1 Sep
2012
1 Sep
'12
3:04 p.m.
I am sure I've found reference to this in the past, but my Google-fu
is failing me.
IIRC in the fairly early stages of the GNU OS project, they looked at
and considered using the kernel from BSD - 4.3 Net/1 or thereabouts, I
think. I'm looking for more detail or a web reference.
I've even found mention of them considering and discarding the UZI
kernel, which is what Allison was mentioning in the thread about
CP/M-compatible OSs the other day.
Could anyone give me a pointer, at all?
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
1 Sep
1 Sep
4:13 p.m.
On 9/1/12 12:04 PM, Liam Proven wrote:
IIRC in the fairly early stages of the GNU OS project,
they looked at
and considered using the kernel from BSD - 4.3 Net/1 or thereabouts, I
think. I'm looking for more detail or a web reference.
In the very earliest days, they were talking about using Trix, which was
developed at MIT.
http://www.gnu.org/software/hurd/users-guide/using_gnuhurd.html
1.3 History
Richard Stallman (RMS) started GNUin 1983, as a project to create a complete free
operating system. In the text of the GNU Manifesto, he mentioned that there is a primitive
kernel. In the first
GNUsletter, Feb. 1986, he says that GNU's kernel is TRIX, which was developed at the
Massachusetts Institute of Technology.
By December of 1986, the Free Software Foundation (FSF) had ?started working on the
changes needed to TRIX? [Gnusletter, Jan. 1987]. Shortly thereafter, the FSF began
?negotiating with Professor
Rashid of Carnegie-Mellon University about working with them on the development of the
Mach kernel? [Gnusletter, June, 1987]. The text implies that the FSF wanted to use someone
else's work, rather
than have to fix TRIX.
In [Gnusletter, Feb. 1988], RMS was talking about taking Mach and putting the Berkeley
Sprite filesystem on top of it, ?after the parts of Berkeley Unix... have been replaced.?
Six months later, the FSF is saying that ?if we can't get Mach, we'll use TRIX or
Berkeley's Sprite.? Here, they present Sprite as a full-kernel option, rather than
just a filesystem.
In January, 1990, they say ?we aren't doing any kernel work. It does not make sense
for us to start a kernel project now, when we still hope to use Mach? [Gnusletter, Jan.
1990]. Nothing significant
occurs until 1991, when a more detailed plan is announced:
6:12 p.m.
On 1 September 2012 21:13, Al Kossow <aek at bitsavers.org> wrote:
On 9/1/12 12:04 PM, Liam Proven wrote:
Interesting stuff! Many thanks!
In the end, after /much/ more digging, I found this:
http://www.groklaw.net/article.php?story=20050727225542530
So candidates were TRIX, Sprite, BSD, UZI, then they tried to build
HURD and in the end it became Linux.
There's a line about tangled webs that springs to mind.
I can't help but wonder what would have happened had they gone with BSD.
I think that memoir might be a bit out though. It mentions BSD
4.4-Lite. That was 1994, I think, by which point Linux was up to
version 1.0 after 3y of hard work. Too late; doesn't fit the story.
Looking at the timings, I think Net/1 seems more likely - that was
4.3BSD, in 1989, the first unencumbered release and it fits the
chronology better.
If the GNU project had discarded most of Net/1 except the kernel, all
the userland and so on, there could have been a working free Unix by
about 1990. Linux would never have happened - Linux has said as much -
and maybe the Free/Net/Open BSD splintering wouldn't, either.
If there had been a viable free Unix out the door in the era of
Windows 3.1, before NT or 95, the computing world would have been
/very/ different, I think.
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
IIRC in the fairly early stages of the GNU OS
project, they looked at
and considered using the kernel from BSD - 4.3 Net/1 or thereabouts, I
think. I'm looking for more detail or a web reference.
In the very earliest days, they were talking about using Trix, which was
developed at MIT.
http://www.gnu.org/software/hurd/users-guide/using_gnuhurd.html
1.3 History
Richard Stallman (RMS) started GNUin 1983, as a project to create a complete
free operating system. In the text of the GNU Manifesto, he mentioned that
there is a primitive kernel. In the first GNUsletter, Feb. 1986, he says
that GNU's kernel is TRIX, which was developed at the Massachusetts
Institute of Technology.
