Where is Minix 3 going ? - Minix

This is a discussion on Where is Minix 3 going ? - Minix ; Hello everybody Where is Minix going or whats the target of this OS? After reading Summer of Code project suggestions I`m a bit confused. I thought Minix will go into direction like FreeBSD (which I use a lot) to a ...

+ Reply to Thread
Results 1 to 10 of 10

Thread: Where is Minix 3 going ?

  1. Where is Minix 3 going ?

    Hello everybody

    Where is Minix going or whats the target of this OS? After reading
    Summer of Code project suggestions I`m a bit confused. I thought Minix
    will go into direction like FreeBSD (which I use a lot) to a user
    friendly personal plaform for X software, Mulitmedia, developing,
    gaming etc. But GSo suggestions points to embedded hardware and
    keeping the system small which is a complet other direction.

    By the way, would it not better to use a forum than a old usergroup
    with a lot spam. That`s the best from of communication in my opionion
    Are people working right now at the project? When is the next
    release?

  2. Re: Where is Minix 3 going ?

    Erik van der Kouwe
    >> gaming etc. But GSo suggestions points to embedded hardware and
    >> keeping the system small which is a complet other direction.

    >
    > The latter direction is the intended one. Actually, the main goal is
    > education, for which goal it is required that Minix remain relavtively
    > small.


    > Don't expect it to ever become user-friendly or a multimedia and gaming
    > platform, as that would not only go counter to its goals,


    Excuse me for this interference but I'm strongly interested in this topic.

    As long as I can see, Minix3 is "just" a kernel that in the near/far
    future could be actually used to replace the Linux or the BSD kernel in
    a quite "normal" distribution, like Debian. There is nothing in Minix3
    that makes it "doomed" to embedded systems. It is perfectly possible to
    adapt it to ambedded systems /and/ desktop ones.

    It is already possible to install Minix3 in a emulator or in real PC and
    add to it X11 (with TWM or Equinox) and TCP/IP support. The X11 layer
    just uses the same code of Linux and BSD and makes possible to have the
    same level of user-friendliness. Once you have X11, developing a
    user-friendly desktop-oriented system should not be very hard (thanks to
    Equinox or maybe, in the distant future, thanks to
    KDE/Gnome/Enlightenment). Moreover, Minix3 is designed to ease the task
    of developing device drivers and other non-kernel code so it should not
    be very hard to port Alsa or other multimedia/games components to it.

    I mean: Minix3 could became very similar to any normal Linux or BSD
    distro in the near/distant future. It is just a matter of manpower.

    As a demostration of this, just think to the fact that Minix3 is
    currently being ported to OLPC, so it will be /forced/ to became
    desktop-oriented and user-friendly at least as much as Linux and BSD
    currenlty are. It is also possible that this will promote the porting of
    Minix3 to UMPC and mobile telephones (a là android/openmoko/limo). There
    are people working to a ARM porting, already...

    Beside this, it is perfectly reasonable to keep Minix3 as small and
    simple as possible. This does not conflict with the goal to be
    user-friendly. Think to QNX and its small GUI (Photon), for example. It
    should not be impossible to adapt a small GUI library like PicoGUI to
    Minix3 (what about DirectFB?).

    BTW, QNX is a good example of what could became Minix3 in the future.

    > but there
    > also simply is not enough manpower in the Minix community to achieve
    > such goals. AFAIK nearly all development is done by people from the
    > Vrije Universiteit Amsterdam; there does not seem to be a large
    > community outside the university as with other OSS Unices.


    Unfortunatley, that's true. But it does not have to be so forever.
    Minix3 is being reviewed more and more on web sites and paper magazines
    and is more and more used as a educational tool in many universities. It
    is just a matter of time to have a reasonably large community of
    developers working at it.

    IMVHO, the two main reasons why Minix3 is not collecting a larger
    community of users and developers are these:

    1) A desktop-oriented (x11+equinox), usable distro is missing. There is
    nothing like Debian or Ubuntu in the Minix3 world and this makes
    relatively hard to try Minix and experiment with it for the end user.
    Having a smaller community of end-users means to have a less-appetizing
    system for developers.

