
the reason is beacaus Mac OS applications are PREDICTABLE and they (usually) only install stuff into one place so you don't have to go hunting to uninstall them... of course it goes without saying that if the developer does a bad job and puts shit all over your hard drive you will have problems... the difference is that putting shit all over your hard drive is standard procedure in XP thats why you need to have an uninstaller... Mac OS programs (good ones anyway) don't need them...
also hacking the os and installing programs are 2 different things...
In any case, we are diverging from the topic at hand...
Originally posted by Amorph
Isn't that great at living up to the hype.
This is simply techier-than-thou BS. You seem offended that Linux isn't living up to your high technical standards and yet you're happy to jump onto the bandwagon of a barely coherent troll. What's up with that?
And for future reference, being better than any MS OS *is* a complement.
Right, but BSD is more free (the license is less restrictive than GNU's), more secure, more stable, and it ran on just about everything back when Linux was still stuck on x86.
Care to explain how this extra 'freeness' provides *any* benefit.
You appear to have some disdain for open source software in general (if you pay money for it, it must be good!) and a grudge against the GPL in particular. (Later you bizarrely claim BSD licenced software is "held to a higher standard" than GPL'd software.)
You may want to meditate on what about the GPL helped Linux overtake BSD (in a variety of ways) from a standing start. Hint: why do the companies pouring billions(!) of dollars into Linux not want that extra 'freeness'?
As for 'stuck on x86', I'll come back to that later.
The administrative software runs like crap on Linux, because of limitations in Linux - in the core and elsewhere.
Now *that* is a lame cop-out. What the hell is this admin software doing that Apache, Mozilla, Emacs, many massive supercomputer clusters, Sharp Zaurus, Tivo etc. etc. aren't?
Fair enough. I'll point you to a hands-on report by a guy with years of in-depth Linux experience:
http://chesterforums.keenspot.com/viewtopic.php?t=50202
So much experience that he lost a month of data in a server crash?
Note that some of the reasons simply come down to discipline: The folks behind BSD wait two years before declaring a codebase stable, so when you install a stable build of BSD, you can be pretty damn sure that the bugs have been worked out.
Which BSD? Which Linux distribution are you comparing it with? Wouldn't you have more confidence if large numbers of people had actually been using the system for two years? You do realise that you are allowed to use other versions of Linux than the one in Linus's cvs tree?
That's a good working thesis, and it's certainly a big reason why MS is afraid of free software in general. But the same arguments apply to BSD.
No they don't, because BSD isn't threatening MSFT. You'll notice that (like yourself) MSFT reserve special malice for anything GPL'd.
But the "expensive" criticism really doesn't apply.
Yes it does. Running commodity OS's on commodity hardware (the x86's that you appear to rate as much as Linux) takes a major chunk out IT budgets and gives you leverage with suppliers, who no longer have you tied to their expensive hardware and niche OS and homebrew software tools. They are increasingly being asked to justify the cost differential and *failing*.
And we do it by paying for software that we can just drop in and expect to work.
How does "paying money == better" any more than "free == better"? Why does paying for some software mean paying for all software?
Your fervent enmity for Linux in particular and open source software in general is simply incomprehensible, particularly given your praise for the BSDs. Did the Linux geeks bully you at school or what?
I don't know if that will be enough to get Mac OS X noticed by IT professionals, but it has worked wonders with UNIX developers.
Originally posted by klinux
Yes, I am now aware it did but why is cursor part of that all important CoreGraphic file?
Because the cursor is a fundamental part of the interface. Because Apple never intended for anyone to change the cursor graphics. Therefore, you change it, and the system won't boot.
This monitors uptimes of companies servers. The first example is apple (switching over to X back in 2000) and the second example which is of more interest to this discussion shows at least one company dropping linux for HP-UX. I don't know much about the other OS's, but they do seem to be liking their update.
If you look at their current systems, it looks like through a bunch of sellings and stuff, they ran the gamut with OSes
Originally posted by dobby
Most Corporate IT's have heard about Apple.
At my previous job, our Apple gurus tried to push OS X in a big way. A single problem hindered rolling out Mac to PC for over 3000 Staff members.
You can't get your password to expire after 30 days!
I don't know what version of Mac OS X Server you were using but we're using v 10.2.6 and in Workgroup Manager under password options you can definitely have a password expire.
In Windows, this is quite hard to see as the UI is welded to the OS. However in linux and MacOS X its quite easy to see the difference. Your linux kernel and support files = the OS, KDE, XFree4, Gnome etc = the UI or Darwin = the OS, and Aqua = the UI.
