That's his point. Instead of a monolithic developer/support source, you've got Novell, RedHat, IBM, Sun...
This can be a good thing, for sure. One company goes out of business, but the whole thing isn't ruined. But there are true downsides to Linux; everything has its flaws. The thing that impressed me about this is that there are really no ``bad'' quotes here. Ballmer says that Linux is just the next new thing, a potential flash-in-the-pan (which, personally, I think may be right, if we aren't lucky)--and that Microsoft must win by responsibly offering a better product and better customer care.
The issue here is tricky, though. I don't think the MPAA or the DVD Assn. could prevail in court with the claim that watching DVD's through DeCSS on a Linux machine after paying a license for the decoder anyway. It's even a little disingenuous to claim that that's the big issue, because it isnt. But they can win against those who distribute DeCSS, on the grounds that it allows people to not even buy the license for the decoder, a license that is required to play DVDs.
Personally, I don't understand why I even need to pay for the decoder. If they sell me a DVD that I can't play, that's my own damn problem. If I find some ingenious other use for it--as a coaster, a wall decoration, or input for DeCSS--all the better. And if I'm lucky enough to even find a way to put it to the original use, without paying for the license, so what? The point of copyright is to commoditize intangibles, so that they can be sold in the albiet-poorly-fitting manufacturing economic model. Which is why it's illegal to copy DVDs and distribute them. Which I agree with fully. But why I can't do whatever I want to do a DVD after I buy it--so long as I don't redistribute a copy of the original--should be fine.
I know that Maya is not Windows-only, but the licenses we have for it are. So if we want to have Photoshop (for texturing), Maya, and, say, Macromedia Director on the same machine (a common toolset for certain classes), we can use Windows or Mac. And Macs cost more for the hardware. My other examples were also perfectly fine; XILINX, I believe, is Windows-only. All I was saying is that while most general programming and computer science courses rely on Linux or Sun, a few do rely on Windows for obvious reasons (such as not having seperate machines for Maya and Photoshop). I should have been more clear in my examples.
What's your point? That VisualBasic is in fact better than I thought, or that those who use it aren't always bad coders?
If the former, I readily admit all I know of VisualBasic is by reputation. I've worked with it perhaps once, and while I wasn't impressed, that's hardly enough experience for me to comment.
If the latter, I'll admit that the stereotype is probably often unfair. The point he was making, though, is that the apparently objective technical inferiority of VisualBasic does not mean there is not, sometimes, a subjectively good reason to use it. You would be proof of that; despite the technical experience to use something ``better'', for rapid application development of some simple GUI tool, there's just no reason to do it in C.
I don't think this fully resolves the question of wether VB is truly as good as it can be--not to say its not sometimes the best for the job--which is really the more pertinent question if one is trying to discuss Microsoft's technical expertise. However, from what he told me--he's going to be working in the compilers division, apparently on VB--VB has some nifty tricks that do deserve some credit.
Interesting argument, and certainly novel compared to many. As you probably guessed, I'm not a huge fan of their products, either; in fact, I don't own or use any. As for your comments about academia specifically, when it comes to computer science, this has not been my experience in the least. PowerPoint being a rare exception--most materials are distributed in PDF, PS, or Tex--nearly all the computer science classes I've taught focus on platform independent technologies or, when specific platforms are necessary, frequently Linux on x86 (formerly Solaris/SPARC). If Linux has gained ground in one area the most, that area is surely academia.
That is not to say Windows is never used, of course--certain applications (3D modelling software like Maya, chip-designing software, and so forth) are Windows-only. But while I certainly have encountered the attitude you are talking about, it is, in my experience, far rarer than the smugness one encounters in non-Windows users (and I'll admit, I demonstrate that same smugness).
Regardless, the users are hardly to blame for the company; the company seems to have made, if not always the right business decisino, than at least enough of the right decisions. That was my point, that even though I don't use or like most of their products, Microsoft is, apparently, the very picture of corporate success.
