Slashdot Mirror
BSD to Leapfrog Linux?
photozz writes "New from ZDNet about the coming of OSX and how some people see this as the rise of BSD, perhaps passing up Linux in numbers of users. " I'm still excited about OSX. I still am considering buying a mac to play with OSX... I mean, I can always install LinuxPPC if OSX sucks goat.
Normally I'm an advocate of "It's not 'FreeBSD vs Linux', it's 'FreeBSD and Linux' -- different tools with different strengths, use the right tool for the job", but frankly sometimes when I'm trying to use FreeBSD as the right tool for the right job I find myself wishing I could do it with Linux instead.
Linux is the de facto superior workstation OS (easier to work with, more features, more apps (not all of which run under FreeBSD's linux support)), but FreeBSD is the de facto superior server OS (stability, stability, stability).
I like FreeBSD for its stability under load, its
I've been poking around at the various Linux distributions for several years, and none of them that I've tried managed to shore up Linux's shortcomings, either, so I've continued to prefer FreeBSD for heavily-loaded servers, but I still think that The Real Solution is to develop a Linux distribution which makes it as good as / better than FreeBSD for server work. I've been wanting an excuse to work on the kernel anyway, and I might be able to do it on company time if I can pitch it to elbossman right. Look for Annie-Linux (I'm a big Eurythmics fan) in, oh, three or five years.
-- Guges --
I want to do this because I read some posts (like the parent; nothing personal, you just happened to remind me of this question of mine) that complain about some view of the traditional Slashdot user and I wonder what posts they are talking about. Sure, you see some "linux r00ls" posts, but if you exclude the blatant trolls, I don't remember reading that many. To be fair, I realize that I may have a mental filter when I scan the posts that makes the stupid ones (like what the parent post complains about) recede into the background and the posts that complain about them stick out.
Hopefully that makes sense. I can't tell if our perceptions of Slashdot are colored by preconceived notions of the types of posts we expect to find.
If someone gets bored enough to do this for a psychology/sociology assignment, let me know. :)
Perhaps they will grow up. :-)
The interesting thing is that many of these people tend to judge by the looks. This year I had several people who perceived my FreeBSD box running under Windowmaker as a Linux box. Obviously they associated something totally alien with a BSD box and did not really grasp the idea that under UNIX the GUI is portable. For them KDE or GNOME is synonymous for Linux (yes, arghll!)
So both "worlds" have many in common, but also important differences (BSD license vs GPL, rather centralistic development vs loose development..)
My belief is, that the BSD will benefit from the BSD license. It is more free and allows for cooperation with industry. The GPL might have its advantages, but is not really the optimum for world of free and commercial/closed software.
Indeed it can run all of the traditionial Mac OS applications as well as any ported specifically to OS X.
From the user's point of view aside from a somewhat different interface Mac OS X will run all of the same applications it always has. Some things like Control Panels & Extensions may not operate (it *is* a different OS after all) but applications have no problems.
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
I don't buy that argument.. the same could be said for Sun, but they make an X86 Solaris. Why? At least partly because they know if people use it on x86 and get experience with it, they might then buy Sun computers knowing how much better it will/could be on it. There are other reasons too (so that admins who work on Sparcs can use Solaris at home too without the high cost of sparc machines), but that's one of the primary ones IMO.
Well, that works for Sun. Sun's advantage is it's highly scalable server software. This isn't the same as the desktop market. The kind of person who would buy a high end Sun server wouldn't be able to find the same performance in any current x86-based server. It's not just more, faster chips. There are a number of other reasons too.
However, Apple is in the desktop market. The speed advantage of Apple hardware is a slight thing at best nowdays, while the price is still at a premium. I don't regret my purchase of a PowerMac G4 at the time I made it. It was far better than any x86 machine on the market at the time. Times have changed, but the prices have not. The price ratio is no longer worth paying on the merit of higher performance alone, and recent trends over the past few years should show anyone knowledgeable in computers that the public cares far more about low price than high performance. Witness the explosion of sub-$1000, then sub-$800, then sub-$500 markets.
