Slashdot is powered by your submissions, so send in your scoop


Forgot your password?

Simply GNUstep Delivers UNIX, Simply 396

Eugenia writes "A new, Linux-based operating system released recently, called Simply GNUstep and it is based on the GNUstep architecture, originally built by NeXT (OpenSTEP) and is now also used by MacOSX (Cocoa). The alpha version of the x86-based OS is available for download and boots off the 110 MB bootable CD. The cool thing about Simply GNUstep is its partial source compatibility with MacOSX programs (further compatibility is still worked on) and its clean infrastructure, as it only includes GnuSTEP graphical applications like WindowMaker, etc. You can read an introduction article of the OS at OSNews."
This discussion has been archived. No new comments can be posted.

Simply GNUstep Delivers UNIX, Simply

Comments Filter:
  • The sad part is.... (Score:4, Interesting)

    by ACK!! ( 10229 ) on Thursday January 10, 2002 @02:19PM (#2817775) Journal
    This is almost what I want.

    I am getting tired of my Gnome and KDE. I am starting to long for the days when I used WindowMaker, Postilion and FSviewer together with a cobbled up list of other xapps to get my job done simply.

    The problems are paramount. Fsviewer barely works on my updated SuSE 7.3 Postilion does not like my cutting edge versions of tcl/tk and I am not yet ready to give up the laundry list of apps I need to do business for a barebones environment. Plus, I like unified look and feel that I get with say KDE or Gnome.

    If I got a distribution with a laundry list of apps centered around those apps with a Nextish look and feel then I would be a happy man.

    The problem with Simply GNUstep is that it is what it says it is. It is Linux with GNUstep already built and configured but it has nothing else.

    If it was supplemented with other X apps with a Next feel or gtk apps with a Next theme maybe into a usable package then I think I would be in love.

    ________________________________________________ __
    • by ( 450073 ) <xanadu&inorbit,com> on Thursday January 10, 2002 @02:34PM (#2817880) Homepage Journal
      (I hate to sound like a troll here, because I don't mean to, but...)

      I am getting tired of my Gnome and KDE. I am starting to long for the days when I used WindowMaker

      So... simply don't use them. Remove the packages if you don't even what to support apps written for them. Install your prefered GUI/WindowManager and don't look back. Seriously. It's that easy. You don't need a "special" distro to do what you want. Just simply install / remove what you want / don't want.

      That's what Linux is all about: choice
      • The reason that geeks fail in meeting customer/end-user needs so often: they take someone else's preferences as an attack on their own. Really, why did you bother to post this? You think that the parent poster was calling for the annihilation of KDE/Gnome? He was expressing preferences and the reasons for them. A distro/environment that provides what he needs out-of-the-box means a lot fewer admin hassles and the freedom to just get working: it's why I used Debian instead of compiling and configuring every damned little thing by myself.

        And besides, the customizability of the KDE/Gnome environments is somewhat restricted by very complex interdependencies.

      • That was not was he meant I assume.

        What happened is more that the applications he wants, like Postilion are not much being used and developed, and are suffering from bitrot.
        You could even say that the attention the gnome and kde desktops get diminish the attention for others.

        So what's happening is that choice is disappearing.

        Maybe that's just because of GNUstep that is slowly developing. I don't know that.
        But I do hope GNUstep becomes/stays a choice.
      • The problem is I am an interface whore. I admit it, so there.

        I want a unified look and feel to my application. I tried to find alternatives in WM or WINGs apps but had trouble with dependency issues Postillion with tk particularily and FSviewer was screaming for an older version of a library that has already been updated a zillion times.

        GNUstep compile-all on core kept dying with what I think might be linker issues with the version Objective C I have on my box.

        Like what another user said the older WindowMaker look and feel apps a lot of them anyway have not been updated in awhile. Other apps like GNUstep itself just barf probably because from the requirements SuSE comes with 2.95.3 gcc instead of gcc 3.0 and it will not compile against the /opt/experimental version I put on from SuSE's gcc3.0 rpm.

        I have even thought of just running a bare bones sawfish windowmanager without the gnome stuff but have not gotten around to trying it on for size.

        After all, right now I use mostly gtk+ apps. Just a preference BTW not a religious point of view (no pro-KDE flames please I know its advantages).

        I ran WindowMaker for two years and had little trouble besides the fact it annoyed me that all the apps had a different look and feel.

        ________________________________________________ __
    • Glib response (Score:3, Interesting)

      by 2nd Post! ( 213333 )
      Get a Mac.
      DisplayPDF, Aqua window manager, a Dock, Finder (with three view modes),, and all for a low price of $1,799!