You'll notice I didn't argue that there was nothing wrong with those activities;)
Nonetheless, what MS did is something that, essentially, would not have been nearly so illegal had it not been so successful; many who are not MS apologists might still admit that others who have been wronged by Microsoft have done just the same thing--Apple, RealNetworks, and Netscape. Never is this justification for illegal activities, but it should be noted, too, that what Microsoft did--the strictly illegal parts--were statutorily illegal, in the sense that speeding is statutorily illegal; I personally do not find much of it to be morally repugnant in the way that rape or murdur are morally repugnant (to get some perspective here). A corporation has no spirit, no ethos, and no criminal culpability (despite the claims of corporate criminal law). A corporation exists for the sole pursuit of profit.
As for morality, it's a tough call what morally is right. If Microsoft stimulates the economy, provides jobs, and pushes innovation, perhaps they aren't so bad. On the other hand, if they drive rivals out of business, crush competition, and stifle innovation, they seemingly leave a trail of broken enterpreneurs in their wake. In other words, I don't know.
Regardless, I was arguing initially that Microsoft aren't so stupid. Not that they are the ``good guys''.;)
You know, it's funny that this is a discussion now, because I had the same conversation with a friend of mine last week. My friend--who's really a very good programmer and generally intelligent person--is going off to a promising job at Microsoft in a short while. And just before the conversation, I'd attended a Microsoft recruiting information session. and I have to admit, I was impressed. Their recruiters were intelligent; their corporate culture--if we can talk about such an elusive property--was compelling, and every single person I know who has or does work for them has been far above average for smarts, not just grades, accolades, or resume points, but for true, geeky, smarts.
So I asked my friend, if MS seems to be such a great company--and they really do, from this angle--why are their products so damn bad? His answer was, who says they're bad?
I told him, ``well, I'll admit Windows fills its niche just as well as anything else--if only OS X worked on x86--but c'mon, look at some of their other stuff!'' Like VisualBasic, he said? Sure, you or I wouldn't use it, but we aren't the target market. The guys using it for rapid development in the corporate world aren't coders. They don't care about performance or semantics or control; they care about something they can use and be done with. And (according to my friend--I've barely used it) VisualBasic fits the bill.
What about Windows Server, I asked. I mean, here's a product that simply sucks. Compare it to Linux, FreeBSD, or any number of commercial Unix offerings, and while it might win on usability or TCO, the effort of making it actually work as well as the alternatives is hardly worth it. He said, yeah, or PocketPC, right? I said, absolutely. Hell, I doubt MS even made money on PocketPC.
He agreed with me, but said, well, think about this. They don't want to make a whole new OS--and they could--they want to make a Windows for a server. They're tied to the legacy code. So while it might not be pretty, Windows Server likely truly is the best they can do with that legacy code, just as is PocketPC.
I suppose Microsoft might seem to be the bumbling 300 pound gorilla at times, but I can't argue with results. Can you?
So in other words, its the same as the locate database on Unix systems. Which I and most other Unix users leave on. I mean, seriously, how much is a few more megs?
Yeah, I know. I was just trying to clarify that the grandparent poster didn't likely mean the laws of physics would be turned upside-down or anything. And that he likes Warcraft, apparently.
I think his point was that if the Bible were to be regarded as containing at least some historical fact, some scientists (not physicists or computer scientists, but anthropologists and archaeologists, who probably want us to be respectful of them and still refer to them as scientists) would be more interested in what the Bible has to say, not because they'd suddenly become believers, but because they may feel that it has historical value in the same way that the Ilead and the Odyssey have historical value, and that he doesn't really care because either way he's going to keep on playing Warcraft. Whatever the hell that is.
I seem to recall a few security bugs in PostNuke, though I'm sure it's a big step from PHP Nuke. I also recall that the main developers left the project.
Anyway, as for security, a lot of these PHP systems have issues. PHP itself isn't the most secure, but it's not really at fault, I don't think. I think, to be honest, that a lot of these systems are written by guys who aren't coders, and it shows in the code (the one's I've looked at, at least). I messed with Drupal, incidentally, for some contract-work; I found it to be quite mature, stable, and reasonably organized; the code was nicely object-oriented and the whole project was quite professional. Comparitively, I think the ones by web-designers who learned to code on PHP simply show the lack of organization in terms of bugs, efficiency, and security.