What are the two reasons to buy from Apple, then? If you're new to computers, fashion is probably a factor. If you're not, then the Mac OS is the more important reason. You make the Mac OS available on cheaper albiet slower and less reliable hardware and most people won't care. They wouldn't pay the premium for Apple hardware, unless they liked the looks. Remember, I'm talking about normal, uninformed people -- not computer geeks like us.
That guts Apple's hardware sales. We saw the exact same thing when Apple allowed clone makers. Questionably superior or inferior hardware, offered for much less than Apple's offerings. Basically, it ate Apple alive. So, what does Apple have left? While dropping hardware would cut their costs, the vast majority of their profits come from their hardware. There are only a few OS-only companies. They usually either:
1) Try to sell their OS at a premium, with limited success (NeXT)
2) Lead some particular profitable niche market (QNX)
3) Supplant their OS research with other products (Apple, Microsoft)
4) Offer support for arcane, user-unfriendly OSes (*cough* Linux distros *cough*)
5) Flounder until they can do one or more of 1-4 or die (Be)
Apple has next to no other real compelling software to sell, like MS, and they no longer dominate any markets that Windows or other can't really compete in. The whole goal of the Mac OS is to make living off of support nigh-impossible, so their choices are to jack up prices -- which will chafe most cheap consumers (who are probably already paying for Windows anyway on x86 hardware) -- or they can take the nosedive to death. It really is that limited.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
You know, wanting to keep parts of their system closed has nothing to do with the choice of BSD over Linux. Everything that Linux would've done (the BSD layer) is already open source. Everything else in the OS is just a user-space app. You can drop in a new copy of the Darwin kernel and not disturb a single thing in the system. The GPL in no way bars closed apps from being run on an open system.
The real reason Apple went with BSD is because Mac OS X is basically an updated version of NeXT's OPENSTEP OS that can run Carbon apps, has a new graphics sub-system, Java, and includes a few new apps like the emulation environment that runs Classic Mac OS programs. NeXTSTEP and OPENSTEP were based on BSD/Mach. All Apple did was update the Mach kernel and update the BSD 4.3 code to BSD 4.4 from some Free distributions.
Using Linux would've involved ripping out the UNIX underpinnings and replacing them for negligible gain. It was a smarter move to stay with what had been proven to work for NeXT. (Though I really, really wish that they'd adopt GNU syntax in the command-line utilities. It's so much better.)
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
By the same token, I'm pretty sure that most people would notice, and probably regret, if they were to be moved, without their consent, from one of these platforms to the other.
Carefree highway, let me slip away on you.
Linux was created by geeks, and despite the immense brainpower of said geeks, it's been difficult to get large numbers of non-geeks to use Linux as their primary home or buisiness desktop computer.
Think about it - how many friends/family do you know who use Linux as their primary desktop machines to use the Internet, file their recepies, write memos, do their taxes, etc.?
Apple is and always has been the most well-known innovator on the consumer computer hardware side of the market. When they lead, others follow. Cast aside arguments about the merits of their technology, but think about how many times their lead has been followed by other OEMs.
Now imagine the leader in consumer computer hardware using a BSD variant as their OS. Now imagine other hardware vendors thinking to themselves - hmm.. we could take a cue here, cast off the M$ shackles, and use a BSD variant on OUR machines.
Until BSD and Linux can get past the stigma of being seen as useful only to geeks, neither OS will take a significant bit out of M$ in the consumer market. Right now the only hardware vendor capable of spearheading that charge is Apple.
Read the EFF's Fair Use FAQ
Here's a better article (I think) on BSD's rising star.
--
As a matter of fact, I am a lawyer. But I play an actor on TV.
----
Quark, Adobe, Macromedia and MANY others have said that they will support OSX. Of course they will - they dont want to lose their mac user base. Mac software is lucrative - Microsoft Office for mac is one of their most profitable products. See this page for more.
LinuxPPC has the excellent implementation of Applixware and hopefully that will continue to happen, though of course there is some doubt with the recent decision by Applix to focus on server-side instead of the desktop app market. Supposedly StarOffice will be available for LinuxPPC but I don't think that has happened yet.
There was a reasonable comparison between Mac on Linux under LinuxPPC and the Classic environment under OS X which basically said that OS X does a better job intergrating the earlier OS's (just in a window) but they both are very reasonable.
It's psychosomatic. You need a lobotomy. I'll get a saw.
