Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Apple Businesses

Usenix: Darwin Welcomed by BSD Community 92

Posted by CmdrTaco from the stuff-to-read dept.
An anonymous submission tells us that MacCentral has a story about the start of the annual Usenix conference, where Apple's Darwin Open Source OS reportedly received a warm welcome as a new member of the BSD operating system family.
This discussion has been archived. No new comments can be posted.

Usenix: Darwin Welcomed by BSD Community More

Usenix: Darwin Welcomed by BSD Community

Comments Filter:

  • It would be nice... but not a sound business move (Score:3)

    by Anonymous Coward on Tuesday June 08, 1999 @05:21AM (#1862209)
    It would be nice to have an alternative like MacOS X for the Wintel machines. However, it is not likely for reasons like the following:

    1) Apple makes money from making boxes, not software: The bulk of Apple's earnings come from the boxes it sells. Making their OS available to Wintel machines would cut into those revenues.

    2) MacOS X on Wintel makes Apple compete with MS directly: Ok, they are already competitors in a way, but this will bring that competition to a different level. The Mac relies on MS Office suit to make their platform more attractive by making documents from Mac Office compatible with documents from Windose Office. They cannot afford to lose MS as a developer of their platform. At least now.

    3) Maintaining two flavors of the OS is costly: Their R&D budget will pretty much double in order to maintain and develop an Intel version of their OS.

    This been said, there are "rumors" that Apple is secretly continuing development of an Intel version of MacOS X Server. This is very likely since until last year there were builds of the OS made for Intel and PPC as well.

    The Intel version was quietly discontinued after MS announced their support of the Mac at the famous Macworld keynote that year. I would not be surprised if MS had something to do with halting OS X Server development on Intel since it competes directly with NT Server.

    Comments to this posting are welcome...
  • The kernel is actually Mach 3.0, a 'micro-kernel' architecture, which is also open source. BSD comes in the form of a compatibility layer on top of Mach.

    The fuss is more about all of the other parts of the OS they are opening up, like HFS+ (Mac's second generation file system) and others.
  • Alpha is the fastest slab of silicon money can buy.

    no it isn't, not by a long shot. Why do you think an S/390 with half a dozen CPUs is so much faster than a top-end AlphaServer 8400? and the more recent PA-RISC jobs are very close in performance - I'd rate a K-class against the equivalent Compaq kit anyday.

  • However, Mach is derived from 4.2BSD.
  • > It's cheap, but that's -all- it has going for it.

    Um, no. It has a code base that OS X could be built from: FreeBSD. It also has Windows: people are more likely to try OS X if they can dual boot into Windows when they need to--good way of converting the WinNT folks. And there is a really, really large x86 market, which means more customers.

    x86 does suck, and many people would like to see it dead, but it isn't about to happen, and it would be stupid to throw away that market because you prefer Alpha's.

    (BTW, following sql*kitten, I'm going to fall for the flamebait and point out that while the Alpha may be the 'fastest' (for now), it isn't the most powerful. Right up there though, and probably the most common chip in the 'fastest' catagory.)
  • Mach is actually a microkernel, which acts as an
    abstraction layer between the hardware and the OS
    kernel. In other words, for Darwin, your programs
    run on top of the BSD kernel, which runs on top
    of the Mach microkernel, which runs on the actual
    hardware.
  • I heard that too, supposedly, NT is based on Mach, but after they tried to make it more "windows-like" it ceased to be Mach anymore... (like Like A. Skywalker--after turning to the darkside, he ceased to be A. Skywalker and became D. Vader.)

    Also, Microsoft is responsible for Xenix, one of the first ports of Unix (system V) to the x86 platform.
  • If MacOS X is going to be basically a unix variant with the MacOS user interface (and none of the crappy underlying old MacOS code) Will there be a port to x86?
  • Well, Microsoft bought their networking code from BSDi, instead of taking it directly from FreeBSD/NetBSD (was OpenBSD around then?). Lord knows why.
  • What I heard is that Windows NT is actually more VMSish than anything else. The guy responsible for most of the design of NT was ex-DEC, and one of the original VMS programmers.