As for my own site, I use a hack of a kluge of a PHP knockoff of Blosxom. It's far, far from elegant, and could really use a total rewrite (which I keep telling myself I'll make time for, but hey, if it ain't broke...). Ah, well.
Strange, I did forget it. I mentioned in in my post just prior, though.
Have you had a lotta luck with it? I was able to run Linux on Linux (but I may as well use UML then), and it was quite fast, but I didn't have Win98 handy to try, which as I heard it was the only Windows to work on Qemu (and even then, quite unreliably). Know anything about that?
VMWare is the most popular commercial one (for Linux and Windows; VirtualPC would be the one to try on Apple--unless you just want to emulate PPC on PPC, i.e. run OSX on PPC Linux, in which case Mac On L inux is for you). Bochs is the leading open source contendor, in that it emulates a complete x86 machine, and works on any architecture (SPARC, Alpha, PowerPC, etc). However, because of that, it's quite slow, and is far more useful for things like reverse engineering or OS testing than actual desktop use (i.e., if you wan't to see registers in use, it'd be great; if you want to watch a movie or use Photoshop, don't bother). And of course, there's always WINE, which runs a number of Windows programs on Linux quite well.
That's odd. Admittedly, it's not unheard of (UML requires a kernel module, but then it's not a hardware emulator, either). But you'd think if it runs with host OSes of FreeBSD, Linux, Windows, and OS/2, it wouldn't rely on anything needing a kernel module; I have no idea how one gains that functionality in Windows or OS/2 (though perhaps this is easier than I am guessing). And I can't help but wonder, then, whether it does hardware emulation in the vein of Bochs, or system call translations in the vein of Qemu. I suppose it's a fair guess that it's the latter, though, since it appears to be x86 only (no mention of support for OSX, which should be easy enough if FreeBSD is supported).
Ah, well. Trust OS News to be short on technical details. Or even on proper grammar.
I take it back. According to the Washington Post article, they do indeed mean supercomputer clusters.
A little background on my mistake; I received a notice about this at the end of last week, commented that my university had been targeted as well, but it was implied (details were very scarce, and the link I had been given to the stanford site was wrong) that the targets were simply general purpose computing resources.
That's simply not true. Read the freakin' article.
In a number of cases, the attacker gained access by sniffing a password off the wire (granted, insecure services should be eliminated, but try telling that to users who rely on them) or simply guessing an insecure password. After gaining local access, in some cases access was escalated through a known vulnerability, but in other times simply through password cracking.
So what excuses are there for the admins? First, securing against authenticated users is reasonably hard. It certainly can be done, but in many cases it's understandable that admins might not fully achieve such security. Coupled with the inherent trust most grant their users, this is to be expected. Second, users choose bad passwords. It happens. One can enforce secure passwords--many places do--but just like eliminating insecure services, it ticks off the users, and as a result some servers won't implement this. And as for patching, again, a department may not have budget for backup machines and failover; a certain patch--say, one against a local vulnerability only exploitable by authenticated users--may not get applied for a little time.
Point being that even while reasonable precautions are usually taken, there are often still vulnerabilities for the determined attacker to exploit. Typically, the issue is much less laziness than a desire to keep accessible to the users, which is the entire point, really, of the machines to begin with. As for your comment about rootkits, we only know about the machines that we know about, and even those were often quite well covered.
Try not to blame the victim. You or I could be next.
I don't think they mean clusters as in MOSIX, etc. The term seems to be used frequently in academia to refer to a group of machines, with load balancing between them, used for services like shell access, web and mail serving, etc. Additionally, individual servers are being attacked as well. Many schools have a very, shall we say, fragmented IT infrastructure; I'm at a medium-sized private university (about 10,000 undergrads, perhaps) with four different undergraduate schools and perhaps twice as many graduate schools. Each has its own IT department. The larger ones are well-run, but some of the smaller ones aren't even on the newsgroup of which all the IT departments are supposed to belong because they can't figure out how to use the news server (or so it's been said, at any rate). Point is, academia has some great admins, and some psych professors running servers out of their classrooms.