I honestly don't understand the hype surrounding MacOS X. So far, the fact that Apple is producing a UNIX seems to be the only lure to people like Slashdotters.
At best, we're being handed another UNIX that could possible have more compatability issues than any other. We already have four ring leaders in free UNIX and UNIX-like operating systems as it is, and they've all been around long enough to be proven perfectly functional in production environments.
I guess what I'm trying to ask is, can someone rationalize the community interest in MacOS X? What does it offer over other UNIX environments, other than yet another OS alternative? Why should I take interest in this? Apple has never offered anything to me personally or professionally. Why should this change just because they're releasing an operating system that beats with the heart of BSD? This seems like nothing more than a toy to keep people with Mac hardware laying around entertained, coupled with a boat-load of hype.
Look at it this way: I want to be interested in MacOS X, but I can't find any reason. Please help! :^)
Relating to the topic of the article, I don't see hype being quite enough to divert people's attention away from existing free UNIX operating systems (and non-free in the case of Solaris, which seems to be most popular down commercial avenues). All hope for MacOS X lies in whether or not it can perform as well or better than other UNIX alternatives when it finally hits x86. Providing adequate support for hardware, both natively and from manufacturers, will also be an important contributing factor. If MacOS X can't do this initially, I doubt people will ever take it seriously. Initial disappointment could have lasting repercussions if it does eventually become a scalabale and stable operating system. Linux and *BSD have more than proven themselves, but MacOS X hasn't been given that opportunity.
To say that 'BSD is likely to rival Linux very soon in total number of users' while MacOS X is still in development seems absurdly silly to me, more so considering that it currently operates only on a statistically unpopular (and thus far workstation-oriented) platform. I feel that Henry Kingman has grossly underestimated the popularity and portability Linux offers, both as an extremely stable serving environment, and as an evolving desktop environment. Not to mention the other (currently available) BSDs.
NickIn embedded devices, I'd bet. Does that matter? If OSX is BSD based, but Joe Average mac user barely knows about it, does that matter? What can Apple's involvement do to help proliferation of BSD? Apple isn't known for SMP, server market, scalability, etc, and on the desktop side, all the eye candy is being kept proprietary. So, other than a talking point, I can't see Apple aiding the BSD cause all that much.
Yes, streaming media services may be nice, but I am looking for something that could really cause heightened BSD deployments.
ostiguy, openbsd firewall user
MacOSX = Rhapsody 5.7+ Rhapsody = OPENSTEP for Mach (product code name change as of Apple buyout) OPENSTEP for Mach = NeXTSTEP (product name change as of Sun-NeXT co-released OpenStep spec.)
therefore (transitive property)
MacOSX = NeXTSTEP
The series, each of which is comprised of some version of Mach, BSD, Display Postscript, and Objective-C Frameworks:
NeXTSTEP 1.x
-BSD4.4-lite
-Mach 2.5
-DPS
-Objective-C + Appkit Framework
NeXTSTEP 2.x
-BSD4.4-lite
-Mach 2.5 + extensions
-DPS
-Objective-C + Appkit Framework
NeXTSTEP 3.0..3.3
-BSD4.4-lite
-Mach 2.5 + more extensions
-DPS
-Obj-C + Appkit + Foundation Kit (early kit)
OPENSTEP 4.0..4.2
-BSD4.4-lite
-Mach 2.5 + more extensions
-DPS
-Obj-C + New OpenStep frameworks + EOF
Rhapsody 5.x (Early Apple prototype)
-BSD4.4-lite
-Mach 2.5 + blah blah
-DPS
-Obj-C + OpenStep core frameworks (Codenamed Yellowbox) + extensions + EOF
MacOSX Server 1.x (Rhapsody 5.7) same as the above, but stabler.
MacOSX 1.x (Rhapsody 5.x [where x    -BSD4.4-lite
   -Mach 3 + fidly bits
   -DisplayPDF (Quartz)
   -Obj-C + enhanced OpenStep frameworks (Now called Cocoa) + EOF
BSD bits were taken from NetBSD and FreeBSD, with (I thought) some userland from OpenBSD.
EOF = Enterprise Object Framework - an Object-to-Relational Database adapter layer (very very good.)