By December of 1986, the Free Software Foundation (FSF) had ?started working
on the changes needed to TRIX? [Gnusletter, Jan. 1987]. Shortly thereafter,
the FSF began ?negotiating with Professor Rashid of Carnegie-Mellon
University about working with them on the development of the Mach kernel?
[Gnusletter, June, 1987]. The text implies that the FSF wanted to use
someone else's work, rather than have to fix TRIX.
In [Gnusletter, Feb. 1988], RMS was talking about taking Mach and putting
the Berkeley Sprite filesystem on top of it, ?after the parts of Berkeley
Unix... have been replaced.?
Six months later, the FSF is saying that ?if we can't get Mach, we'll use
TRIX or Berkeley's Sprite.? Here, they present Sprite as a full-kernel
option, rather than just a filesystem.
In January, 1990, they say ?we aren't doing any kernel work. It does not
make sense for us to start a kernel project now, when we still hope to use
Mach? [Gnusletter, Jan. 1990]. Nothing significant occurs until 1991, when a
more detailed plan is announced: 
7:10 p.m.
On 09/01/2012 06:12 PM, Liam Proven wrote:
So candidates were TRIX, Sprite, BSD, UZI, then they
tried to build
HURD and in the end it became Linux.
Well, to the very end of that sentence, I say "yes and no". The
Linux crowd used a lot of GNU userland stuff from the very beginning,
with little or no involvement from the GNU project, and certainly no
*official* involvement. When RMS realized (well, more like "admitted to
himself") that the Hurd was essentially stillborn, they pretty much laid
claim to Linux. As far as I saw, they didn't go so far as revisionist
historianism, but they did try VERY hard to get everyone to call it
"GNU/Linux". That riding-on-coattails thing is something they should've
been pretty embarrassed about, but they certainly didn't seem to be at
the time. RMS was not at all bashful about discussing Linux as
"GNU/Linux" at every turn, simply because they adopted the GNU userland
utilities...as did several other OSs.
Windows, for example, doesn't somehow become "GNU/Windows" when one
installs Cygwin. ;)
The GNU OS project went on and on for a long time, as the dream of
several people, so I think it was very tough for them to just throw in
the towel and be content with the fact that their *utilities* were what
the world depended on, and they didn't build the entire OS.
Of course this is all just useless "what the hell would I know, I was
only THERE" type of information, but I could put it on a webpage if you
like. ;) All the happenings from that era, which wasn't really very
long ago at all, well less than a decade, and all of the above is pretty
generally known.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
7:16 p.m.
On 9/1/2012 7:10 PM, Dave McGuire wrote:
The GNU OS project went on and on for a long time, as
the dream of
several people, so I think it was very tough for them to just throw in
the towel and be content with the fact that their *utilities* were
what the world depended on, and they didn't build the entire OS.
I remember bootstrapping the GNU/HURD circa 1998 for grins. It was a
real PITA, though. It may be worth trying to run as a VM these days just
to play with.
7:27 p.m.
I tried it last about a year or two ago. I didn't manage to get it working. It's
still very rough and needs a lot more work.
--------
Paul Anderson -- VE3HOP
On 2012-09-01, at 7:16 PM, Jonathan Katz <jon at jonworld.com> wrote:
On 9/1/2012 7:10 PM, Dave McGuire wrote:
The GNU OS project went on and on for a long
time, as the dream of several people, so I think it was very tough for them to just throw
in the towel and be content with the fact that their *utilities* were what the world
depended on, and they didn't build the entire OS.
I remember bootstrapping the GNU/HURD circa 1998 for grins. It was a real PITA, though.
It may be worth trying to run as a VM these days just to play with. 
7:39 p.m.
Well, to the very end of that sentence, I say
"yes and no". The
Linux crowd used a lot of GNU userland stuff from the very beginning,
with little or no involvement from the GNU project, and certainly no
*official* involvement. When RMS realized (well, more like "admitted to
himself") that the Hurd was essentially stillborn, they pretty much laid
claim to Linux. As far as I saw, they didn't go so far as revisionist
historianism, but they did try VERY hard to get everyone to call it
"GNU/Linux". That riding-on-coattails thing is something they should've
been pretty embarrassed about, but they certainly didn't seem to be at
the time. RMS was not at all bashful about discussing Linux as
"GNU/Linux" at every turn, simply because they adopted the GNU userland
utilities...as did several other OSs.
These are the kinds of actions that make people shake their head, but
that's probably exactly what RMS wants to have happen.