Academic computing is the epitome of *available* computing, in the sense that availability is the highest priority. Financial institutions may prioritise (or at least, should prioritise) security and a good administration over availability, but by its nature, academic computing involves disparate infrastructures, various levels of admins with various goals, and so forth. All students, faculty, and staff need access; frequently, granting loose, unsecure access is simply more efficient for the time being than making things secure. Such is life.
Considering that the Google Toolbar only works on IE, Google doesn't actually come across as THAT Linux-friendly (as a client; obviously, they use Linux industrially-speaking).
Re:Google: the Apple of the internet
on
Google's Next Steps
·
· Score: 2, Insightful
R&D candidates: sharks with lasers, giant death-ray on Moon, secret nuclear-powered carribean island laboratory...?
Sorry. Something about
So. They have this huge map of the Web and are aware of how people move around in the virtual space it represents. They have the perfect place to store this map (one of the world's largest computers that's all but incapable of crashing). And they are clever at reading this map. Google knows what people write about, what they search for, what they shop for, they know who wants to advertise and how effective those advertisements are, and they're about to know how we communicate with friends and loved ones. What can they do with all that? Just about anything that collection of Ph.Ds can dream up.
This has been around for well over a year. I work in such an organization, and we get MS licenses for $5 a pop. Granted, not as free as Linux, but I don't think our recipients would appreciate Linux PCs quite as much. Hey, I use Linux on my desktop full time. I don't even own a Windows machine. But those are the realities, I'm afraid.
I'm a very recent (about a month and a half) convert to OSX. My desktop is Gentoo, which I also use on my PC at work. I run a Debian server, use RedHat far more than I'd like at work, and run OpenBSD on my own server. I've also got similar experience running FreeBSD. I used to have Windows on my desktop as well, up until late last year, simply for the software compatibility. It was only recently that I decided that everything I did in Windows could be done without significant extra effort on Linux.
I had, prior to buying my 12" Powerbook, used OS9, and found it to be quite unpleasant in that the interface is perfectly friendly, if a bit archaic-feeling, but that there seemed to be a complete and utter lack of any real features in the area of networking, no multi-user capabilities, and generally poor reliability and usability. I had also used OSX a small amount, and it seemed to be, if a bit slow (and still does feel that way, even on my Powerbook), a perfectly tolerable Unix-ish OS.
Nevertheless, I suspect I'll have a Linux desktop around for a long time (and in some respects truly do prefer it over OSX). Firstly, I found switching to a closed-source OS to be a bit of a bother. Certain minor things I'd like to change I can't, at least, without significant work. While the initial installation is easy, and getting a working desktop for basic stuff like email, web browsing, etc, I can't change certain things how I'd like. The sec ond issue I have is more anticipatory--sooner or later, I just know Apple is going to bite me in the ass with upgrades. OS 10.1 users are apparently expected to upgrade to Panther. When 10.4 comes out, am I going to be expected to plop down another $120 just like that? And finally, in terms of usability, yes, OSX has many nice features. And it's pretty. But it runs X11 apps clunkily at best, Fink and OpenDarwin ports and all are great, but there's not NEARLY the range of Free software available (e.g. the lack of a non-alpha level, non-X11 Free, or even free, word processor). Closed Broadcomm drivers mean I can't put my Airport Extreme card into passive mode. And of course, I'm simply less familiar with OSX, as well. If I decide to implement GRSec and PAX protection on my Gentoo box, I can do it. Being a bit of a security hobbyist, how do I know my Apple is as secure?