I just can't imagine a Mac user:
- Editing and customizing System and app resources.
- Overclocking his Mac.
- Building a cross-platform network.
- Running traceroutes.
- Still remembering the file copy command from CP/M. (PIP, of all things)
- Reading and posting to Slashdot.
Oh, wait. I can imagine. That would be me. (And I know I'm one of many.)Oh well. Nevermind.
--
NetInfo connection failed for server 127.0.0.1/local
I've never seriously used it, but BSD: lacks a journaled file system
... but they've actually lost some utility in memory management that is more relevant for workstation users. So you could argue linux is now really good for big iron tasks, but the BSDs have really clean memory management that allows them to be heavily loaded and perform well.
But has soft updates, which are a substantial improvement over ext2 with respect to recovery from crashes. More seriously, what advantage does journalling have over soft (atomic) updates, and what advantages do soft updates have over journalling ?? I think reasonable arguments can be made that soft updates are faster, and at least as crash tolerant as journalling, and a heck of a lot easier to program and maintain wrt the VFS layer.
only recently migrated to the ELF format
So???
still uses Bind 4 (OpenBSD specifically)
And this one is really relevant for those 0.01% of machines begin used as name servers.
Does BSD have a multi-threaded IP stack? How does BSD perform on Mindcraft? Linux has been playing catch-up in this space for some time, and may have a big lead.
Actually, *BSD does much better with their IP stack than linux 2.2.* and 2.0.*. *BSD also does much better under heavy loads because it has a more mature memory management scheme. Linux is supposed to work on this in the next devel series now that more fine grained SMP locking is present.
But seriously, how much advantage do you reckon a multi-threaded IP stack makes on a single processor machine ??
The free BSDs are a very fine choice for a kernel and base utilities. There are some areas in which linux is better, and others in which the BSDs are better. Generally common server tasks work out better for *BSD than linux, and application availability and marketing are stronger for linux.
Linux's big recent push has largely oriented around big hardware - SMP, multiple NICs,
OK, I'm about to find out if the points I've already given out in a thread get trashed now that I'm posting in it. (Note: If they do, then I apologise to the only person who's said anything
/.:
/., and they know (at least the good ones do) where the different systems shine. If you're nodding along with this, you know what I mean. If you're going, "but I'm not like that--he's an asshole!" then you're quite possibly _exactly_ who I'm talking about.
worthwhile so far on this article.) I just can't ignore this stupidity any longer.
Here's what I hear most of the time on
"I am a unix guru. I work with Linux linux linux linux linux linux linux (solaris) linux and linux.
I tried to install BSD once, but it didn't work so it sucks. I know, because I am a unix guru."
Most of you haven't even touched Unix as a field--you've played with different distros of a _single_ variant of Unix (which very pedantically isn't even Unix at all), which is about as significant as playing with the different versions of Win95. THERE'S A WHOLE SHITLOAD MORE OUT THERE, and almost NONE of you posters have even seen it, let alone know enough to comment intelligently about it.
Anyone played with HP-UX? How about AIX (eek!)?
Tru64 is kinda different too. Then there's SCO Unix (if you can afford it), IRIX, and so on.
It comes down to this: I'm sick of evangelists who Know the One True Path to Enlightenment, and feel that sullying themselves with broader knowledge (not to mention history) would be Blasphemy. Or maybe they're just scared to find out that Linux (or BSD, or even Solaris or ANY single OS) isn't the perfect, magic, foolproof solution to everything.
Naturally, there _are_ true Unix professionals on
"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban
OS X Workstation will be nice. But it really is a "Mac" approach to BSD. OS X Server on the otherhand, is what most of us think of when we compare NetBSD, FreeBSD, etc. to OS X.
We have an OS X Server (Rhapsody) here, supporting the networked filesystems. At first glance it seems pretty cool. But the SCSI drivcers (for a very popular PCI SCSI card, probably the most popular) are really alpha quality. They don't even support disconnect, which causes a kernel panic when you remove a tape from the tape drive, in use or not. And the only tape backup solution for OS X Server was pulled off of the shelves for legal problems. Even with that removed, under loads the computer randomly crashes, even without third party software (of which there exists little). To get a compiler for Rhapsody, you have to shell out tons for a developer's kit and membership. Linux is soon replacing that OS X Server here.