    -lee...oh, and Microsoft didn't write Xenix, they paid SCO to do it. PS: VMS + 1 = WNT.
  • Well, probably not for a long time.
    Keep in mind that apple still gets the majority
    of it's money from the hardware business. This OS would have to be very profitable for them to give up the hardware side of things.

    -T
  • "How much has Apple donated to *BSD projects?"

    However much it cost them to develop Darwin. I'd rather they keep donating man-hours to open source projects rather than money.

    Darth Binks

    How wude!
  • Have you ever used OS X? The client is nothing special GUI-wise. It actually looks a bit like Win3.1. The server, although decently stable due to a BSD kernel, is a hodgepodge of NeXT apps, and Mac apps on the same system, completely devoid of any graphical consistency, long the one strong point of the Mac.

    OS X is a good example of how to hack stability into a system, and how to allow the world to once again enjoy the bliss of boinkout (I remember boinkout on my old NeXT, needless to say it rules), but a poor example of how to make a GUI. It ranks far behind Windows, KDE, GNOME, the old Mac, the old NeXT, ect. in that regard.
  • Apple already did this on the 68K archictecture -- A/UX. It was a BSD-ish OS, with lots of GNU tools available too. It had a pre-MacOS 7 style GUI. It bootstrapped out of MacOS and had preemption, memory protection, a decent filesystem -- the works. Way outperformed the MacOS servers well into the PowerPC era. Technically, way ahead of NT in that era, IMO. They should have ported A/UX to the PPC and left MacOS behind.

    I'm reserving judgement on OS X; Apple has a long and ignoble history of developing innovative products and abandoning them due to marketing ineptitude, technical glitches obstinately ignored, and just plain corporate inertia. Frankly, I don't see why they're getting the buzz they are on this project. The basic OS doesn't seem to do anything special (Mach microkernel with BSD 4.4, Java).

  • Re:Will this stuff eventually be ported to x86 ?? (Score:3)

    by Anonymous Coward on Tuesday June 08, 1999 @06:18AM (#1862229)
    Cocoa was supposed to be cross-platform. Way back when we were supposed to have Yellow Box for Mac OS 9 about now, and Apple would probably be preparing Rhapsody 1.5 and announcing CyberDog 5, Apple promised OpenStep developers that, if we wrote for Yellow Box (now Cocoa), then we would be able to deploy (and I quote) "on Rhapsody for Macintosh, Intel, and Yellow Box for Windows."

    Apple has been, shall we say, dismal in carrying out that promise. They proceded to scrap Rhapsody for Intel, then Yellow Box for Mac OS 8.5 in favor of Carbon, then Yellow Box for Windows. So, now, Yellow Box developers have nothing to write for except Mac OS X Server. I wouldn't exactly call that cross-platform.

    How has Apple violated its promises? Let me count that ways: OpenDoc, QuickDraw GX, Yellow Box for Windows, Rhapsody, Display PostScript, QuickDraw 3D, eWorld, six-slot G3s, Pippin, Copland, Gershwin, ...

    You'll pardon me if I tell you that my team is now working on Linux.
  • >Have you ever used OS X?

    No he hasn't and neither have you.

    You have used either Mac OS X Server or a beta veriosn of OS X. In either case it's unfair to judge the eventual consumer version of the OS from it's 1.0, first to market Server version, or from it's current beta.
    (I think that's one BIG reason Apple hasn't trumpted OS X Server the way could/should. They know it's not pretty but it works - the year+ bewteen the rollout of X Server and the consumer version will be ALOT of beating the Unix bits with a shovel and painting them over with nice high gloss paint. I beat there won't even be a CLI access for the consumer OS when it ships - although some one will add one in about a day... 8)

    I'd agree that some of your points are valid but as a server environment goes it's WAY ahead of about everything else in terms of GUI friendliness. Some people will grumble that Server OSs _shouldn't_ be friendly the same way I grumble that, "I used 8.5 inch floppies that lost data all the time. And I was _grateful_, not like you kids today with your 2gig removables...." when I'm in that mood.