I think I would probably strangle him if I had to be in the same room with
him for any length of time. But I respect that he is consistent in his
views and not hypocritical about implementing them, even if that makes him
unpopular (and even if his views are batsh*t crazy sometimes).
--
------------------------------------ personal: http://www.cameronkaiser.com/ --
Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com
-- God made the integers; all else is the work of Man. -- Kronecker -----------
7:46 p.m.
On 9/1/2012 4:39 PM, Cameron Kaiser wrote:
I got a pretty good chuckle out of this, RMS's "tour rider." It starts
off fairly sane and then gets downright bizarre. "If you can find a
host for me that has a friendly parrot, I will be very very glad." (but
DO NOT give him a parrot.)
https://secure.mysociety.org/admin/lists/pipermail/developers-public/2011-O…
- Josh
Well, to the
very end of that sentence, I say "yes and no". The
Linux crowd used a lot of GNU userland stuff from the very beginning,
with little or no involvement from the GNU project, and certainly no
*official* involvement. When RMS realized (well, more like "admitted to
himself") that the Hurd was essentially stillborn, they pretty much laid
claim to Linux. As far as I saw, they didn't go so far as revisionist
historianism, but they did try VERY hard to get everyone to call it
"GNU/Linux". That riding-on-coattails thing is something they should've
been pretty embarrassed about, but they certainly didn't seem to be at
the time. RMS was not at all bashful about discussing Linux as
"GNU/Linux" at every turn, simply because they adopted the GNU userland
utilities...as did several other OSs.
These are the kinds of actions that make
people shake their head, but
that's probably exactly what RMS wants to have happen.
I think I would probably strangle him if I had to be in the same room with
him for any length of time. But I respect that he is consistent in his
views and not hypocritical about implementing them, even if that makes him
unpopular (and even if his views are batsh*t crazy sometimes).
8:03 p.m.
I think I
would probably strangle him if I had to be in the same room with
him for any length of time.
You don't want to be in the same room with him for other reasons.
2 Sep
2 Sep
5:14 a.m.
On 09/01/2012 08:03 PM, Cameron Kaiser wrote:
I had heard he had "personal issues" but I'm not sure what to say about
those.
It's very common in those circles. (the MIT super-genius crowd) I
don't know what the deal is, but it can get downright powerful.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
I think I would probably strangle him if I had to be
in the same room with
him for any length of time.
You don't want to be in the same room with him for other reasons. 
4 Sep
4 Sep
5:06 p.m.
On Sat, Sep 01, 2012 at 04:47:29PM -0700, Al Kossow wrote:
On 9/1/12 4:39 PM, Cameron Kaiser wrote:
Especially if you are female (as reported to me by a reliable first hand
source).
Kind regards,
Alex.
--
"Opportunity is missed by most people because it is dressed in overalls and
looks like work." -- Thomas A. Edison
I think I would probably strangle him if I had to
be in the same room with
him for any length of time.
You don't want to be in the same room with him for other reasons.
1 Sep
1 Sep
8:41 p.m.
On 2 September 2012 00:39, Cameron Kaiser <spectre at floodgap.com> wrote:
These are the kinds of actions that make people shake their head, but
that's probably exactly what RMS wants to have happen.
I think I would probably strangle him if I had to be in the same room with
him for any length of time. But I respect that he is consistent in his
views and not hypocritical about implementing them, even if that makes him
unpopular (and even if his views are batsh*t crazy sometimes).
I have heard that he is an intolerable PITA in many ways, personally
and professionally, but he is possibly an actual genius, certainly
remarkably far-sighted and clear-sighted (at least some of the time),
and at the end of the day, without him, we wouldn't have Linux and the
computer world would be a much poorer place.
GNU/Linux is a truly transformative piece of technology, and I think
its greatest potential still lies ahead of it. I suspect it will be in
the time when the 5 billion or so people of the developing world come
online & get access to the Internet and the Web. That is going to
truly change the world, and I don't believe any proprietary tech is
going to do it.
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
8:38 p.m.
On 2 September 2012 00:10, Dave McGuire <mcguire at neurotica.com> wrote:
On 09/01/2012 06:12 PM, Liam Proven wrote:
[Nod]
So candidates were TRIX, Sprite, BSD, UZI, then they tried to build
HURD and in the end it became Linux.