      You even get a DVD-R and LCD screen out of it :)
    • by hexix ( 9514 ) on Thursday January 10, 2002 @03:25PM (#2818292) Homepage
      I think you totally missed the point of this. It's not a lame attempt to make a NextStep look-alike. It is an attempt to recreate the OpenStep API for nice object oriented objective-c programming.

      If you want Gtk+ apps with a NextStep theme, then use Gtk+ apps with a NextStep theme, any current linux distro can give you this.
      • Too true. A NeXT look without the consistency or feel (no matter your skin, GTK+ apps always end up inconsistent!) is like putting lipstick on Bugs Bunny to make a beautiful woman. However, GNUsteps goal isn't just to implement the OpenStep API. It's also to create a usable set of applications that mimic the look, feel, and huge usability of the NeXT environment.
  • by Anonymous Coward
    When you do an install of FreeBSD, you can just choose the 'windowmaker desktop' option.

    Why pick yet another Linux Distro (This is this weeks Distro of the week) when you can pick FreeBSD, an OS that has shipped with a NeXTish interface for years.

    (Oh, and you get to avoid GNOME or KDE bloat, at least until you install gnumeric.)
    • I don't understand why they build GNUstep on linux. The logical choice would be to build it on Darwin. Then you would have the ability to create an interface which "...aims to be a user-friendly version of UNIX for the PC, similar to what Apple's OS X is to the Macintosh." (

      Not only that, but it would be a Mac OS like OS for PC that is based on the Mac kernel. Using the same kernel and API set as Mac OS would be really useful. There is already starting to be quite a presence of GNU software for Darwin (can be run on Mac OS X too, not just plain Darwin installs). You can check it out at []
      • I don't see that it really matters. Any lower level calls are just abstracted through the toolkit anyway. And Linux probably supports a lot more hardware and has better performance.

  • Why Linux?? (Score:5, Insightful)

    by aztektum ( 170569 ) on Thursday January 10, 2002 @02:21PM (#2817794)
    That's not meant as a flame. I'm just curious why they chose Linux instead of a BSD. I don't exactly care to follow the licensing terms behind all the different open/free software so I'm unaware if that's an issue, but using a BSD would seem to be a wiser choice being that they're "trying" to get an OS X on x86.

    I dunno.
    • Re:Why Linux?? (Score:2, Interesting)

      by kawlyn ( 154590 )
      I would agree, why Linux. Specifically why not Darwin, which arguably is OSX.

      OK perhaps darwin isn't ready for prime time, but hey....

      • From the article:
        • If Apple ever licenses Darwin as free software, Simply GNUstep could switch.
        This is a little misleading. The Free Software Foundation doesn't consider Apple's license to be free [], but puhleeease...Darwin is free software. It's just silly how RMS thinks he owns the word "free." Next he'll be expecting us to call it GNU/MacOS X.

        AFAICT, it doesn't matter that the APSL is GPL-incompatible, because a Darwin-based GNUStep wouldn't have to link to the Darwin kernel; Darwin is a microkernel design, so you don't need to link anything to it.

        Anyhow, GNUStep sounds like a great project. I'd kinda like to help out with it myself. I like MacOS X, but I figure Apple's going to be out of business within 5 years, and I need an exit strategy. Once GNUStep is running on Linux, it doesn't sound like it should be a big problem to get it running on Darwin (famous last words!) I really don't want to mess with an OS that forces me to recompile the kernel periodically.

        • Re:Why Linux?? (Score:2, Informative)

          by jslag ( 21657 )
          The Free Software Foundation doesn't consider Apple's license to be free [], but puhleeease...Darwin is free software. It's just silly how RMS thinks he owns the word "free."

          Would you think it was silly if you put lots of work into modifying Darwin for internal use, and then realized that you were legally obligated to publish your changes? Because, according to the link you provide, that's why the FSF doesn't call Darwin free.

          • I would think I was silly if I put lots of work into modifying Darwin for internal use, and then realized that I was legally obligated to publish my changes. It would be silly not to read the license before undertaking such a big project.

            IMO, BSD-style licenses are more free than GPL-style ones. Therefore, all GNU software is unfree. But so what? Why should you care what I think, and why should I care what you think? Nobody's forcing anyone to join a free-software project under a license they don't agree with. It's just misleading to state that Darwin isn't free, when really all that's meant is that it isn't free-as-in-RMS.

        • Re:Why Linux?? (Score:3, Insightful)

          by TWR ( 16835 )
          I like MacOS X, but I figure Apple's going to be out of business within 5 years, and I need an exit strategy.

          Put up or shut up.