    Trust me Apple's painting and spackeling as fast as they can towards the roll-out. It may not be perfect when it hits the streets but it'll be alot better than OS X Server from a GUI point of view.

    =tkk

  • BeOS may not use unix code, but it is mostly posix compliant....
    I wouldn't get too exited about POSIX compliance. While it's certainly a good thing to have, keep in mind that Windows NT has been fully POSIX compliant for ages now.

    cjs

  • How much has Apple donated to *BSD projects? It hasn't given anything to OpenBSD. I suspect similar tallies for FreeBSD and NetBSD.

    Hooray for free advertising.
  • Forking the code has worked rather well with the BSD kernel though hasnt it? FreeBSD is rather portable and in several ways more powerful than the linux kernel. NetBSD is one of the most portable OSes I have ever seen, it will run on nearly everything. OpenBSD is damn secure. If the people responcible for for the various BSDes werent able to fork code, everyone would have to use the original BSD kernel. Perl isn'y going to be screwed up by Microsoft, they just are putting money into it because it beats the hell out of ASP. More people use perl because it will run on anything with an interpeter or CGI, ASP only works with NT and properly patched Apache.

  • It's the same as MKLinux that use the Mach Microkernel to abstract the hardware.

    Would it be possible to reboot from MKLinux to darwin and vice versa by coming back to tehe Mach level and not to totally reboot like we do today??

    would it be possible to do this to do speedy soft reboot between the two or between two Linux/BSD kernel???

  • Wasn't it SCO that paid royalties to MS until recentely because they were using some Xenix source code???

    VMS+1= WNT
    I did know this one, it reminds me of HAL+1=IBM (HAL of 2001, a space odissey of course)
  • Merced is RISC. The whole Pentium line is a cross between CISC and RISC, thats why it's become so popular. The MMX and SSE instruction sets, along with the others are all RISC sets used in a CISC environment. The Merced is to RISC that the Pentium is to CISC. It's a full 64 bits which means it will be comparable to Alpha, MIPS, UltraSPARC, ect..
  • Not so fast! Don't you think we should keep the paper clip help wizard that we've come to love in so many MS products?
  • Don't forget Newton!

    I'm _still_ waiting for a PDA that's half as useful as my trusty 2100, and which doesn't have catastrophic bugs in the OS..

    I must admit though those Casio E100s are looking pretty tempting..
  • btw: check out this [orb.com] site.. You can get refurbed NeXT equipment, though I don't recall if it comes with a full development kit..

    I recall the folks at NYC's MindVox used NeXT..
  • MacOSX Server decently stable?

    Two words: CGI scripts.

    I have hardcore Mac'ers here twitching apoplectically about this, particularly when I show them my CelticGreen KDE laptop.. ;)
  • I wouldn't be so sure about that... apparently they took, and returned, alot of code from the NetBSD group and FreeBSD group.

    If you read that Maccentral article I think it quoted one of the NetBSD guys actually... saying exactly the opposite to what you said. Besides, do you REALLY think that the BSD community would "embrace" Darwin if it was jsut stealing all their code.

    Also, if you look at the PDF documentation that comes with MacOS X Server you will notice at the end all the credits. They include, Net, Open, FreeBSD as well as Apache, and various other OS groups.

    I don't think they took so much from the OpenBSD group, but they did take some.
  • >so maybe it'll be good for the existing BSDs too.

    If you look at older coverage, you'll find that apple sent gaggles of bug fixes to the netbsd base they borrowed from.
  • So if the kernel code is open anyway (BSD), what's all the fuss about opening it?
  • Search the FreeBSD site for "Wilfredo Sanchez". He is the BSD (and now Darwin) leader at Apple. You'll find his (i.e. Apple's) contributions.

