Minix Jobs for Me ?

[Patrick]

Thanks yet again Jean-Baptiste ! I have a CD image off and building now.

Hi Everyone

This experience makes me feel that the wiki needs some updating too.

I would like to help out.

Is there a way to revert changes on the wiki if I royally screw
something up?

What if I started going through things and I posted to the list with a
summary of any changes? You could roll back anything dumb.

Could I just go off an a bit of a tangent.......

I stopped using Windows as my primary desktop in 2004. I have been on
Linux since. I tried GhostBSD and PCBSD but they did not work well, this
was 7-10 years ago, I have no idea where they are now. Linux and a bit
of Minix is all I know.

I feel a bit squeezed between the BSD and GNU camps.

I love GnuCOBOL but it seems to need gnu autotools. I was able to get it
up and running on Minix but only by clobbering BSD packages with GNU ones.

I have several things I have been working on for years. It's often been
a lot of planning and progress has been very slow. One topic is
controlling scientific instruments.

China and India now have a huge marketshare in terms of laboratories. I
think it's quite hard to sue anyone in India and even Microsoft can't
sue anyone in China. I would never violate anyone's license but if I GPL
something, it's actually public domain as I could never enforce the GPL
where my software would likely end up being used. The GPL actually
forces me to public domain my work.

I know software licenses get everyone upset and I don't want to upset
anyone. It's just that I have an idea.

I have 15GB of sources from Trisquel Linux. Trisquel is a hard-core FSF
approved distro and again there might be grief between the camps.

However the developers have put a lot of effort into making sure that
their users will not get into software trouble(codecs etc) and it's a
good distro.

I am trying to build Minix with the goal of creating a second build
script to clobber some packages and to add a lot more.

I am hoping to get a desktop environment up and running, a web browser,
email client, word processor etc.

I feel like the project is getting squeezed between NetBSD which would
be a more popular choice for embedded and Linux which is ready for
everyday computing.

If I could get an everyday computing setup with Minux, I could keep
exploring it everyday while I work on other things and so could everyone
else.

I like the Ada language but this is where I think it falls on it's face.
The Ada community is strongly anti C language but C let's you get up and
running fast and can be used for many things. If something really needs
to write reliable software, Ada might be a good choice but is it ? The C
programmer that uses C everyday for lots of things might end up writing
safer software than the Ada programmer that is stuck with a limited set
of options.

If people could use Minix day in and day out, they might know it better
and might be more likely to use it in an embedded space and that
embedded deployment does not have to have any of the GNU packages I am
trying to install.

If I did something like this and I did not tamper with the BSD friendly
build that is the core of the project, I would not be violating the
goals of the project right?

I also feel like we should be getting the word out that Minix is much
smaller then BSD and Linux and easier to understand and manipulate. I
feel that at this point, this is the central selling feature. The
re-incarnation sever is nice but with with so little uptake of the
project, would someone feel more confident to use it or to use a more
mature project like NetBSD, which has been used in space ?

Sorry for the rambling, the other thing is how can I help? I really love
the project. I am back for the 4th time. Obviously I can't seem to get
away from it :)

I really appreciate all the help, I hope this post doesn't upset
anyone-Patrick

[Jean-Baptiste Boric]

Is there a way to revert changes on the wiki if I royally screw
something up?

Yes. The wiki keeps old revisions, so rolling back is always possible. 

 

What if I started going through things and I posted to the list with a
summary of any changes? You could roll back anything dumb.

I'll say preemptively to just go ahead. You're more likely to improve things, even if it takes a couple of subsequent edits to clean up.

 

I am trying to build Minix with the goal of creating a second build
script to clobber some packages and to add a lot more.

I am hoping to get a desktop environment up and running, a web browser,
email client, word processor etc.

I've managed to get enough stuff working to do that. It definitively did feel like a throwback to the 90s and X11 stability was spotty under stress, but it did work.

If people could use Minix day in and day out, they might know it better
and might be more likely to use it in an embedded space and that
embedded deployment does not have to have any of the GNU packages I am
trying to install.

It's been a while since my last true testdrive, but it does feel like no one's using MINIX3 as their primary, everyday OS these days. It works, but it doesn't mean the experience is as polished as any of the BSDs out there.

If I did something like this and I did not tamper with the BSD friendly
build that is the core of the project, I would not be violating the
goals of the project right?

There's hardly anything you can do to violate the goals of a BSD-licensed project. 

I also feel like we should be getting the word out that Minix is much
smaller then BSD and Linux and easier to understand and manipulate. I
feel that at this point, this is the central selling feature. The
re-incarnation sever is nice but with with so little uptake of the
project, would someone feel more confident to use it or to use a more
mature project like NetBSD, which has been used in space ?

I'll prefix this answer by saying I am tainted by looking at xv6 and Fuchsia/Zircon and their modern design choices.

I'll be honest, by default I would have a bit of a hard time justifying MINIX3 over NetBSD if I were in a position to make such a call at work, if only because MINIX3 needs modernization in several areas (not 64 bit capable, no SMP, no kernel threads, no RUMP support for drivers, dated microkernel design). MINIX3 does have very interesting characteristics and advantages over NetBSD for deeply embedded work (hence Intel ME), but it does need someone with quite a bit of free time to bring it kicking and screaming into the 21th century.

Sorry for the rambling, the other thing is how can I help? I really love
the project. I am back for the 4th time. Obviously I can't seem to get
away from it :)

Play around with MINIX3 and try to fix issues you run into. That's how I started here.