strangest systems I've sent email from

Liam Proven lproven at
Thu Apr 28 17:41:08 CDT 2016

On 28 April 2016 at 16:52,  <alexmcwhirter at> wrote:
> On 2016-04-28 10:44, Liam Proven wrote:
>> On 28 April 2016 at 16:35, Mouse <mouse at> wrote:

>>> The depressing (to me) part is that there seems to be a place for
>>> decent-quality restaurants in the same restaurant-food ecosystem that
>>> contains junk-food chains...but there doesn't seem to be the analog in
>>> the computer operating system ecosystem.
>> Absolutely! This is *the* key question, really.
> We get closer to that analog as time passes. The more Linux becomes the next
> Windows, the more people jump ship (mostly to FreeBSD).

That's true, and I am choosing to interpret it as an encouraging sign.
I have nothing against systemd -- I have looked into it and I think I
more or less grasp both the pros and the antis. I have no horse in the
race; whereas I used to play around with init scripts and whatnot,
mostly, that was decades ago now and these days, for my own use, I
favour anything which "just works".

Which is why I'm tying on a (cheap, used) Mac, on which I mostly run
FOSS and freeware.

> Gentoo Linux is my
> distro of choice simply because i can pick, choose, and compile everything i
> want for just about any arch.

I tried it years ago. I found it a very unpleasant experience, and
went to some communities, both online and real-world, to ask for
guidance. One of my questions was "how can I choose a simpler init? I
don't like SysV init, I prefer the BSD one. How do I switch?"

At first they didn't understand the question at all. When I got it
across, the reaction was incredulity: "why on Earth would you want to
do that?!"

I could tweak the compile flags and optimisation of KDE, but not
change the init. That's not customisation in my book: that's
yak-shaving, painting the bike shed a new colour. Basically pointless;
I wanted more profound change, and Gentoo didn't offer it and the
Gentoo community couldn't grasp /why/ I'd want to change something so
profound that it was a given, an axiom, a fundamental.

So I gave up on it before I even had a fully-working system.

> However, if i could choose any OS i would
> probably go with illumos.

Last time I tried, probably one of the last versions of OpenSolaris,
it couldn't understand either my chipset-integrated Ethernet port /or/
the on-motherboard 3rd party Ethernet. So I didn't get very far; it
worked, but I couldn't get online.

>  Unfortunately The man power needed to maintain the
> software repo is the biggest challenge when wanting to go you're own
> direction. I would say that it's the primary reason it seems like we're
> dining in a world of McDonald's and Steak Houses, but nothing in-between.

Well, perhaps. But I'm interested in something *way* more different
than just a different Unix. I like to play with things like Haiku,
AROS, Syllable, Morphos. Interesting, but none are complete enough for
me to work in them, sadly.

I'm intrigued by projects like Movitz, Interim, TempleOS -- the
/really/ different stuff. But all are prototypes, demos, so
dramatically limited they can't be /used/ yet.

I lack the skills to make Plan 9 or Minix 3 do anything useful.

And I read about now-dead historical OSes such as Taos (later Intent),
which I played with all-too-briefly at an Acorn World show about 20y
ago, or Parhelion's HeliOS for Transputer machines. Nothing even
_remotely_ like them exists any more.

>>> I can't help wondering how many people use Linux because "Open Source"
>>> but have never once even tried to build anything from source.
>> Some, certainly.
> Indeed, but more end users help drive demand for better software. Eventually
> someone will get frustrated a build a better tool or improve and existing
> one.

Here's hoping. ISTM that what it's actually leading to is polishing
the same turds. (OK, "turd" is grossly unfair, I actually hugely
admire Linux, but I hope my general meaning is clear.)

We are continuing to refine and tweak a 1970s-style OS -- a
technologically conservative monolithic Unix. FOSS Unix hasn't even
caught up with 1990s style Unix design yet, the early microkernel ones
like NeXTstep or OSF/1, AKA Digital UNIX. It's roughly 2 decades
behind the times.

That's what Minix 3 is trying to fix, of course.

And commercial Unix in general has totally failed to pick up the ideas
of even Plan 9 -- which I think of as Unix Version 2.0: "even more
things are files now, and the kernel is networking aware".

Inferno -- Plan 9 Version 2.0, or perhaps Unix Version 3.0: "Like Plan
9, But Now We Have A Friendlier GUI And Everything's
Processor-Independent", AKA "it's time, let's move on from C" -- is
languishing, AFAICS.

Liam Proven • Profile:
Email: lproven at • GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at • Skype/AIM/Yahoo/LinkedIn: liamproven
Cell/Mobiles: +44 7939-087884 (UK) • +420 702 829 053 (ČR)

More information about the cctalk mailing list