OSX is great, to be sure, for a desktop. It patches the weakness of desktop Linux and is, in many ways, a paragon for that endeavor. And XCode and all are certainly good enough that I have no real issues doing development ON OSX, though most of what I write is written FOR Linux/x86. But I'd never use it for a server, rarely for anything truly serious, certianly not on a desktop where the price of Apple hardware is prohibitive, and not simply because Aqua apps look ``a hell of a lot slicker'':P
Slashdot Mirror
This can be a good thing, for sure. One company goes out of business, but the whole thing isn't ruined. But there are true downsides to Linux; everything has its flaws. The thing that impressed me about this is that there are really no ``bad'' quotes here. Ballmer says that Linux is just the next new thing, a potential flash-in-the-pan (which, personally, I think may be right, if we aren't lucky)--and that Microsoft must win by responsibly offering a better product and better customer care.
Nothing to see here. Move along.
I betcha plenty of Senators don't use SSL on their e-mail clients.
Personally, I don't understand why I even need to pay for the decoder. If they sell me a DVD that I can't play, that's my own damn problem. If I find some ingenious other use for it--as a coaster, a wall decoration, or input for DeCSS--all the better. And if I'm lucky enough to even find a way to put it to the original use, without paying for the license, so what? The point of copyright is to commoditize intangibles, so that they can be sold in the albiet-poorly-fitting manufacturing economic model. Which is why it's illegal to copy DVDs and distribute them. Which I agree with fully. But why I can't do whatever I want to do a DVD after I buy it--so long as I don't redistribute a copy of the original--should be fine.
I know that Maya is not Windows-only, but the licenses we have for it are. So if we want to have Photoshop (for texturing), Maya, and, say, Macromedia Director on the same machine (a common toolset for certain classes), we can use Windows or Mac. And Macs cost more for the hardware. My other examples were also perfectly fine; XILINX, I believe, is Windows-only. All I was saying is that while most general programming and computer science courses rely on Linux or Sun, a few do rely on Windows for obvious reasons (such as not having seperate machines for Maya and Photoshop). I should have been more clear in my examples.
If the former, I readily admit all I know of VisualBasic is by reputation. I've worked with it perhaps once, and while I wasn't impressed, that's hardly enough experience for me to comment.
If the latter, I'll admit that the stereotype is probably often unfair. The point he was making, though, is that the apparently objective technical inferiority of VisualBasic does not mean there is not, sometimes, a subjectively good reason to use it. You would be proof of that; despite the technical experience to use something ``better'', for rapid application development of some simple GUI tool, there's just no reason to do it in C.
I don't think this fully resolves the question of wether VB is truly as good as it can be--not to say its not sometimes the best for the job--which is really the more pertinent question if one is trying to discuss Microsoft's technical expertise. However, from what he told me--he's going to be working in the compilers division, apparently on VB--VB has some nifty tricks that do deserve some credit.
That is not to say Windows is never used, of course--certain applications (3D modelling software like Maya, chip-designing software, and so forth) are Windows-only. But while I certainly have encountered the attitude you are talking about, it is, in my experience, far rarer than the smugness one encounters in non-Windows users (and I'll admit, I demonstrate that same smugness).
Regardless, the users are hardly to blame for the company; the company seems to have made, if not always the right business decisino, than at least enough of the right decisions. That was my point, that even though I don't use or like most of their products, Microsoft is, apparently, the very picture of corporate success.
Nonetheless, what MS did is something that, essentially, would not have been nearly so illegal had it not been so successful; many who are not MS apologists might still admit that others who have been wronged by Microsoft have done just the same thing--Apple, RealNetworks, and Netscape. Never is this justification for illegal activities, but it should be noted, too, that what Microsoft did--the strictly illegal parts--were statutorily illegal, in the sense that speeding is statutorily illegal; I personally do not find much of it to be morally repugnant in the way that rape or murdur are morally repugnant (to get some perspective here). A corporation has no spirit, no ethos, and no criminal culpability (despite the claims of corporate criminal law). A corporation exists for the sole pursuit of profit.
As for morality, it's a tough call what morally is right. If Microsoft stimulates the economy, provides jobs, and pushes innovation, perhaps they aren't so bad. On the other hand, if they drive rivals out of business, crush competition, and stifle innovation, they seemingly leave a trail of broken enterpreneurs in their wake. In other words, I don't know.