    2) It is not clear what makes Minix3 "better" or just "different" from
    Linux and BSD. Put aside the more readable code (useful for educational
    purpose), Minix3 is very similar to a scaled-down version of Linux, like
    the ones you can see inside your router or inside many embedded systems.
    As a consequence, developers devote their efforts to Linux.

    As long as I can see, it is time to develop a small, desktop-oriented
    distro in order to promote Minix3, to create a larger user basis and to
    stimulate developers. It is also time to develop a "reference
    application" that makes clear the overall goal of Minix3. The OLPC
    porting could be this application.

    JM2C

    --

    Alessandro Bottoni
    Website: http://www.alessandrobottoni.it/

    "The opposite of a correct statement is a false statement. But the
    opposite of a profound truth may well be another profound truth."
    -- Niels Bohr

  3. Re: Where is Minix 3 going ?

    On Sat, 03 May 2008 18:01:32 -0700, kforstho wrote:

    > Where is Minix going or whats the target of this OS? After reading
    > Summer of Code project suggestions I`m a bit confused. I thought Minix
    > will go into direction like FreeBSD (which I use a lot) to a user
    > friendly personal plaform for X software, Mulitmedia, developing, gaming
    > etc. But GSo suggestions points to embedded hardware and keeping the
    > system small which is a complet other direction.


    I don't believe that the "target" issue makes much sense. At the moment
    minix lacks fundamental functionality like USB support. As soon as the
    fundamental features are implemented then more people will end up giving
    minix a try and also contributing to it. To put it in other words, things
    will start to pick up momentum. Then it is just a matter of scratching
    your own itch, which was what took other kernel projects from small
    desktop systems to multiple architectures and both servers and embedded
    systems.


    > By the way, would it not better to use a forum than a old usergroup with
    > a lot spam. That`s the best from of communication in my opionion Are
    > people working right now at the project? When is the next release?


    If you happen to believe that webforums are somehow better than
    newsgroups then I have to say that something is fundamentally broken with
    your perception.


    Rui Maciel

  4. Re: Where is Minix 3 going ?

    > As long as I can see, Minix3 is "just" a kernel that in the near/far
    > future could be actually used to replace the Linux or the BSD kernel
    > in a quite "normal" distribution, like Debian. There is nothing in
    > Minix3 that makes it "doomed" to embedded systems. It is perfectly
    > possible to adapt it to ambedded systems and desktop ones.


    This could become reality, but I think it applies more to the far than
    to the near future. Getting a full user-friendly environment requires
    porting much software to Minix. I must admit to not having tried
    Equinox, but TWM is not what most people would describe as a
    user-friendly environment.

    Some software can easily be ported, sometimes as easy as running
    "configure; make install", but getting all the software that is needed
    for a user-friendly experience will take some time.

    This is especially the case due to the lack of virtual memory and a
    number of system calls. When a stable release comes out including these
    features, this may speed up porting dramatically. The people working on
    this told me there will most likely not be such a release for at least
    another year.

    > IMVHO, the two main reasons why Minix3 is not collecting a larger
    > community of users and developers are these:


    One major issue besides the ones you mentioned is drivers. Most of the
    ones posting here are willing to modify some of the drivers to get
    their hardware, especially ethernet cards, to work. Most of the public
    at large is neither willing nor capable of doing this. Again, this can
    be resolved in the future, but for now it inhibits adoption on a large
    scale.

    > As long as I can see, it is time to develop a small, desktop-oriented
    > distro in order to promote Minix3, to create a larger user basis and
    > to stimulate developers. It is also time to develop a "reference
    > application" that makes clear the overall goal of Minix3. The OLPC
    > porting could be this application.


    That is very true, and I hope this would make Minix more appealing to a
    large audience. One major advantage of the OLPC project is that the
    hardware is uniform, reducing the need for new drivers. Still I expect
    it to take much time before there is a new stable release including
    virtual memory which I think is currently the major dividing line
    between Minix and other kernels.

    --
    With kind regards,
    Erik van der Kouwe

  5. Re: Where is Minix 3 going ?

    First of all thanks for your replies.

    quote
    As long as I can see, Minix3 is "just" a kernel that in the near/far
    future could be actually used to replace the Linux or the BSD kernel
    in
    a quite "normal" distribution, like Debian. There is nothing in Minix3
    that makes it "doomed" to embedded systems. It is perfectly possible
    to
    adapt it to ambedded systems /and/ desktop ones.