          I'll wager good money ($1,000 sound good? How about $10,000?) that Apple will not be out of business within 5 years. We can define "out of business" as "having declared Chapter 7 Bankruptcy and in the process of selling off remaining assets."

          If Apple is purchased by another company and still making Macs/Mac OS X/etc., that doesn't count as "out of business" in my book.

          So, you willing to bet?


    • Re:Why Linux?? (Score:2, Insightful)

      by watchmaker1 ( 540289 )
      Because it's GNUstep. GNU. As in GPL. BSD is not GPL'ed. As far as I know the only official GNU blessing of the use of the BSD license was when RMS approved of ogg switching from LGPL to BSD to aid in acceptance with hardware makers.

      With FSF driving the bus, I don't see it making any forays into software which doesn't support the GPL.

      I, personally, have grown weary of Linux distros. Redhat is chock full of bloat, Two attempts at debian have left me frustrated and angry. I read on the "You can roll your own" like LinuxFromScratch, but dont have that much time to invest.

      When OpenBSD 3.0 came out, I installed it on a spare box just for giggles. I was shocked to find that it was EXACTLY what I expected from an OS. That box became my new cable modem router. I can't quantify it, but OpenBSD just FEELS right. If it did SMP, I'd have it on every box I own by now.

      • If you're still interested in finding a decent Linux distro, give Slackware a look. No fancy configuration wizards or package management, just good clean Linux.
  • GNUStep (Score:2, Interesting)

    by Anonymous Coward
    GNUStep is arguably more impoartant than KDE or GNOME for the future of linux, and it deserves a larger audience (for testing, etc), and it can be a pain to configure, compile, and install, so a distro is good for those reasons.

    However, I'm not sure it's ready yet. redraws are slower than mollasses headed uphill in January, and the sample applications (what little there is) are characterized by a lack of features and a tendency to crash.

    When GNUStep is ready for prime time, I'll be happy. Hopefully, this can help that day come sooner
  • Good to see (Score:4, Insightful)

    by Laxitive ( 10360 ) on Thursday January 10, 2002 @02:23PM (#2817802) Journal

    I remember trying to get GnuStep to work a few months back. The code compiled pretty cleanly, and I played around a bit with the development framework for GNUStep (which is rather cool btw, makes writing build files for apps extremely clean, and ObjectiveC is an extremely nice language).

    I just wish there was a better way of integrating GNUStep, KDE, and Gnome. I really think a concerted effort by all three teams to support a common base (common component interfaces, clipboard, look&feel configuration files) would be beneficial for all involved.

    GNUStep brings with it a good, tried&true development framework.

    KDE & Gnome are both more evolved, with more and better applications.

    Getting these to work together would be a worthwhile proposition.

  • by Arcturax ( 454188 ) on Thursday January 10, 2002 @02:23PM (#2817811)
    Looks like it is x86 for now. Anyone know if they are aiming for a PPC version as well?

    Of course that may be a bit pointless given you can get Mac OS X which is more mature, but it would still be interesting to see it.

    What willbe really interesting is if this becomes kine of like a "Mine", allowing PC users to run some select Mac OS X software along with their Linux apps. Think of this as a way for Apple to take a more back door foray into the x86 world, to expand use of Apple's software and show people how cool some of Apple's software on the Mac is by getting iTunes and such to run on this thing. It would certainly give PC users who used this (who I admit would likely not be your run of the mill PC user) a taste of Apple's world without them having to go out and purchase a Mac right away.

    Maybe wishful thinking, but any alternative to the current status quo (i.e. Windows) is welcome at this point even if it doesn't do all I hope it eventually can.
    • What willbe really interesting is if this becomes kine of like a "Mine", allowing PC users to run some select Mac OS X software along with their Linux apps.

      Pretty unlikely. Remember, its partially *source* compatible. OS X PowerPC binaries compiled against Cocoa cannot possibly run under Linux/GNUstep on x86. Since most Mac software is distributed commercial in binary-only form its unlikely that any of it will be instantly available on linux, and even if it is, they are as likely to port it using something "standard" like Qt even if it means rewriting much of the program, than to try to get it to work with GNUStep in its present form.

      And certainly software sold by Apple itself, like iTunes, is about as likely to be ported to linux as, say, Microsoft Office, GNUstep or no.
      • by Rob Parkhill ( 1444 ) on Thursday January 10, 2002 @02:49PM (#2818010) Homepage
        [sigh], if only OS X and GNUstep were just a little bit more like NeXTstep, and this wouldn't be a problem.

        NeXTstep has something called "fat binaries". It means that the same application that runs on my 68040 NeXT machine will also run on my HP and Intel machines running NeXTstep. When I compile my application, I simply tell the compiler to compile for all these platforms.