  • Can you imagine what would happen if someone put a fully implemented GUI like Apple on a UNIX box

    Such an OS already exists -- it's called 'Mac OS X'. It is Darwin + Mac-ish/NeXT-ish GUI (or rather, Darwin is Mac OS X without the GUI components). I expected great things from it, but so far, I am not very impressed. My NetInfo database got corrupted a few days ago, and recovering it is a pain. vi does not work in single-user mode, and single-user mode terminal does not do with VT100 or XTerm emulation, so I cannot use Emacs in full-screen text mode. I am editing files with echo, cat, and sed!!!

    (curses OS/X dirtily and stalks away)

    --

  • Well, sort of...
    The real news with OS X Client is "Cocoa". Check out
    Macintouch [macintouch.com] for notes from May's WWDC. Its oversimplfying somewhat, but essentially, its a cross-platform API for x86, IA-64, and PowerPC.
  • You seem to be misunderstanding RISC. There is not necessarily anything inherently superior about it. If it were all machines would only have a few instructions. There is a balance to be reached between how much work the CPU, compiler, and programmer does. RISC is about making the compiler (assembler) do more work. In fact MMX and Altivec are a tendency back to CISC type architectures. More complex instructions. Otherwise they would just give you multiply and add instead of dotproduct.

    Remember RISC executes more instructions to do any operation. Since there is a smaller granularity though, effective pipelining and out-of-order execution allows greater throughput.

    Hopefully that made some sense.
  • Long long ago in a galaxy far far away Steve Jobs actually did this.

    It was a consistent, well designed, object oriented UI on top of a Mach microkernel with a BSD subsystem. It also came with one of the best development environments ever, and was a testament to superior design. It had cd quality sound, publishing quality graphics, built in networking, even pre-emptive multitasking. All of this in the days of MS windows 3.0. NeXT's hardware was even easy to use. One screw undoes the case, almost no internal wires, the case is cast magnesium and acts as it's own heat sink. They just dont make 'em like this anymore.

    It's interesting to note that MS put them out of business by the way of compaq and dell selling ultra-cheap crapfests, and vendors being lured away by economies of scale.

    It's unfortunate that Mac community has forced Apple to water down the superior technology they bought from NeXT with Mac compatibility layers instead of an incremental update of the great OS they had. NeXT really only required incremental upgrades to bring it into 2000, but Apple has spent most of it's time trying to keep Mac developers happy rather than updating the OS they bought from NeXT.

    -Rich
  • Uh...
    The Kernel is Mach, not BSD.
    It was Apple's chief tech officer's PhD thesis.

  • Re:Advertising (Score:4)

    by 1millionmhz ( 34257 ) on Tuesday June 08, 1999 @04:51AM (#1862260) Homepage
    Wow. And here I was thinking all along that the Open Source community was more interested in what kind of code you give back to the community, not how much money you donate. If we're talking about code, man-hours, etc. I think Apple's given us quite a lot, with even more to come. If we're talking money, then let's give Intel and IBM the applause they deserve for sinking all that dough into Red Hat.
  • Can you imagine what would happen if someone put a fully implemented GUI like Apple on a UNIX box (not talking about kde's mac look either). Make a *nux box as easy to use as the Macs are supposed to be. Not only would it have the stability of UNIX but it woud have the ease of use of Mac. If there was software to go along with that, software that would be able to read M$ formats correctly, it would have a chance at putting M$ out of business.
  • It's also good for M$ because that stock they bought is now worth 3-4 times as much.

    Maybe that'll almost make up for the undisclosed sum M$ paid Apple to settle the patent disputes Apple was privately threatening to sue them over.