Linux OS is good. It has the same roots as FreeBSD, and is maintained by some of the best programmers on the planet. As is BSD, and Darwin. However some of the "bolt-on" UI's for linux are crap.
Simarly, some people forget where darwin comes from, and continually assume that MacOS X cannot do something just because there isn't a nice GUI for such a task.
If I were responsible for a network, I would use macos X purly because in the last 4 months of owning this iMac, it hasn't crashed once!!
Originally posted by Code Master
http://uptime.netcraft.com/
This monitors uptimes of companies servers. The first example is apple (switching over to X back in 2000) and the second example which is of more interest to this discussion shows at least one company dropping linux for HP-UX. I don't know much about the other OS's, but they do seem to be liking their update.
If you look at their current systems, it looks like through a bunch of sellings and stuff, they ran the gamut with OSes
Code Master, it's easy to manipulate Netcraft's "what's that site running?" function by having an intermediary server running a different OS. Check www.goatse.cx's uptime if you don't believe me. (But whatever you do, don't actually visit the site. I can't stress this enough!)
Rumor has it that Microsoft still runs Hotmail off of FreeBSD because it's simply impossible to get the stability and flexibility for such a huge site to run on WinNT/2000. They just put an intermediary Win 2000 server in the way so that anyone who asks "what OS" will be fooled into thinking that MS eats its own dog food.
stupider...likeafox wrote:
This is simply techier-than-thou BS. You seem offended that Linux isn't living up to your high technical standards and yet you're happy to jump onto the bandwagon of a barely coherent troll. What's up with that?
Huh?
You appear to have some disdain for open source software in general (if you pay money for it, it must be good!) and a grudge against the GPL in particular. (Later you bizarrely claim BSD licenced software is "held to a higher standard" than GPL'd software.)
You're reading all kinds of really bizarre statements into my post. First of all, for the record, I've known of and admired and used the FSF's cause and software for about 15 years now. I do not disdain it. Furthermore, I specifically said that most GNU software is written to a much higher standard than Linux is, so your parenthetical claim is simply false.
Also, the "if you pay money for it, it must be good" is false from my post, trivially, or I would have said that Windows was preferable to Linux and BSD both. I didn't. However, when you're looking at a solution, you should look at the overall cost. A solution with an upfront cost and low support costs might be better than a solution with no upfront cost and higher support costs. It depends on what you need. But the original poster's assertion that Linux is automatically cheaper just because it has no upfront cost is false.
You may want to meditate on what about the GPL helped Linux overtake BSD (in a variety of ways) from a standing start. Hint: why do the companies pouring billions(!) of dollars into Linux not want that extra 'freeness'?
You might want to meditate on what the AT&T suit did for BSD's momentum. Before Linux got anywhere, Sun, Digital and many other companies were pouring money into BSD. Just to stay on AT&T's good side, they all migrated to SVR4.
Linux' success had very little to do with the GPL, although the GPL didn't hurt.
The administrative software runs like crap on Linux, because of limitations in Linux - in the core and elsewhere.
Now *that* is a lame cop-out. What the hell is this admin software doing that Apache, Mozilla, Emacs, many massive supercomputer clusters, Sharp Zaurus, Tivo etc. etc. aren't?
Talking to an expensive, high-bandwidth piece of hardware at the driver level?
So much experience that he lost a month of data in a server crash?
Read the article and draw your own conclusions.
Which BSD? Which Linux distribution are you comparing it with? Wouldn't you have more confidence if large numbers of people had actually been using the system for two years? You do realise that you are allowed to use other versions of Linux than the one in Linus's cvs tree?
That doesn't mean the older releases are as rigorously tested as FreeBSD's (the BSD the article refers to, with the two year testing period). Last I checked, Yahoo.com ran on FreeBSD, so I'd say their releases are very well tested.
You can't ignore the basic problem posed in the article: The fix-by-incremental-upgrade model doesn't work well. This isn't a free software issue, it's a discipline issue. It's the same reason AIX is far more secure than Solaris, even though they're both variants on the same codebase.
No they don't, because BSD isn't threatening MSFT. You'll notice that (like yourself) MSFT reserve special malice for anything GPL'd.