        Sure, the binaries end up being bigger (about 50% larger for each platform, if I recall), but it was completely painless, and you could use "lipo" to reduce the binary size (and make the application only run on a single platform) if you wanted.

        Yes sir, this was state-of-the-art back in 1992. 10 years later, it's all but disapeared.
        • by xil ( 151104 ) on Thursday January 10, 2002 @03:17PM (#2818224)
          Mac OS X still has fat binaries, actually. You should be able to build binaries which will run on Mac OS X or Darwin (PPC) and Darwin (Intel).

          Obviously the Darwin/Intel contingent is pretty small, so I haven't seen this in practice.

          On my Mac OS X 10.1.2 box:

          % which lipo
          • Multiple executables are supported in the OS X package system, so you could in theory build a Cocoa app which you could run on a Mac, copy across the network to your PC running OS X, double click on the same icon, and run it there. But this is all academic as Cocoa is only on Macs.
        • by Anonymous Coward on Thursday January 10, 2002 @03:17PM (#2818231)
          GNUStep has the framework for fat binaries in place. An application is actually a folder with an executable, TIFFs, and property lists. Running an application is done by running a shell script that searches common application locations, checks what processor and OS you're using, and runs the appropriate binary. The biggest shortcoming is GCC, which can only compile for 1 architecture at a time.
  • openstep (Score:3, Insightful)

    by Syre ( 234917 ) on Thursday January 10, 2002 @02:26PM (#2817828)
    Back when Apple bought NeXT, I wrote a letter to Jobs suggesting that he release NeXTstep for every platform, make it open source, and become one of the standard APIs that developers use.

    My point was that if he did this, and developers did adopt the platform, he'd end up with lots of apps that would run on the Mac, and would thereby neutralize the Windows API proprietary boondoggle.

    He ignored my advice at the time, and this is nice, but too little too late to solve the problem of creating a true platform-independent API that developers would want to write to.
  • Corrections (Score:4, Informative)

    by MouseR ( 3264 ) on Thursday January 10, 2002 @02:35PM (#2817890) Homepage
    based on the GNUstep architecture, originally built by NeXT (OpenSTEP) and is now also used by MacOSX (Cocoa).

    The above is wrong. The original NeXT Computer OS was called NeXTSTEP (notice capitalisation--it's important in what follows). When NeXT Computer ditched hardware, it became "NeXT Software", and spun off it's OS (in the 3.x version) into a cross platform OS called OpenStep (4.x).

    This OS was to run on Sun, Intel and NeXT boxes. The API was modified, and made public (the API, not implementation).

    This API specification was called OPENSTEP (capitalisation differs from the NeXT Software OS name).

    GNUStep is therefore based on the OPENSTEP specification. No other permutation of name and inheritance is correct.
    • > The original NeXT Computer OS was called NeXTSTEP
      > (notice capitalisation--it's important in what
      > follows).

      Oh so close, but not quite. The original OS was NeXTstep. Got changed to NeXTSTEP around 3.0.
    • Re:Corrections (Score:3, Informative)

      by droleary ( 47999 )
      Clearly the people modding you up don't know you're clueless yourself. The original OS was called both NeXTSTEP and NEXTSTEP by NeXT, and OPENSTEP was the follow-up OS. With it came a cross-platform API called OpenStep. Perhaps it's asking too much for people to actually go to the GNUstep site and click on the "GNU & OpenStep" link?
    • Another correction.

      OpenStep ran on x86, sparc, motorola (68k) and hppa architectures.

      OpenStep tended to have poor integration and "feel" on all of the above archs except the native 68k on real black hardware.
      • Correct the correction. NeXTSTEP ran on those architectures (I used to do fat binaries for all of them). Then the OpenStep stuff happened to support other base OS's - Solaris, Windows, others. NeXTSTEP changed along with it. That's about the time I left the NeXT world but watched from the sidelines (the people I worked for owned a chunk of NeXT).

        But you're right about the poor integration. Here's a comment from some code I wrote when NeXTSTEP for Intel was first shipped,

        #ifdef NSFIP
        * memcmp (as per ANSI definition) was shipped broken on
        * NeXTStep for Intel Processors. This should be removed
        * sometime after it is fixed.

        Oh boy. When memcmp doesn't work what else is broken :) The C library version of memcmp assumed big-endian [68K] and did word accesses to compare ints rather than bytes, IIRC it messed up the end case, comparing the wrong bytes of the last word. There was no problem when optimizing, gcc uses a (good) builtin, but turn off -O and things broke.

