“An early preview of the next version has been released under version number 3.1.3, an interim release intended for developers who want to see the bleeding edge of MINIX and perhaps then track the current source. For more information, see the 3.1.3 release notes and the download page.” MINIX3 will now also release weekly snapshots.
Originally, Minix 1 and 2 were simply learning tools. Minix 3 is focusing on being an OS for embedded systems or resource limited computers. I’ve seen a few decent ports lately for Minix (Links comes to mind) and hopefully soon it will be a viable alternative to Linux for embedded systems.
Minix could only be a viable alternative if the embedded system used x86 architecture.
But IIRC, ARM is used much more frequently.
>Minix could only be a viable alternative if the embedded system used x86 architecture.
What do you mean? Does Minix only run on x86?
If this is the case, I find it very funny, remembering that in the flamewar Tanenbaum vs Linus, one criticism against Linux was that it was x86 only at the time.
> What do you mean? Does Minix only run on x86?
I’ll quote from the main page of minix3.org:
> To run MINIX 3, you need a PC driven by a 386, 486,
> or Pentium CPU or compatible.
and the FAQ:
> What hardware do I need to run MINIX 3?
> You need an Intel 386 or higher with 4 MB of RAM, an
> IDE hard disk with100 MB of free disk space, and an
> IDE CD-ROM for booting. It is not possible to boot
> off a USB CD-ROM drive (yet).
And from the Tanenbaum vs. Torvalds debate (Tanenbaum quote):
> MINIX was designed to be reasonably portable, and
> has been ported from the Intel line to the 680×0
> (Atari, Amiga, Macintosh), SPARC, and NS32016.
> LINUX is tied fairly closely to the 80×86. Not the
> way to go.
I’ll omit a quote that tells what hardware Linux runs on. Seems like one of AST’s arguments has turned against him.
Mr Tanenbaum still deserves credit for engineering an architecture that makes bug-free programming and porting simpler by design. Yet he obviously missed that good engineering is not all you need. From actually looking at the Linux code, I can tell that the code is rather messy, but I can’t deny the fact that it works well and was ported to more platforms than Minix can ever hope to reach.
“Mr Tanenbaum still deserves credit for engineering an architecture that makes bug-free programming and porting simpler by design.”
Actually, it turns out that microkernels are notoriously hard to debug (debugging an async kernel is much harder than a heavily threaded one), and that’s the reason microkernel projects have such slow progress. Even RMS admitted this as the primary reason why Linux prevailed and GNU/Hurd didn’t.
I actually used Minux as my OS for a time being, but the install included GPL software. I didn’t feel like going through hundreds of small programs and risking the chance of leaving some GPL software, so I uninstalled it.
It may be handy for embedded systems, but it doesn’t have to be limited to that.
Just out of curiosity, unless you plan on redistributing said software, why would you care if the software was GPL, BSD, MIT, Apache, CPL, etc. licensed? Those are copyright licenses and only apply to modifications and redistribution of the code and/or binary. As far as actual usage is concerned, they are indistinguishable to the end user.
Minix 3 is also focusing on being a highly reliable, general purpose operating system. Prof. Tanenbaum emphases reliability over other things in his recent publications and this is what he aims to achieve with Minix 3.
Really? It doesn’t boot from CD??
It does boot from a CD normally, unless in this build its broken.
3.1.3 booted & installed fine in VMWare for windows. Perhaps you have a corrupt iso?
Looks very cool.
What i don’t like is always the need for X windows.
Oops! It doesn’t boot from a USB CDRom. Haha, I misread that to mean it doesn’t boot from CDRom at all. Doh!