For Apple to propel OS X (and thus BSD) beyond Linux, they need to devote more attention to OS X Server than they have before. I understand Apple plans to do this when OS X Workstation is released, to make OS X Server more compatable with the workstation product, and therefor more useable. But I don't hold my breath- last I heard they were delaying it. Hopefully Apple will see the potential and avoid ignoring the Server product.
As to price, OS X should run comfortably on the MSRP US$799 'Indigo' IMacs. While you may not have that cash laying around it's not a bad price for the hardware one gets (15" Sony monitor, PowerPC, fast Ethernet, Modem, etc.)
As to Intel x86 support, no one has yet to describe a viable way for Apple to sell this & not cut their own throat. 1000th repetition: Apple is a hardware company - they make their money on hardware - they couldn't survive as an OS house. OS X may well exist on x86 (& Alpha) but until there's profit in it don't look for it to come out of the labs.
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
- popularity
- availability of a rich suite of apps
- whether the platform is stable (doesn't crash)
- speed/efficiency
- cost
At this point, Linux is way ahead of the other UNIX-like siblings on popularity, and also leads in app availability.I'm an old UNIX and BSD hacker and bigot (since the '70's), and I personally think that BSD is a better, more robust, more secure, slicker software distribution than Linux, but I understand that Linux has the upper hand in the market just because of more effective hype.
It might sound sensible to say "Why can't we all just get along?" or even "Why can't these Linux kids give UNIX/BSD their props?" but that matter is no more objective than OS preference.
Actually, there is quite a difference between
"UNIXalikes" maybe at the command line they are all similar, but even then there are significant differences by default. I'll just compare Solaris and standard GNU/Linux distros. The shell of choice in solaris is ksh, which behaves differently somewhat from bash, the linux favorite. Also, linux usually ships with GNU fileutils, while solaris does not. Just use ps on both and you'll realize that there are some fundamental differences in the way they interface.
In the GUI front, solaris still sticks by CDE, while Redhat tends to like to use GNOME, and mandrake uses kde.. Also, under solaris you basically get no VCs, which is also quite a large diff.
From the standpoint of development, the systems have really different behaving environments. Just try to use dlopen() under both and you'll see. Also, try to write a multi-threaded X app in Solaris and then port it to linux. Chances are, in linux, you'll get tons of async replies before you add mutexes, semaphores, and special X calls for threading. Solaris X environment is *much* more thread safe than XFree86.
No matter how you look at it, while "UNIXalikes" may be very similar, there are many fundamental differences that distinguish them and cause preferences one way or the other. All in all, things are probably about equal, just some prefer one style over another...
XML is like violence. If it doesn't solve the problem, use more.
The differences between UNIXalikes are not worth quibbling about. From a user perspective and a programmer perspective, they are just about the same beast. It's like arguing about the radical differences between driving Ford Explorer and a Chevy Blazer (hint: there are none).
The only trouble here is that some people have religiously latched onto Linux and don't want to hear that it is 95% of something called UNIX, which has been around for nearly 30 years. They want Linux to be some kind of l33t inside secret. But in truth BSD and Linux are identical twins with different hobbies. That's not a put down of either system--or any of the other UNIXalikes such as Solaris--just a suggestion that this shouldn't be a feud.
Well, linux has already outnumbered *BSDs and OSX combined.
But that is largely irrelevant. With the introduction of autoconf, the open source components of these technologies will complement each other. Openssh was taken from openbsd to *BSD and linux. GNOME and KDE are largely linux developed, but work fine on *BSD.
The largest linux companies look at big business UNIX and Microsoft as the competition for different markets. *BSD and linux will both continue to grow at the expense of Microsoft and mainframe Unices. The market dynamics may have a few people going from linux to *BSD (and fewer going the other direction), but the changes in user base for linux and *BSD are coming not from each other but from Microsoft and mainframe Unix.
The media loves to play up battles, like KDE/GNOME, Redhat/Mandrake... but the reality is that KDE and GNOME help each other more than they hurt each other through competition. The same is true of Mandrake and Redhat. Any improvements made by open source companies in software lead to strengthening of all open source companies' software.