Well, to the very end of that sentence, I say "yes and no". The Linux
crowd used a lot of GNU userland stuff from the very beginning, with little
or no involvement from the GNU project, and certainly no *official*
involvement. When RMS realized (well, more like "admitted to
himself") that
the Hurd was essentially stillborn, they pretty much laid claim to Linux.
As far as I saw, they didn't go so far as revisionist historianism, but they
did try VERY hard to get everyone to call it "GNU/Linux".
Sounds about right. I was following with mild interest until the Linux
kernel passed version 1.0, at which point I started paying attention,
attempting to get the thing to install, etc.
That
riding-on-coattails thing is something they should've been pretty
embarrassed about, but they certainly didn't seem to be at the time. RMS
was not at all bashful about discussing Linux as "GNU/Linux" at every turn,
simply because they adopted the GNU userland utilities...as did several
other OSs.
I'd agree with that, but as RMS sees it (I reckon), it was built using
his tools to work with his userland; it would be nothing without the
GNU Project, and thus it's de facto /part/ of the GNU Project.
Windows, for example, doesn't somehow become
"GNU/Windows" when one
installs Cygwin. ;)
Ha! Good point!
The GNU OS project went on and on for a long time,
as the dream of several
people, so I think it was very tough for them to just throw in the towel and
be content with the fact that their *utilities* were what the world depended
on, and they didn't build the entire OS.
Well, in the longer run, I think Andy Tanenbaum was right. The
monolithic-kernel ideal was past its time in 1991 and it's
ridiculously so 20+ yr later.
Even now it's based on L4 and not Mach, I think the microkernel idea
has merit. There are others out there - Minix 3 springs to mind.
I suspect that eventually the Linux project will just run out of
steam, falter and stagnate, and maybe, just maybe, by then, the
distributed/loosely-coupled-array-of-servers type of OS will rise up
from its ancient slumber and take over the world.
I have also read that the design of Plan 9 makes the whole monolithic
kernel/microkernel debate moot, but I don't know enough about Plan 9
to judge & anyone who does is unable to talk down to my level to
explain it to me. ;?)
Of course this is all just useless "what the
hell would I know, I was only
THERE" type of information, but I could put it on a webpage if you like. ;)
All the happenings from that era, which wasn't really very long ago at all,
well less than a decade, and all of the above is pretty generally known.
The real mostly-forgotten story, as I remember it, is here:
http://www.h-online.com/open/features/GNU-HURD-Altered-visions-and-lost-pro…
... as sourced from here:
http://www.groklaw.net/article.php?story=20050727225542530
That's one of the biggest missed opportunities in the history of
computing right there.
It took 4 or 5 /years/ from 4.3BSD Net/1 to BSD 4.4-Lite, when the
real work could really begin. By then, Linux + GNU had stolen the
thunder.
If GNU had adopted the BSD kernel in the late '80s, then the whole
AT&T code issue would have gone away. (AIUI.) There might not /be/ a
BSD family of Unixes (which would be a shame) but the Free Unix world
would have got bootstrapped before Microsoft got in there.
For me, the greatest missed opportunities in the last 30y of computing
were (in approximate chronological order):
* Sinclair's Spectrum 128 *not* being based on the Timex TS2068 improved ULA
(Sinclair was focussing on the abortive QL & missed the chance to keep
the Spectrum on top of the 8-bit heap)
* Apple castrating DR GEM on x86, leading to the split from the 680x0 version
(GEM as a multitasking GUI on top of a multitasking CDOS386-based OS?
Yes please!)
* OS/2 1.x not being 386-specific, in which case, it coulda been a contender
(Allegedly IBM wanting to support its 286 customers, but visibly
stupid /even at the time/)
* GNU not adopting the BSD kernel in 1989-1990
(Stalling the development of a Free Unix for 4-5 vital years)
* Quarterdeck not getting Desqview/X out until after Windows 3.1
(DVX was fantastic, but too late)
* Apple rushing the Newton out before it was ready
(Abandoning Dylan, which could have been a truly transformative
language, I still feel)
* Apple not spotting that the answer to Pink, Copland etc. was A/UX on PowerPC
(But that would have killed the NeXT deal, which saved the company, so... moot)
* Be and BeOS getting shafted by MICROS~1
(Big players like Hitachi were bundling it but couldn't dual-boot
because of MS' OEM contract on MS-DOS & Windows)
Anyone got any other suggestions for the list? :?)
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
9:39 p.m.
On 9/1/12 5:12 PM, Liam Proven wrote:
So candidates were TRIX, Sprite, BSD, UZI, then they
tried to build
HURD and in the end it became Linux.