        But it all lives on... I bought a Mac a while ago to run Mac OS X. Immediately after getting it I install OS X and reboot. And a big smile comes to my face. It was like stepping back in time. The blue background, the spinning color wheel, the boot graphics, the login panel. The look's updated a bit but it's NeXTSTEP. Services, NetInfo, .app's, Interface Builder. Finally, its all out there again.

  • by 2Bits ( 167227 ) on Thursday January 10, 2002 @02:40PM (#2817926)
    What's so hard to grab some screenshots and put them up there? I've used the NeXT machines before, and have used WindowsMaker, so I know what to expect. But still, I'd like to see screenshots. And if it looks ugly, I won't even bother. And I want to see the boot up screen too.

    And you can get more users to try it out too, if they can see something before they download that 110MB of data. Even at that "small size", it's still a lot, for people like me who don't have access to high speed internet.
    • by Kaypro ( 35263 )
      Was lucky enought to grabe this minutes before it was posted.

      Bootup is EXACTLY like RedHat/Mandrake.... guess they didn't implement that part yet.

      After booting it starts up X in vesa mode and up pops the GUI... It looks...well..exactly like WindowMaker... I'm guessing cause thats what theyre using.

      As far as apps go it's got a terminal, email, and stuff...nothing out of the ordinary.

      They do have a nice IDE Development though.

      I'll check it out next release probably.... it seemed pretty fast. Worth checking out.... just make sure you burn this on an CD-RW.... ;-)

    • Go here [] if you want screen shots. The link on the story post was just for the download.

      If you follow the link to OS News [] there's a link to the left for screenshots and a bunch of other stuff
  • by xtremex ( 130532 ) <cguru AT bigfoot DOT com> on Thursday January 10, 2002 @02:40PM (#2817930) Homepage
    I went to the site and the ISO they have for download doesnt exist! isnt there!
  • Not there yet. (Score:4, Informative)

    by foonf ( 447461 ) on Thursday January 10, 2002 @02:41PM (#2817932) Homepage
    I downloaded the ISO and tried to boot into it. Immediately the kernel fb support gave an "unsupported display mode" error. I checked the available modes and there was nothing but text modes available. I have a Matrox G400, which is very well supported by the kernel framebuffer drivers. In fact I use the fb console at 1024x768 on my real linux installation without problems.

    The startup looked interesting, at any rate. It failed to detect my NIC (a pretty standard DEC Tulip card) and gave a few other errors I can't remember. Then it tried to run X, but since it was configured to use the framebuffer driver, which wasn't working, it choked. Needless to say if they had at least allowed the option of using the XFree86 accelerated drivers, it would have been fine, but they don't. So then the system shut itself down. This worked ok, except their kernel is compiled without APM support, so it didn't actually turn itself off.

    Oh well. I'll try it again later, seems like a decent idea really.
    • Re:Not there yet. (Score:5, Informative)

      by cehardin ( 163989 ) on Thursday January 10, 2002 @03:24PM (#2818291)
      The problem is that many of the newest video cards do not support the VESA 2.0 standard.

      As this was mainly a demo cd, I was more concerned with getting it to run on as many machines as possible with minimal effore (therefore, VESA)

      The actual installation disc I'll end up creating will not try to use the VESA framebuffer.

  • Simplicity is key (Score:2, Insightful)

    by xt ( 225814 )
    I think the really interesting part is the IDE; If you have a fairly simple way to extend your computer experience, you get a little more attached to what you use and enjoy it a lot more.

    Don't forget there are a lot of intelligent people who enjoy tinkering with things (computers and OSes included) but can't afford to spend too much time...

    If the developers also choose a well rounded set of applications, then we'll have an interesting alternative to packing a zillion apps (almost) noone will use and creating yet another distro that confuses users about the choices, rather than being itself an alternative choice.

    By having something simple you can use and extend, you are also a lot more motivated to actually use it and stick with it, rather than observe at amazement and then go back to .

    Perhaps, simplicity is itself a choice sorely missed all too often nowadays...
  • by David Ishee ( 6015 ) on Thursday January 10, 2002 @03:01PM (#2818098) Homepage
    If this is a Linux based system, will RMS flame us is we don't call it GNU/Linux/GNUStep?
  • by Glock27 ( 446276 ) on Thursday January 10, 2002 @03:04PM (#2818120)
    I'm a Java fan, drawn to it by (relative) simplicity, decent OO and speed.

    That said, Objective-C also has many fine attributes, and has never gained the popularity it deserves. Objective-C (gcc is Apple's Obj-C compiler also) is fully compiled and has great legacy compatibility with C, both desirable attributes when compared with Java. There are other tradeoffs between the languages, but Objective-C looks like a great Java alternative in certain circumstances. It also looks like a fun 'recreational language' for side projects.