BSD doesn't have the "buzz" or whatever that Linux has, but MS is going after open source in general, and the GPL in particular, because of the restrictions against commercial use - and, since you seem to have gotten the impression from somewhere that not being impressed by Linux equates to a raging hatred of the GPL and everything it stands for, I will emphasize that I have no problem whatsoever with the GPL. I admire its means, and its ends, and unilaterally oppose MS' attempts to attack it. They might not care so much about the BSD license (and licenses like it, such as Perl's) just because those licenses don't prevent or restrict interaction with commercial code.
And we do it by paying for software that we can just drop in and expect to work.
How does "paying money == better" any more than "free == better"? Why does paying for some software mean paying for all software?
It doesn't. You're drawing a complete non sequitur from my argument. I'm am arguing against the two points that "free == better" (by saying only that "free doesn't necessarily mean better," and using my own shop as an example), and that Linux is some exemplar of either free software or free OS' that can run on commodity hardware.
On the other hand, I'm currently trying to leverage the Apache Foundation's XML efforts to the greatest degree possible, with an eye toward eventually getting something like Mozilla to render the front end. Because, you see, I've used and admired open source software for 15 years or so, whenever it's been the best tool for the job.
Your fervent enmity for Linux in particular and open source software in general is simply incomprehensible, particularly given your praise for the BSDs. Did the Linux geeks bully you at school or what?
Linux didn't exist when I was in school.
Originally posted by Amorph
You're reading all kinds of really bizarre statements into my post.
My apologies for getting the wrong end of the stick, I was thrown off by your inital agreement with inkhead (who by the way, does not have a point).
It's easy to get lost when you have to change gears from a rant about Linux that criticises the GUI (i.e. Gnome/KDE or others) and a mish-mash of other classic trolls, to one that criticises Linux kernel development but praises the BSDs which share (or cross-pollinate) a great deal of code, philosophy, software, users etc. with the Linux distributions that inkhead was attempting to rubbish.
I think therefore the much subtler point you were making (that I missed completely) is that the Linux (kernel) development model of continual improvement from a low starting point isn't as good as the engineering excellence of the BSD and others (not-necessarily open-source).
You ask
It's "good enough." Haven't we had enough of "good enough" at this point?
And the answer is no. We've had enough of "good enough" software that puts its competitors out of business with shady practices, of "good enough" software that gets sold on the basis of powerpoint presentations and locks you into a becoming the cash cow of a lazy company with no motivation to innovate or improve.
If your future choices aren't going to be impeded by the current choice then good enough software is exactly what you want.
Also, you claim that the fix-by-incremental-upgrades model doesn't work but sometimes worse *is* better
more on worse-is-better...
Originally posted by stupider...likeafox
My apologies for getting the wrong end of the stick, I was thrown off by your inital agreement with inkhead (who by the way, does not have a point).
I was implicitly ratcheting down his language, which was probably the source of your confusion. Insofar as he states that Linux is not a great choice for a lot of enterprise work, I have to agree. You can massage it up to the point where it is, sometimes, but that's not really a recommendation.
It's easy to get lost when you have to change gears from a rant about Linux that criticises the GUI (i.e. Gnome/KDE or others) and a mish-mash of other classic trolls, to one that criticises Linux kernel development but praises the BSDs which share (or cross-pollinate) a great deal of code, philosophy, software, users etc. with the Linux distributions that inkhead was attempting to rubbish.
I don't know how much philosophy BSD shares with Linux, or for that matter how much Linux shares with much other GNU code. There are choices beside the cathedral and the bazaar, which both the BSDs and the FSF itself have taken (Apple, of course, is erecting a cathedral with its doors flung open to the bazaar outside). I am unimpressed with GUI efforts like Gnome for many of the same reason that I'm unimpressed with kernel efforts like Linux. (Which, again, is not to say that Gnome is worthless, or should be discarded, or anything similarly absolutist. Lots of neat tools and libraries have come out of that effort, just as many cool and useful things have come from the Linux development effort. But the "bazaar" development model is better suited to tools and libraries than to large-scale projects like kernels and GUIs.) When Linux came out, and I saw the development model, I wondered what happened to the GNU Project's once-reknowned coding standards. I wonder how many recent converts to open source even know that the GNU Project has coding standards.
You ask ["Haven't we had enough of 'good enough'?"]
And the answer is no. We've had enough of "good enough" software that puts its competitors out of business with shady practices, of "good enough" software that gets sold on the basis of powerpoint presentations and locks you into a becoming the cash cow of a lazy company with no motivation to innovate or improve.
If your future choices aren't going to be impeded by the current choice then good enough software is exactly what you want.