  • Re:Advertising (Score:3)

    by nikc ( 11398 ) on Tuesday June 08, 1999 @04:31AM (#1862263)

    Vulcan wrote:

    How much has Apple donated to *BSD projects? It hasn't given anything to OpenBSD. I suspect similar tallies for FreeBSD and NetBSD.

    The story doesn't mention it. However, about half the userland code is from NetBSD, and Apple have supplied improvements and changes which have been integrated back into NetBSD. If OpenBSD track the NetBSD userland then those changes are probably in OpenBSD as well.

    Apple are planning (or already have) used code from the FreeBSD kernel in the mach kernel. I assume this is to get ideas and possible implementations, rather than simple cut-n-paste. As Apple's engineers refine those ideas and implementations (where refinement is possible, obviously, FreeBSD already has a very good kernel) those refinements will be fed back to FreeBSD (from which the other BSDs can grab them). Again, this probably won't be a simple cut-n-paste job, but more a "We were able to get x% speed up here by doing x, y, and z. Here's diffs, so you can reimplement it in your code".

    N

  • With Darwin this is more than possible.
    The developer releases of MacOS X Server (Rhapsody) all had x86 versions.

    The code is there. People just need to (and are) clean it up and make it work. OpenStep (on which much of MacOS X is based) was renowned for it's portability. It ran on x86, Sparc and Alpha I think. There is nothing stopping people from porting Darwin to MacOS X.

    The interesting part is whether Apple will heed the call and sell their GUI and APIs to go on top of Darwin wherever it is ported.

  • Nice analysis, but don't forget that NextStep was near commercial death, with less than 10 ISVs writing Next software (or so I've heard). The 'carbon' API is a necessary evil for those with huge MacOS code bases, such as Adobe, Microsoft, and others.

    In the long run, this just forstalls the inevitable, which is Yellow Box/OpenStep. Don't worry - you NeXT people have won.
    --
  • I think it's already been decided, except for the timing. What do you think would happen to Apple's stock price (hinges on profitability) if they ported their crown jewels to x86? They'd get HAMMERED, and then they'd be suck supporting the tired poor hardware that's so common in x86 land.

    There IS a version of Mac OS X for Intel - I've seen it run and I wasted a whole week trying unsuccessfully to get it to run on spare work PC's (prior job).

    I think once the hardware sales took off they didn't need OS X Server/x86... this was "plan B".

    PowerPC has it all over x86 for elegance and engineering but not price. They'd be completely stupid if they did not have a migration path laid out, but first they have to win developer trust and prove they can move us onto a UNIX base and do it much more quickly than moving us off Motorola 68000's took. Once their customer base is secured they can think about opening the hardware up to competition again.

    The difference between offering OS X on x86 and BeOS's move to Intel, is Apple *actually* has a chance to pull it off, especially in light of Microsoft's troubles (post-Y2k, DOJ). Apple gets to tap into the "alternative" development, since they're really just another UNIX, meanwhile Microsoft further isolates itself on the crappy NT kernel (yeah, can't wait for "Consumer NT" ;-)
  • I mean, obviously all the *bsd variants... and some commercial unices are bsd based. but i've heard rumors that some m$ code is bsd based too. do they belong to the family?

    in other news, i'm jealous. rich got to go to usenix, and i had to stay in indiana. and it's really humid here these days... ugh.

  • From what I've read/heard about OS X Client that's exactly what Apple's in the process of doing. It's based on a Mach Kernel and it has a MacOS GUI with all the smiley prettiness and ease of use that entails.
    We'll have to wait for betas to be released before we can see how good a job they do, but if OS X Server is any indication of the direction they're headed I think it looks very promising.
  • >MacOSX Server decently stable?
    >Two words: CGI scripts.

    Yes, the binary of Apache that was bundled with OS X Server 1.0. How quickly Apple responds to this and fixes it will be a good test of Apple commitment to Servers and Serving in general.