    I was considering one of the new iMacs anyhow, it's good to see that much code might port to an open source setting also! :-)

    299,792,458 m/s...not just a good idea, its the law!

    • Objective-C is great, but try using it when you need to link in some C++ libraries some time. For fun, try linking in C++ libraries built with the Sun WSPro compiler.

      Java is the hands-down winner here. In fact, if you are developing WebObjects code, and need to use C++ libraries, Java is the best way to go! It's that, or write C wrappers for your C++, and call those from Obj-C. Ugh.

      To be fair, the problem is most likely that Obj-C has been all but abandoned, and never has worked properly alongside C++. Had Obj-C received the development resources of Java, it wouldn't have these drawbacks.
      • by bnenning ( 58349 )
        Apple has made modifications to gcc to support "Objective C++", which allows Objective C code to use C++ syntax and classes. Hopefully these changes will eventually be merged into the main gcc sources.
        • by cbv ( 221379 )

          Apple has made modifications to gcc to support "Objective C++", which allows Objective C code to use C++ syntax and classes. Hopefully these changes will eventually be merged into the main gcc sources.

          Take a look at that proposal [] by Ziemowit Laski ...

        • Objective-C++ doesn't allow you to use C++ syntax to program in Obj-C, nor does it allow you to use the Objective-C runtime and syntax to instantiate and send messages to C++ objects, nor does it allow you to use C++ to instantiate and send messages to Objective-C classes and objects.

          What it does allow you to do is mix Objective-C and C++ in the same source file, using C++ syntax to deal with C++ classes and objects, and the Obj-C runtime/syntax to deal with Obj-C classes and objects. They're still quite seperated. However, I don't imagine it would be hard to have an automated conversion process going. The only real blocking point on that one is the static and opaque nature of C++, where Objective-C is dynamic and reflective enough to allow such tricks.
      • by eweu ( 213081 )
        To be fair, the problem is most likely that Obj-C has been all but abandoned, and never has worked properly alongside C++.

        That just isn't true. Obj-C is not popular or widespread, but Apple continues to develop the language. Take a look at the release notes from Project Builder:

        Mac OS X 10.1 introduces the Objective-C++ front-end to the Mac OS X version of the GCC compiler. Objective-C++ allows you to freely mix C++ and Objective C code in the same source source file. Using Objective-C++, you can directly call Objective-C objects from C++ code, and you can directly call C++ code from Objective-C objects. Thus, Objective-C++ allows you to use C++ class libraries directly from within your Cocoa application, or to use Cocoa or Foundation objects directly from within your C++ application.
        Since I do this quite a bit, I'd have to say Obj-C and C++ play quite nicely together.
        • Hmmm, I should have been a bit clearer. Obj-C and C++ work together just fine, and I have mixed-and-matched C, C++, and Obj-C all in the same source file, just as Apple suggests. And even calling C++ libs directly from Obj-C works great, when you are always using the same compiler for both (ie. gcc)

          However, when you have a precompiled C++ library that used a -different- compiler from the gcc you are currently working with (such as the Sun WSPro compiler, or an older version of gcc), this is where the headaches begin.

          Perhaps this is just a compiler issue, and not an Obj-C issue, but seeing as the only compiler that supports Obj-C is gcc, I tend to (incorrectly) lump the two together.
  • Close... (Score:2, Interesting)

    by Fnkmaster ( 89084 )
    I like the general approach. While GNOME and KDE are nice, I don't think either of them does it for me, or anybody with a more aesthetic view of the world. Go use OS X for a while, and try going back to GNOME or KDE. You will feel a serious sense of loss. Same goes for Windows, frankly.

    I like the Linux core, I enjoy and am comfortable with Linux as a kernel and the GNU/Linux combo as an OS. I want a better GUI on top of it that pleases my aesthetic sense, makes my life easier when I want it to be easier, doesn't feature at least two separate tracks of font management systems, lots of apps of massively disjoint look-and-feel and more widget toolkits than I care to think about. In other words, Simply GNUStep is a good move, but why don't we consider dropping the X windows? Furthermore, why don't we consider taking this a step further? Hell, OS X took the old NextStep stuff and improved it dramatically. Why don't we do the same, and not be constrained by OS X or attempt to parrot or copy it, and see if we can improve on it?

    I agree that source level compatibility with OS X is a nice feature at this point in time since lots of Cocoa apps are being written (primarily because OS X is doing so well), and I like the *Step environments. But I'd like to see some innovation from the Open Source world too.