No, not at all. RMS fiercely fought (may still do) Linux's
"takeover" of the GNU environment.
The whole "GNU/Linux" thing wasn't an effort to coopt Linux as part
of GNU, but rather RMS's attempt to force Linux users and vendors to
acknowledge GNU as the OS and Linux as "just the kernel".
I was involved in the local Linux Users' Group when all that fight
was fresh. The LUG invited RMS to speak here, and he accepted the
invitation - with the minor condition that we must change the name of
the group from AustinLUG (Austin Linux Users' Group) to AustinG/LUG
(Austin GNU/Linux Users' Group).
I really don't think he even saw the humor in the suggestion....
Doc
10:50 p.m.
On 2 September 2012 02:39, Doc <doc at vaxen.net> wrote:
On 9/1/12 5:12 PM, Liam Proven wrote:
How so?
He wants credit for GNU - hell, he wants joint billing, essentially.
But he has got a point. No GNU, no Linux; but no Linux, still a GNU,
someway, somehow.
X.org isn't a Linux thing any more, either - it also runs on the BSDs
and Solaris and probably other Unices too.
So candidates were TRIX, Sprite, BSD, UZI, then
they tried to build
HURD and in the end it became Linux.
No, not at all. RMS fiercely fought (may still do) Linux's "takeover"
of
the GNU environment. The whole "GNU/Linux" thing wasn't an
effort to coopt Linux as part of
GNU, but rather RMS's attempt to force Linux users and vendors to
acknowledge GNU as the OS and Linux as "just the kernel".
Again, he has got a point. I don't agree with him, but his arguments
are not entirely without merit.
I was involved in the local Linux Users' Group
when all that fight was
fresh. The LUG invited RMS to speak here, and he accepted the invitation -
with the minor condition that we must change the name of the group from
AustinLUG (Austin Linux Users' Group) to AustinG/LUG (Austin GNU/Linux
Users' Group).
I really don't think he even saw the humor in the suggestion....
Not famed for his sense of humour.
But the thing is, the threat we all face to our freedoms from the big
proprietary vendors is immense. If MICROS~1 and Apple get their way,
we won't own our computers or the content on them - they will and they
will have total control at all times, from bits on disk to display on
screen. We will only be allowed to do what they permit us to do, and
anything we thing we own or create, they can keep, take away from us,
use as they wish and there won't be a damned thing we can do about it.
And there is, at the end of the day, only one man who warned about
this and who led a project for many decades to ensure that there /is/
an alternative. That man is not Eric Raymond, Linus Torvalds or Mark
Shuttleworth. It's Richard Stallman. The one, the original, and
however rude, inconsiderate, hostile, intransigent, uncooperative or
anything else he is, he has never wavered and remained true all the
way.
If in another decade people still have control of their own computers
and can do whatever they want with them and the data on them, it will
be because of RMS and nobody else.
And frankly, for all that he can be ornery or whatever, at the end of
the day, the world could not thank him enough for that. People have
won Nobel prizes for a lot less.
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
2 Sep
2 Sep
9:59 a.m.
On 9/1/12 9:50 PM, Liam Proven wrote:
X.org isn't a Linux thing any more, either - it
also runs on the BSDs
and Solaris and probably other Unices too.
X.org, nee XFree86, never was a Linux thing, any more than GNU. Like
the GNU environment, it started as a cross-platform Free implementation
of commercial unix tools. As far as I know or can remember, neither
X.org or XFree86 ever existed as a Linux-only implementation.
If in another decade people still have control of
their own computers
and can do whatever they want with them and the data on them, it will
be because of RMS and nobody else.
And frankly, for all that he can be ornery or whatever, at the end of
the day, the world could not thank him enough for that. People have
won Nobel prizes for a lot less.
I won't argue that at all. As a matter of fact, I may have *said*
all that a time or 20...
Doc
11 a.m.
On 2 September 2012 14:59, Doc <doc at vaxen.net> wrote:
Fair enough.
Your comment that ?RMS fiercely fought (may still do) Linux's
"takeover" of the GNU environment? seemed harsh to me, but looking
back, I think you were misinterpreting /me/ and thus /in context/ it
was not so harsh.