    That being said I'll put it up against both Linux 1.0 and NT 1.0 for a stability test. ;)

    >I have hardcore Mac'ers here twitching apoplectically about this, particularly when I show them my CelticGreen KDE laptop.. ;)

    Is that a thinly disguised sexual reference/threat? ;)

    "Oh my god! Mr. Burns is coming onto me!"

    =tkk
  • I've been using BeOS since R4 was released, and I haven't wandered across it...perhaps I'm not looking in the right places ;-)
  • Ah well, you probably belong to the usual suspects who also post to comp.sys.next.advocacy incessantly whining about Apple and the demise of the YellowBox/Objective C. Get it through your thick skull that we don't care about your endless tirades!

    It's interesting how Linux has devolved into a rallying place for bitter people with an axe to grind. First it appealed to the Anti-Microsoft clique, now they get the Anti-Apple people too. What a lovely crowd.
  • Wasn't it SCO that paid royalties to MS until recentely because they were using some Xenix source code???

    I don't know if they paid royalties, but SCO UNIX (and some other SVR3/SVR4 UNIXes) did include Xenix-compatibility code; there was a suit wherein, as I remember, SCO sued to be allowed to yank that code out of one of their UNIXes.

    As far as I know, Xenix was originally a Microsoft product; I think they ended up handing it off to SCO.

  • The GPL is the license that keeps people from forking the code. The BSD license allowed the code to fork like crazy (BSD, FreeBSD, NetBSD, OpenBSD) before.

    Given that FreeBSD, NetBSD, and OpenBSD are all free software, I'm not sure how "the BSD license" allowed them to spin off from BSD and "the GPL" would have prevented that.

    The GPL doesn't say you can't take the Linux kernel or GCC or... and make your own separate version thereof; you just have to make source available to everyone to whom you make the binary available, "cause the modified files to carry prominent notices stating that you changed the files and the date of any change", allow anybody who got your separate version to freely give it away, etc..

  • BeOS may not use unix code, but it is mostly posix compliant and Be is shooting for full compliance in r5. While Mac's may have MacOS X as their next generation consumer OS (the client), the PC platform has BeOS which has all the basic features of good unices like smp, pre-emptive multitasking, protected memory, gnu tools (almost if not all of them). Right now on a technological scale Windows can only compete with BeOS in 3d acceleration because BeOS doesn't support it yet. However VERY soon when r4.5 comes out there will be 3d acceleration for those of us who took the plunge to install the flying batmobile (BeOS).
  • HFS+ is Apple's third-generation FS. The Macintosh File System was the first Mac FS; HFS was Apple's second.

  • AU/X was System V, and there was *no* GNU stuff available, as far as I could tell, apparently due to a FSF boycott.

    AU/X was pretty cool for it's day though. I believe the OSX "Blue Box" works essentially the same way that Mac programs (including the Finder) ran on AU/X.
    --
  • Darwin is a *very* portable system so eventually we should see it on Sparc and Alpha too. [...] Then Apple can sell the MacOS X GUI on top... and there will be a REAL MS killer ;).

    Apple's revenues come almost exclusively from hardware, not software. That means Apple's software R&D comes from hardware. That means every times they can sell the Mac OS (X or otherwise) for less than the minimum $500 premium they see from a hardware sale, they're pissing away money.

    Yes, as long as one doesn't post any math one can make the claim that if they sold gazillions of seats, they could make up whatever revenues they would have lost through hardware sales. Well, even discounting the far greater costs of supporting hardware the company didn't build, it's such a wildly dangerous gamble that Apple would be looking at a shareholder lawsuit removing its executive officers before anybody could even say, "I've hacked in a replacement for the Start Bar."

    It will simply never happen. The economics are prohibitive.

    bumppo
  • Mainly I'd like to question point 3 --
    maintaining multiple flavors of the OS
    is only costly at certain levels, i.e.
    parts of the kernel, compiler, and drivers. A portion of
    the kernel and almost the entire rest of the OS,
    at least on Unixlike systems and several other
    OS's (like NT, BeOS, etc) tends to be portable.