This smells of fatalism, which is basically just the passive tense blown up into a worldview. It also seems to miss one of the whole advantages of open source, which is that it simply doesn't go away. The BSD community doesn't have to be immense, it simply has to be big enough to keep the codebase up. There's really not that much development left to do on it. It's been mature for over a decade. So I wouldn't hesitate to adopt it, even if Linux runs away with mindshare. I certainly didn't abandon the Macintosh just because Windows ran away with all the market share, and the Mac is tied to one company's fortunes.
Also, you claim that the fix-by-incremental-upgrades model doesn't work but sometimes worse *is* better
Certainly, advantages and disadvantages are situational. However, I don't believe that worse is better in the case of Linux, and in particular in the case of Linux as a useful alternative for enterprise. I doubt the platform would suffer if someone decided to take a given branch, battle-test it for two years, and then release a "stable" distro with the result. Many of BSD's advantages are simply the result of discipline over time. There's nothing preventing their best practices from being applied to Linux or Gnome or what-have-you. You wouldn't even have to change the way the main trunk was managed (although, obviously, that would help).
[edit: Having read the article above, I don't see the relevance. That's less about quality of implementation and more about philosophy of design. For the record, I have always fallen firmly into the New Jersey camp. But if you do implement something, you do it right. The original UNIX source was compared to poetry and assigned literary value - that quality, in fact, is what spawned the open source movement in the first place. It is also not a common trait within the Linux codebase, even allowing for Dennis Ritchie's NJ-ish maxim that any nontrivial program contains at least one hack.]
Now I want to argue that worse-is-better is better. C is a programming language designed for writing Unix, and it was designed using the New Jersey approach. C is therefore a language for which it is easy to write a decent compiler, and it requires the programmer to write text that is easy for the compiler to interpret. Some have called C a fancy assembly language. Both early Unix and C compilers had simple structures, are easy to port, require few machine resources to run, and provide about 50%--80% of what you want from an operating system and programming language.
Half the computers that exist at any point are worse than median (smaller or slower). Unix and C work fine on them. The worse-is-better philosophy means that implementation simplicity has highest priority, which means Unix and C are easy to port on such machines. Therefore, one expects that if the 50% functionality Unix and C support is satisfactory, they will start to appear everywhere. And they have, haven't they?
Unix and C are the ultimate computer viruses.
A further benefit of the worse-is-better philosophy is that the programmer is conditioned to sacrifice some safety, convenience, and hassle to get good performance and modest resource use. Programs written using the New Jersey approach will work well both in small machines and large ones, and the code will be portable because it is written on top of a virus.
It is important to remember that the initial virus has to be basically good. If so, the viral spread is assured as long as it is portable. Once the virus has spread, there will be pressure to improve it, possibly by increasing its functionality closer to 90%, but users have already been conditioned to accept worse than the right thing. Therefore, the worse-is-better software first will gain acceptance, second will condition its users to expect less, and third will be improved to a point that is almost the right thing. In concrete terms, even though Lisp compilers in 1987 were about as good as C compilers, there are many more compiler experts who want to make C compilers better than want to make Lisp compilers better.
The good news is that in 1995 we will have a good operating system and programming language; the bad news is that they will be Unix and C++.
Re-read this bit and think Linux, remembering that this guy is insulting (sorta) the Unix and C approach as he comes from a Lisp background.
You think Linux isn't very good. This guy is explaining why it will succeed because it isn't very good, though he's not talking about Linux in particular and it's probably best to substitute "run on cheap commodity hardware" for "portable".
[Linux] provide[s] about 50%--80% of what you want from an operating system
Therefore, the worse-is-better software [Linux] first will gain acceptance, second will condition its users to expect less, and third will be improved to a point that is almost the right thing.
Half the computers that exist at any point are worse than median (smaller or slower). [Linux runs] fine on them.
A further benefit of the worse-is-better philosophy is that the [Linux] programmer is conditioned to sacrifice some safety, convenience, and hassle to get good performance and modest resource use.
... there are many more [kernal hackers] who want to make [Linux] better than want to make [VMS] better.