Regardless, I was arguing initially that Microsoft aren't so stupid. Not that they are the ``good guys''. ;)
Locate does indeed run by a cronjob. I have no idea what the Windows one does, but I doubt it's truly random ;)
So I asked my friend, if MS seems to be such a great company--and they really do, from this angle--why are their products so damn bad? His answer was, who says they're bad?
I told him, ``well, I'll admit Windows fills its niche just as well as anything else--if only OS X worked on x86--but c'mon, look at some of their other stuff!'' Like VisualBasic, he said? Sure, you or I wouldn't use it, but we aren't the target market. The guys using it for rapid development in the corporate world aren't coders. They don't care about performance or semantics or control; they care about something they can use and be done with. And (according to my friend--I've barely used it) VisualBasic fits the bill.
What about Windows Server, I asked. I mean, here's a product that simply sucks. Compare it to Linux, FreeBSD, or any number of commercial Unix offerings, and while it might win on usability or TCO, the effort of making it actually work as well as the alternatives is hardly worth it. He said, yeah, or PocketPC, right? I said, absolutely. Hell, I doubt MS even made money on PocketPC.
He agreed with me, but said, well, think about this. They don't want to make a whole new OS--and they could--they want to make a Windows for a server. They're tied to the legacy code. So while it might not be pretty, Windows Server likely truly is the best they can do with that legacy code, just as is PocketPC.
I suppose Microsoft might seem to be the bumbling 300 pound gorilla at times, but I can't argue with results. Can you?
QoS is a Quality of Service. Presumably the Windows service just allows QoS traffic shaping.
So in other words, its the same as the locate database on Unix systems. Which I and most other Unix users leave on. I mean, seriously, how much is a few more megs?
Yeah, I know. I was just trying to clarify that the grandparent poster didn't likely mean the laws of physics would be turned upside-down or anything. And that he likes Warcraft, apparently.
Whew.
Anyway, as for security, a lot of these PHP systems have issues. PHP itself isn't the most secure, but it's not really at fault, I don't think. I think, to be honest, that a lot of these systems are written by guys who aren't coders, and it shows in the code (the one's I've looked at, at least). I messed with Drupal, incidentally, for some contract-work; I found it to be quite mature, stable, and reasonably organized; the code was nicely object-oriented and the whole project was quite professional. Comparitively, I think the ones by web-designers who learned to code on PHP simply show the lack of organization in terms of bugs, efficiency, and security.
As for my own site, I use a hack of a kluge of a PHP knockoff of Blosxom. It's far, far from elegant, and could really use a total rewrite (which I keep telling myself I'll make time for, but hey, if it ain't broke...). Ah, well.
Have you had a lotta luck with it? I was able to run Linux on Linux (but I may as well use UML then), and it was quite fast, but I didn't have Win98 handy to try, which as I heard it was the only Windows to work on Qemu (and even then, quite unreliably). Know anything about that?
VMWare is the most popular commercial one (for Linux and Windows; VirtualPC would be the one to try on Apple--unless you just want to emulate PPC on PPC, i.e. run OSX on PPC Linux, in which case Mac On L inux is for you). Bochs is the leading open source contendor, in that it emulates a complete x86 machine, and works on any architecture (SPARC, Alpha, PowerPC, etc). However, because of that, it's quite slow, and is far more useful for things like reverse engineering or OS testing than actual desktop use (i.e., if you wan't to see registers in use, it'd be great; if you want to watch a movie or use Photoshop, don't bother). And of course, there's always WINE, which runs a number of Windows programs on Linux quite well.
Ah, well. Trust OS News to be short on technical details. Or even on proper grammar.
A little background on my mistake; I received a notice about this at the end of last week, commented that my university had been targeted as well, but it was implied (details were very scarce, and the link I had been given to the stanford site was wrong) that the targets were simply general purpose computing resources.
In a number of cases, the attacker gained access by sniffing a password off the wire (granted, insecure services should be eliminated, but try telling that to users who rely on them) or simply guessing an insecure password. After gaining local access, in some cases access was escalated through a known vulnerability, but in other times simply through password cracking.