    So while the R&D budget may be somewhat higher, it
    should be a far cry from being double that of
    being single-platform. A good guess might be
    that it'd be an additional 10% of work..
  • It ran on x86, NeXT68k, SPARC, and some PA-RISC (HP) boxes.
  • Because it is Apple, it's a commercial OS. They are also opening things like AppleTalk, NetInfo, their DriverKit (IOKit), filesystem (HFS,HFS+) and a few other things that have never seen the light of day.

    This was really done in response to the outcry that there wasn't an x86 port of MacOS X (All the developer releases, Rhapsody, had a fully functional x86 version). So by releasing Darwin they released all the x86 code that was already there so that we can port it over. Darwin is a *very* portable system so eventually we should see it on Sparc and Alpha too.

    Then Apple can sell the MacOS X GUI on top... and there will be a REAL MS killer ;).
  • Apple wouldn't have done that. According to the GPL (as I understand it), anything in which Apple used GPLed code (the Linux kernel) would have to be GPLed as well. While Darwin is open source, it's likely that parts of Darwin are integrated into the non-free portions of the OS that Apple would not want to GPL.

    -awc
  • merced is such a broken idea. i think when merced is released, people will find that the rare times that a program can be parallelized that much will not be enough to take advantage of merced's level of parallelism(sp). also, it makes compilers that much more complicated.
  • While it might be a little premature to say the open source magic is working in this case, I doubt it can hurt.

    Leave Freecell too! (do they even bundle freecell in win9x?) It was in my NT 3.1 beta all those years ago and I loved it. It was the only thing I liked about NT -- until I turned the CD's into coasters.
  • Of course there are several good Freecell implementations just waiting for you. Search for freecell on freshmeat. I believe the Gnome distribution also has a pretty good freecell implementation. The best part is that Freecell is now free software.
  • The lawsuit against Microsoft points in this general direction. People from IBM have stated that Microsoft forced them to keep OS/2 development/marketing low profile or Microsoft would charge them very high prices for Windows-licenses. Which ofcourse would make it very difficult for IBM to sell their PCs in the mainstream business market.
    This might have been the same with Apple: "Don't compete with NT or no Office for you"
    I think OS X for Intel might still happen, maybe as Darwin for Intel when Apple decides to open up more of its source. With a lot of open BSD's out there for Intel someone will get it into his/her head to put some Intel hw support in there.

    Message on our company Intranet:
    "You have a sticker in your private area"
  • If it uses the NeXT core, then it must be Mach 2.5. The kernel is actually Mach 2.5 with a "monolithic" BSD kernel on top of it. Both Mach and BSD are in the kernel space. This is the architecture of Digital Unix too (now Compaq Tru64 (ugh) Unix).
  • instead linux has redhat, caldera, debian, suse, turbolinux, linuxppc, mandrake, slackware, stampede... Then you have not one GUI with the intention of bringing a consistent look and feel to linux but 2 or 3...with the intention of bringing a consistent look and feel to linux
  • > hell's team will win the stanley cup first

    Well, you got to remember we have Satan playing for the Sabres, and they won tonight.

    Go Sabres!
  • Maybe it is VMS...

    Nope, i remember reading in an offical MS document, MS licenced Unix from ATT, and ported it to x86, then they licensed (and eventually sold, i believe) Xenix to SCO.
  • x86 sux...I'd rather not perpetuate the intellectual monopoly x86 has. It's cheap, but that's -all- it has going for it.

    ARM is cheap, fast, and easy on power consumption. Alpha is the fastest slab of silicon money can buy. You can buy generic system boards for -both- architectures. We already have a low-end workstation/PC platform for MacOS X: the Macintosh.

    Darth Binks

    Meesa gonna bring you to da darkside, yah!

Slashdot Top Deals

No man is an island if he's on at least one mailing list.

Close