First, that this approach has limits. His assertion that the "worse-is-better" approach eventually runs as well on big hardware as on small is false. This approach scales down nicely, and scales up indifferently. Conversely, the "better is better" approach (which VMS nearly exemplifies) scales down poorly. VMS' filesystem is unbelievably robust, but the tradeoff is that it's complex and difficult to write for. When you have a big box running a mission-critical database under heavy load, that complexity and difficulty is a worthwhile tradeoff. But for any other purpose I would much rather deal with UNIX' brilliantly simple "everything is a stream" paradigm. Apparently, I'm not alone in this assessment, because VMS has a significant share of the uptime-critical market (and a good chunk of the high-performance computing market, because its clustering technology has been substantially better than anyone else's for over a decade now), and UNIX and UNIX-like OS' dominate markets where reliability and efficiency are desirable, but not critical.
I would also argue against the idea that C++ evolved C closer to a "perfect" solution. Objective-C did that, New Jersey-style, and look where it went. C++ exemplifies neither the MIT approach nor the New Jersey approach; it exemplifies design by committee and nothing else. It certainly isn't easy to write a compiler for!
Finally, again, there is the question of elegance of implementation. Again, what he's talking about is design. C doesn't do everything, and it doesn't try to, but it's clean and simple. UNIX doesn't do everything, and it doesn't try to (although System V made a credible lunge in that direction) but it was beautifully written. Ditto BSD (and its associated technologies, such as threading and TCP/IP). But the New Jersey style emphasizes elegance of implementation over completeness. Linux emphasizes neither - if anything, it's concentrated on speed optimization in a few areas, on portability (which does sort of dovetail with the NJ style) and on a wealth of user-level tools. But it's fundamentally different in approach and in philosophy from the other examples cited - not to mention in quality - and so I don't think the analogy holds directly. Linux is a different kind of worse, and to the extent that it actually succeeds in getting to the next level (as BSD had, before the lawsuit - it's hard to remember now, but in the late 1980s BSD was a major force with a very high rate of acceptance) it will have to realign itself to the NJ style. Or, it'll succeed by sheer momentum, and we'll have another Windows on our hands - but nothing of the quality produced by the likes of Thompson and Ritchie and Jordan Hubbard. And that's what I meant by "haven't we had enough of 'good enough'?" I consider the NJ style to be excellent, not 'good enough'. The MIT style still has to answer for MULTICS.
I have Redhat 8. I can't see anywhere Linux would be better than BSD since that is all Linux is good for anyway, command line apps. The GUI is much worse than Windows 95. In fact it is a poor copy of Windows. The Virtual Desktops are nice but hey OS X has them. MacOS X is direct competition with Linux on the desktop and it wins. BSD wins on the server side. Why don't the Linux people start writing for Darwin/FreeBSD and help Apple fight M$? I don't see how *NIX divisions help anyone. The BSDs should unite, Linux should die, and everyone should help Apple, at least until M$ is no longer a monopoly. Apple should also return the favor: if Safari is free why isn't it open source, for example? iCal? iChat? iPhoto? Address Book? TextEdit? Hell Stickies? There are many small things that could be enhanced added or fixed that Apple might see in a submitted version and like and add to the apps in an update. As long as there is an agreement not to use the Apple code for commercial gain I don't see how it would hurt Apple, it could only help them and the open source community.
Originally posted by Aquatic
Inkhead was playing Devil's Advocate I admire him. No one ever gets to say anything bad about Linux.
There are plenty of uninformed Linux cheerleaders about. That fact and claims of playing devil's advocate don't give you carte blanche to be wrong about things e.g. the fact that no-one ever gets to criticize Linux.
I see it all the time, just like I see Mac users criticize Apple all the time. However, no-one likes their dirty laundry done in public though, *especially* by people with no clue what they're talking about (see almost any article on Apple by John C Dvorak).
Which is "Line Ux" because it's Linux like Linux and Lucy, not Linn Ux.
That doesn't really make any sense, but if it did it would still be wrong. We covered this recently try the search function.
I have Redhat 8. I can't see anywhere Linux would be better than BSD since that is all Linux is good for anyway, command line apps.
What about if you were creating a platform for home entertainment system that you wanted to share with other big businesses with no fear that one would fork it and take it closed source? Like Sony and Panasonic have. Or does your entire conception of computer OS's and what they are used for stretch from Windows 95 to Mac OS X, with a vague idea about "servers"?
The GUI is much worse than Windows 95.
Which GUI? The one on your phone, on your tivo, on your PDA, on your network appliances, on your home multimedia system, or what? Even if your only talking about Red Hat 8 there are many different GUIs: Gnome, KDE, Ximian etc.
Looking at the beatufilly anti-aliased fonts on my screen I'd say that they are better than Windows 95, but I'm a techy, so I suppose you'd better clarify for who and for what it is so much worse.