  • by Genady ( 27988 ) <gary@rogers.mac@com> on Thursday January 10, 2002 @03:11PM (#2818174)
    Think about it, build Gnustep over Darwin x86 and you have... A Frankenstein's monster version of Mac OS X for x86.
  • Not a new OS (Score:3, Insightful)

    by Anonymous Coward on Thursday January 10, 2002 @03:12PM (#2818178)
    This isn't a new OS, it's a linux distroid designed to focus on and exploit GNUStep.

    GNUStep isn't an OS, it's the API from NeXTSTEP.

    It's supposedly really really cool to program in because Objective-C is a lot more dynamic in its design than C++. (Much less type checking = less recompilation, more rapid development, it's a lot more like working in Smalltalk or a scripting language like Ruby. So I hear from people who use Objective-C in my company.)

    HOWEVER it ain't ready, GNUStep is still laying the foundations. When they're all laid, it should be possible to add a lot of very good apps very fast. (NeXT is most famous for having been something you can develop apps very well and fast in.)

    This is an interesting start.

    GNUStep apps should be relatively easy to port to Cocoa and vice versa, that's the extent of the connection.

    All this yammering about how pretty the window decorations are is silly. It ain't about looking like candy, it's about being pleasant to use and working well.

    All this stuff about "being the next BeOS" is silly too. This isn't about users... not yet. It's about developers. It isn't a new OS, it's a new programming environment and a Linux distro optimized for it.
  • Terms & Definitions (Score:5, Informative)

    by maggard ( 5579 ) <> on Thursday January 10, 2002 @03:21PM (#2818257) Homepage Journal
    For those confused by what's what this might help bring folks up to speed (and keep the discussion coherent):
    • NeXT: The next business founded by Steve Jobs after being pushed out of Apple (to sell the next generation of computers.)
    • NeXTSTEP: The black cube then slab with object-oriented OS based on Unix sold by NeXT Computer.
    • OpenStep: The NeXT OS ported to 5 different architectures and sold as a stand-alone product.
    • Objective-C: The language OpenStep is written in. An object oriented extension of C considered by many to be cleaner then C++.
    • GNUstep: The reimplementation of OpenStep by gnu-folks.
    • Rhapsody: OpenStep after Next buys Apple for -$400 million. Reworked to be their next OS. To run on Macs and under Wintel. Dropped after developers refuse support.
    • MacOS X: Shipped version of next-gen MacOS. OpenStep-derived kernel & Cocoa layer along with legacy MacOS compatabilty & virtualization evironments. Publically PPC only.
    • MacOS X Server: Same thing, different focus on services.
    • Cocoa: The layer in MacOS X that along with the kernel is still closely OpenStep.
    • Darwin: The Open Source PPC & x86 core of MacOS X - doesn't include Cocoa.
    • Simply GNUstep: GNUstep coupled with a Linux distribution.
    • You forgot:

      GoATSTEPx: a wide-open public-domain environment supporting hot-plugging of peripherals soon to be cleaned up and released by Microsoft. (It's easy to tell where the first patch will be applied!)

    • Close... but not quite.

      NeXTSTEP: Not the hardware, but the original OS made by NeXT. Versions 0.3 - 3.3.

      The NeXT cube, NeXT Turbo Cube, NeXT Station, NeXT Color Station, NeXT Turbo Station, and NeXT Color Turbo Station are the names of the computers NeXT sold.

      OpenStep: The OpenStep API was a bit different than the original NeXTSTEP API. The "open" part of the moniker denotes that it was an open API spec, and OpenStep followed this spec. This allows for other implementations to happen, like GNUstep or OpenStep for Solaris. Versions 4.0 - 4.2.

      Yes, I did say OpenStep for Solaris. There was also an OpenStep Enterprise for Windows. OpenStep for Solaris allowed you to run an OpenStep environment and related apps on top of X11, rather like GNUstep does. It ran it's own OpenStep window manager, but still allowed vanilla X apps.

      OSE for Win32 was the OpenStep API for Windows, allowing one to develop and run OpenStep apps on Windows. While it didn't change the shell/wm like in OS/Solaris, you could run OpenStep apps remotely, including the Dock and which was pretty cool to see. The widgets were given a bit more of a Windows look.

      Rhapsody: Wasn't dropped entirely, as you say. The original Rhapsody vision was dropped, but Rhapsody was released as Mac OS X Server 1.x. Mac OS X Server 2.0 is based on Mac OS X 10.0, however.
  • What's the point? (Score:2, Insightful)

    by prototype ( 242023 )
    Okay, maybe I'm confused over a few things here but I'm failing to see the point of this project?

