Minix hardware support? Sparc64, Power8/9, Mips64, ARM64. Competent developers, project led consistently with intention of real world usability?

[tinkg...@gmail.com]

Hi!

Comparing Minix with the BSD:s and Linux, I think Minix' concept of internal segregation and modularization is incredibly compelling, and I would love to actually use it in a production environment e.g. in server, embedded and desktop environments.

Security is only getting more important, and security through segregation is a healthy approach.

For Minix to be interesting to a wider audience, however, it must have excellent platform support, so like many other people this is the first thing I review.

To start with, Sparc64, Power8/9, Mips64, Aarch64 ("ARM64").

This is not a play idea, but it's central in offering something that anyone would actually like to use and invest time in.

AMD64 systems may be cheap, but they are not a panacea platform but come with a wealth of problems, and any credible open source OS will make a primary priority of cultivating platform support.

How has implementing support for these platforms worked out?

When will Sparc64, Power8/9, Mips64 and ARM64 be usable and stable with Minix?

Next, hardware support is paramount too.

The only page with supported hardware I find is http://wiki.minix3.org/doku.php?id=usersguide:hardwarerequirements . I hope this page is an April fool's joke - in total two gigabit ethernet chips supported, and no USB not to mention high-speed USB, are you kidding me?

If I got it right, Minix is deployed on every PC of the planet through Intel's management platform, since 2-3 years.

I presume that one should have some deep hardware integration, if so I presume they keeping a closed tree of drivers for their particular platform.

If the above is not a bad beer joke, then I would get the impression that Minix still is suffering of over-centralization to a minuscule development team

A very small team may be optimal when innovating and designing a concept, but, a mainstream Unix operating system needs a somehow wider community definition, akin to what OpenBSD has, where code review and implementation work is distributed in a community of provenly competent persons.

To make a point rhetorically, European Union grant providers or academic review boards may be impressed with Beaglebone 2 support, but production- and professional users would not even bother to blink for such a proposition.

How much is Minix actually actively cultivated to be usable for non-ultra-specialized applications, in other words as a unix/BSD community member, should I bother?

Tink

[David van Moolenbroek]

Hello,

On Sunday, September 10, 2017 at 9:59:50 AM UTC+2, tinkg...@gmail.com wrote:

How much is Minix actually actively cultivated to be usable for non-ultra-specialized applications, in other words as a unix/BSD community member, should I bother?

With those expectations? No. The MINIX 3 project is run by less than a handful of spare-time volunteers who mostly have (non-MINIX) day jobs. We welcome all contributions, but with very few people interested in contributing at all, the project is currently unable to create an all-round production-ready distribution. Therefore, it is aimed at developers only at this time. We are not short on ideas on how to make MINIX better. We simply lack the manpower.

Regards,
David

[tinkg...@gmail.com]

Hi David,

Thanks for responding.

Minix' present codebase, however, is of a very high quality, right? This should be reflected in Intel's choice of making Minix the most-deployed desktop OS in the world.

And years and years have gone into it, so by comparison, adding more platforms is actually not a major effort, or?

What about asking the government to donate you a couple million EUR to make Minix' platform and hardware support actually useful and adequate?

At some point, open source OS:es fall in the hand of OS coders like you and me, however possibly such an effort could provide a critical push forward?

Tink