MacOS X is direct competition with Linux on the desktop and it wins. BSD wins on the server side.
Did you miss the discussion of how being "the best" doesn't help you "win"?. Reminder: 95% Windows: 5% Mac.
Why don't the Linux people start writing for Darwin/FreeBSD and help Apple fight M$? I don't see how *NIX divisions help anyone. The BSDs should unite, Linux should die, and everyone should help Apple, at least until M$ is no longer a monopoly.
You may have missed it, but basically they are. The vast majority of code above the kernel is not technically Linux and is used in the BSDs and many other crazy operating systems you've never heard of. More importantly they have no reason to not interoperate so I can, for example control several X11 or command line programs running on several *nix boxes through my Mac.
Apple should also return the favor: if Safari is free why isn't it open source, for example? iCal? iChat? iPhoto? Address Book? TextEdit? Hell Stickies? There are many small things that could be enhanced added or fixed that Apple might see in a submitted version and like and add to the apps in an update.
This would mostly be stupid for Apple. As long as they keep using open standards (which almost everything you mention does) and using open source effectively to advance their own economic interests (which they are, see Safari WebCore and Rendezvous for examples) then you can't ask for much more.
As long as there is an agreement not to use the Apple code for commercial gain I don't see how it would hurt Apple, it could only help them and the open source community.
If you can't use it for commercial gain it's not open source; it's part of the definition. #6 to be precise.
http://www.opensource.org/docs/definition.php
Originally posted by Amorph
Apparently, I'm not alone in this assessment, because VMS has a significant share of the uptime-critical market (and a good chunk of the high-performance computing market, because its clustering technology has been substantially better than anyone else's for over a decade now), and UNIX and UNIX-like OS' dominate markets where reliability and efficiency are desirable, but not critical.
Here you think they're doing well because they're hanging onto their most valuable customers: a classic mistake.
Linux is a disruptive technology (I recommend the book of the same name, which doesn't directly address Linux) that steals the customers you least want first. However, in time it grows and steals so many of your customers that those that remain start to switch from the "better" product simply because of the costs of staying in a niche (assuming that the "better" product isn't simply overtaken in quality as the market focus shifts).
Some more examples:
http://www.disruptivetechnologies.com/dtcases.html
I would also argue against the idea that C++ evolved C closer to a "perfect" solution. Objective-C did that, New Jersey-style, and look where it went. C++ exemplifies neither the MIT approach nor the New Jersey approach; it exemplifies design by committee and nothing else. It certainly isn't easy to write a compiler for!
C++ is not "perfect", it's "worse". But then if it isn't "better" than Obj-C then why is Obj-C such a niche technology today, even with NeXT's, then Apple's championing (without which it would have died a lonely death)?
You're right that Obj-C is much, much "better". That is why it is "worse".
The design-by-commitee stuff is by-the-by, it was designed to ride upon the virus of C just like Unix was. The crux however is not technological in this case but psychological as C programmers wanted to keep on writing C while claiming to be doing OOP because they used a C++ compiler. No one wanted to really embrace OO, but no-one wanted to admit that either. (I am aware that plain C compiles as Obj-C btw)
Which GUI? The one on your phone, on your tivo, on your PDA, on your network appliances, on your home multimedia system, or what? Even if your only talking about Red Hat 8 there are many different GUIs: Gnome, KDE, Ximian etc.
My cell doesn't use Linux. Linux PDAs are not as good as PocketPCs, Symbians, or even (blech) Palms.
Did you miss the discussion of how being "the best" doesn't help you "win"?. Reminder: 95% Windows: 5% Mac.
Exactly that is Amorph's point. Linux is "good enough." Just like Windows. Do you want to put up with that? I don't and I don't want to see one mediocre product replace another even if it's not controlled by an Evil Company. I like quality that's why I'm typing on a 12" PowerBook.
You may have missed it, but basically they are.
No they aren't. The average end user doesn't even know what compiling is let alone how to do it. Thus you need to find binaries for each NIX flavor. NIX is very fractured and Linux just makes it worse.
If you can't use it for commercial gain it's not open source; it's part of the definition. #6 to be precise.
Webcore is opensource but it is being put in a commercial product, OmniWeb.
I have not read a single article that calls Linux what it is: mediocre. Slashdot doesn't diss Linux like we diss Apple. And anyway Apple's software quality is lightyears ahead of anything on Linux. They just don't get it. It's about the interface.