So what excuses are there for the admins? First, securing against authenticated users is reasonably hard. It certainly can be done, but in many cases it's understandable that admins might not fully achieve such security. Coupled with the inherent trust most grant their users, this is to be expected. Second, users choose bad passwords. It happens. One can enforce secure passwords--many places do--but just like eliminating insecure services, it ticks off the users, and as a result some servers won't implement this. And as for patching, again, a department may not have budget for backup machines and failover; a certain patch--say, one against a local vulnerability only exploitable by authenticated users--may not get applied for a little time.
Point being that even while reasonable precautions are usually taken, there are often still vulnerabilities for the determined attacker to exploit. Typically, the issue is much less laziness than a desire to keep accessible to the users, which is the entire point, really, of the machines to begin with. As for your comment about rootkits, we only know about the machines that we know about, and even those were often quite well covered.
Try not to blame the victim. You or I could be next.
Academic computing is the epitome of *available* computing, in the sense that availability is the highest priority. Financial institutions may prioritise (or at least, should prioritise) security and a good administration over availability, but by its nature, academic computing involves disparate infrastructures, various levels of admins with various goals, and so forth. All students, faculty, and staff need access; frequently, granting loose, unsecure access is simply more efficient for the time being than making things secure. Such is life.
Considering that the Google Toolbar only works on IE, Google doesn't actually come across as THAT Linux-friendly (as a client; obviously, they use Linux industrially-speaking).
Sorry. Something about
Just makes me think of mad scientists.Oh how I wanted to. Alas, fucking Broadcomm won't release Linux drivers, so I'd have to get a USB WiFi adapter.
This has been around for well over a year. I work in such an organization, and we get MS licenses for $5 a pop. Granted, not as free as Linux, but I don't think our recipients would appreciate Linux PCs quite as much. Hey, I use Linux on my desktop full time. I don't even own a Windows machine. But those are the realities, I'm afraid.
I had, prior to buying my 12" Powerbook, used OS9, and found it to be quite unpleasant in that the interface is perfectly friendly, if a bit archaic-feeling, but that there seemed to be a complete and utter lack of any real features in the area of networking, no multi-user capabilities, and generally poor reliability and usability. I had also used OSX a small amount, and it seemed to be, if a bit slow (and still does feel that way, even on my Powerbook), a perfectly tolerable Unix-ish OS.
Nevertheless, I suspect I'll have a Linux desktop around for a long time (and in some respects truly do prefer it over OSX). Firstly, I found switching to a closed-source OS to be a bit of a bother. Certain minor things I'd like to change I can't, at least, without significant work. While the initial installation is easy, and getting a working desktop for basic stuff like email, web browsing, etc, I can't change certain things how I'd like. The sec ond issue I have is more anticipatory--sooner or later, I just know Apple is going to bite me in the ass with upgrades. OS 10.1 users are apparently expected to upgrade to Panther. When 10.4 comes out, am I going to be expected to plop down another $120 just like that? And finally, in terms of usability, yes, OSX has many nice features. And it's pretty. But it runs X11 apps clunkily at best, Fink and OpenDarwin ports and all are great, but there's not NEARLY the range of Free software available (e.g. the lack of a non-alpha level, non-X11 Free, or even free, word processor). Closed Broadcomm drivers mean I can't put my Airport Extreme card into passive mode. And of course, I'm simply less familiar with OSX, as well. If I decide to implement GRSec and PAX protection on my Gentoo box, I can do it. Being a bit of a security hobbyist, how do I know my Apple is as secure?
OSX is great, to be sure, for a desktop. It patches the weakness of desktop Linux and is, in many ways, a paragon for that endeavor. And XCode and all are certainly good enough that I have no real issues doing development ON OSX, though most of what I write is written FOR Linux/x86. But I'd never use it for a server, rarely for anything truly serious, certianly not on a desktop where the price of Apple hardware is prohibitive, and not simply because Aqua apps look ``a hell of a lot slicker'' :P
Overall, however, I am happy.