    WindowMaker has already been around for awhile and comes with it's own GNUstep like interface (or it is it's own GNUstep interface if you want to split hairs). If you don't like WindowMaker, then use AfterStep which again gives you the NeXT type interface (dock, clip, etc.). Either of these can be installed onto any Linux distro. You can install RedHat and get all the cool hardware detection with it and just don't install KDE/Gnome/etc. then grab the latest WindowMaker/AfterStep files and you have the same thing this is offering. So where's the magic?

    Some of the features it's touting:
    Uses the latest linux kernels and its latest features (ie: pure devfs, framebuffer)

    Great, except according to some people here it has a lot of problems just installing. Besides, in a few weeks (or whenever the next update happens) the latest kernels will be out of date. You may as well just ftp your own kernels and compile them for your own system.

    Graphical Boot-Up (no confusing Linux kernel messages)
    Personally I like seeing the messages boot up so I know what sub-systems and modules are being loaded. If my sound module fails at least I know it.

    Kept as simple as possible (no GNOME, no KDE, etc, just GNUstep)
    Just install any linux distro without KDE/Gnome and slap on WindowMaker/AfterStep and you get the same thing right? So how is this a selling feature?

    So we've already got this if you want it. Just go and grab whatever window manager suits your taste. If this is a move towards Mac OS X compatibility then great, but it seems like a very small step as there is a LOT of work ahead to even get something close to that.

    Personally it just seems like a waste to bundle it with yet another copy of Linux. Separate it out (unless there's something special you're doing with the kernal) as a download so anyone can grab it in less than 10 minutes and let us decide which kernel to use for the base.

    At the very least, toss up a few screenshots, make the download availalbe in a few formats and provide a little more information about what features this has or will have. What's the big picture and where is it leading?

  • Right now, there doesn't seem to be any way of getting a complete installation of GNUstep through the Debian package system (you get some libraries, but that's all as far as I can tell). I wouldn't mind giving GNUstep a try, but I'm not going to throw out my whole Linux installation. I suspect many others are in the same boat, and a complete set of easy-to-set-up GNUstep packages for Debian would probably reach more people at this point than a separate distribution.
  • by redmenace ( 145551 ) on Thursday January 10, 2002 @04:24PM (#2818828)
    I've got to run to work (I'm late!), but if one of you could please put a post on the slashdot article explaining the problem I would be very

    It seems I broke a rule at sourceforge which limits file sizes to 100MB

    If someone could offer a mirror site I would be very thankful for that as well!

    Thanks! Chad

    Here's what sourceforge had to say



    My name is ********; I am the Quality of Service Manager for This message is directed to you since you are
    designated as a project administrator for the Simply GNUstep project on

    First, we would like to take a moment to congratulate you on your recent
    press exposure on -- we love to see Open Source projects
    succeed, and press exposure of this nature is always of great benefit.
    Since the announcement of your project efforts on /., your project has
    received roughly 2241 downloads of the ISO image provided through your
    download page at:

    It has come to our attention that you are making use of
    project web services as a mechanism to release file materials in excess
    of 100MB in size. Each project hosted on is provided
    with project web services as to ensure that they may adequately provide
    an online description and information regarding their project.

    ... etc
  • by cehardin ( 163989 ) on Thursday January 10, 2002 @04:24PM (#2818831)
    Sourceforge has a policy of not allowing DLs of files over 100MB, I broke that limit with my 110MB ISO image, so they took read access form the file to everybody but me.

    If someone would like to offer a mirror I would appreciate it very much!

    Chad Hardin
    • by PMcGovern ( 13300 ) on Thursday January 10, 2002 @09:35PM (#2820959) does indeed have a 100MB limit per project for project web server disk usage.

      We do, however, encourage people to use our File Release System (FRS) which does not have this 100MB limitation.

      The File Release System is recommended for a number of reasons; one of which is
      to allow us to balance the traffic load among many download servers on our high-capacity download network.

      Before we removed the file, Simply GNUstep had two thousand downloads at 110MB each, during the course of the day (totalling over 200 GB of data).

      Obviously we need to distribute this kind of load, which our file release system allows us to do; otherwise this kind of download traffic will impact the other 30,000 projects we host.

      If you have any other questions about this, please feel free to email me at pat (at)
  • compatibility (Score:2, Insightful)

    by Snuffub ( 173401 )
    "The cool thing about Simply GNUstep is its partial source compatibility with MacOSX programs"

    Arent all programs 'partialy source compatible' with an OS that has that particular compiller this could be great in terms of colaberation and future development but it also could be nothing mroe than empty hype.

  • ...and digging into SourceForge this way [] shows no files either. Anyone have a link that works?

"So why don't you make like a tree, and get outta here." -- Biff in "Back to the Future"