    It is already possible to install Minix3 in a emulator or in real PC
    and
    add to it X11 (with TWM or Equinox) and TCP/IP support. The X11 layer
    just uses the same code of Linux and BSD and makes possible to have
    the
    same level of user-friendliness. Once you have X11, developing a
    user-friendly desktop-oriented system should not be very hard (thanks
    to
    Equinox or maybe, in the distant future, thanks to
    KDE/Gnome/Enlightenment). Moreover, Minix3 is designed to ease the
    task
    of developing device drivers and other non-kernel code so it should
    not
    be very hard to port Alsa or other multimedia/games components to it.
    quote

    Great posting Alessandro. I see it the similar way. Minix has some
    potential, and with some better hardware support, better installer,
    bug fixes, fixed Ram, more keyboard layout for the consol and Xfree or
    Xorg it would be possible to make a working multimedia platform.
    That`s how it works I in nearly all opensource community. People get
    involved into an OS if their hardware is supported or if minimal
    software (like Abiword, Xmms, Firefox) is supported. More people
    involved mean hugher community and the chance is high that they give
    something back to the OS, like porting their favorite software to
    minix or writing new drivers. So I was really confused, that making
    the Minix more userfriendly is not really a goal for google`s summer
    of code.

  6. Re: Where is Minix 3 going ?

    On May 4, 5:51*pm, kfors...@email2me.net wrote:
    > The X11 layer just uses the same code of Linux and BSD and makes
    > possible to have the same level of user-friendliness. Once you have X11,
    > developing a user-friendly desktop-oriented system should not be very
    > hard (thanks to Equinox or maybe, in the distant future, thanks to
    > KDE/Gnome/Enlightenment).


    In fact, I have a working, stable GTK+ port. I just need to find the
    time to pack
    things up. With that we could already have Xfce pretty quickly.
    AbiWord and
    Gnumeric should be just around the corner (Gimp's in the works).

    Marco

  7. Re: Where is Minix 3 going ?

    On 2008-05-04, Erik van der Kouwe expressed:
    >> As long as I can see, Minix3 is "just" a kernel that in the near/far
    >> future could be actually used to replace the Linux or the BSD kernel
    >> in a quite "normal" distribution, like Debian. There is nothing in
    >> Minix3 that makes it "doomed" to embedded systems. It is perfectly
    >> possible to adapt it to ambedded systems and desktop ones.

    >
    > This could become reality, but I think it applies more to the far than
    > to the near future. Getting a full user-friendly environment requires
    > porting much software to Minix. I must admit to not having tried
    > Equinox, but TWM is not what most people would describe as a
    > user-friendly environment.


    Isn't it possible to use code distributions (like Gentoo?)

    That would make the porting available to a lot of people.

    Greetings,
    Frank

  8. Re: Where is Minix 3 going ?

    As long as it is BSD it wont gain wide spread adoption by community or
    developers.

    Minix should be under GPL.

    RMS is asking to Unix under GPL, even OpenSolaris will be going GPL
    soon seeing how SUN lately is loving GPL :P

  9. Re: Where is Minix 3 going ?

    > As long as it is BSD it wont gain wide spread adoption by community or
    > developers.
    >
    > Minix should be under GPL.
    >
    > RMS is asking to Unix under GPL, even OpenSolaris will be going GPL
    > soon seeing how SUN lately is loving GPL :P


    That doesn't make sense. Minix can be used much more widely with BSD
    than with GPL. This may be an issue in the embedded market, were one
    specifically targets commercial projects. This could be a competitive
    edge over Linux.

    If developer would not be willing to develop BSD because they want the
    restrictions that GPL provides, then the BSD unices would not be here.

    --
    With kind regards,
    Erik van der Kouwe

  10. Re: Where is Minix 3 going ?

    --{ for.dev.null@gmail.com a plopé ceci: }--

    > Minix should be under GPL.


    Why ? And is your newsreader under GPL ?


    --
    { SIGWHAT?!!, "BACKSIGNAL" },
    /* the processus signal the kill program it disagree
    with the previous signal given. See anarchy(3) */
    --{ f.m.b.l revisite la command kill }--

+ Reply to Thread