When I said "Linux 'took over' GNU", I don't mean that the former
own
or dominates the latter or anything like that. All I'm getting at is
that it's the default pairing. The BSDs have their own userlands to go
with their own kernels (notwithstanding Debian k/FreeBSD et al); the
open-Solaris-distros such as Illumos, Belenix, Schillix etc. I believe
still use a Sun libc and graft a GNU userland /beside/ rather than
/instead of/ the Oracle one.
The various incarnations of the HURD are very minority offerings
indeed and most of the other Unix-like kernels have their own
userland. You pretty much only get the GNU userland on Linux and you
pretty much only get Linux underneath the GNU userland.
That is all I meant when I said "takeover".
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
On 9/1/12 9:50 PM, Liam Proven wrote:
I'll take your word. I am fairly ignorant of the BSD world - the only
BSD I've ever managed to install and get to boot up to a GUI was
PC-BSD, which is kinda cheating.
I was under the impression that for many years, in the XFree86 era,
the BSDs maintained their own X.11 servers, and Solaris still does,
the last time I looked - it offers the Sun X server for Sun graphics
hardware and X.org for non-Sun kit.
I have no experience with AIX on x86, only on RIOS/ROMP, but AFAIAA,
all the commercial Unices still maintain their own X servers, or did
until they died.
Was XFree86 /never/ just a Linux-on-x86 project?
X.org isn't a Linux thing any more, either -
it also runs on the BSDs
and Solaris and probably other Unices too.
X.org, nee XFree86, never was a Linux thing, any more than GNU. Like the
GNU environment, it started as a cross-platform Free implementation of
commercial unix tools. As far as I know or can remember, neither X.org or
XFree86 ever existed as a Linux-only implementation. If in another
decade people still have control of their own computers
and can do whatever they want with them and the data on them, it will
be because of RMS and nobody else.
And frankly, for all that he can be ornery or whatever, at the end of
the day, the world could not thank him enough for that. People have
won Nobel prizes for a lot less.
I won't argue that at all. As a matter of fact, I may have *said* all
that a time or 20...
5:31 p.m.
New subject: X and GNU software, was Re: GNU and the BSD kernel
On 09/02/2012 11:00 AM, Liam Proven wrote:
X wasn't a "BSD world" thing either. It wasn't a "free
UNIX" thing at
all.
First and foremost, X is a protocol, not a piece of software. The
*reference implementation* of that protocol, and that's ALL it is, is
available for free, and it compiles and runs on many different
platforms, notably Sun and DEC workstations.
XFree86 came along to support more of the whiz-bang graphics hardware
in the PC world, all of which tended to be very inconsistent and fiddly
even back then. Who in their right mind changes friggin' sync rates in
order to switch in and out of graphics mode? PC video cards, that's
who. Madness. Anyway, XFree86 was put together to address all that
crap, primarily under Linux.
Eventually, most of the free UNIX implementations started to narrow
their focus to x86, even the still-very-cross-platform ones like NetBSD.
They switched most of their ports over to XFree86 some time ago, when
it became flexible enough to support other hardware, and that's where
most of the development was focused. This wasn't all that long ago.
You'll see these two implementations of the X protocol referred to as
"MIT X11" and "XFree". It's important to note that
"XFree" is no more
"free" than MIT's freely-available X server; that's just what they
chose
to name it.
X.org, nee
XFree86, never was a Linux thing, any more than GNU. Like the
GNU environment, it started as a cross-platform Free implementation of
commercial unix tools. As far as I know or can remember, neither X.org or
XFree86 ever existed as a Linux-only implementation.
I'll take your word. I am fairly ignorant of the BSD world - the only
BSD I've ever managed to install and get to boot up to a GUI was
PC-BSD, which is kinda cheating.
I was under the impression that for many years, in the XFree86 era,
the BSDs maintained their own X.11 servers, and Solaris still does,
the last time I looked - it offers the Sun X server for Sun graphics
hardware and X.org for non-Sun kit.
I have no experience with AIX on x86, only on RIOS/ROMP, but AFAIAA,
all the commercial Unices still maintain their own X servers, or did
until they died.
Was XFree86 /never/ just a Linux-on-x86 project? The various incarnations of the HURD are very minority
offerings
indeed and most of the other Unix-like kernels have their own
userland. You pretty much only get the GNU userland on Linux and you
pretty much only get Linux underneath the GNU userland.
Well not quite. Most commercial UNIX sysadmins, myself included, had
their first introduction to GNU software as improved replacements for
vendor-supplied utilities. This started happening LONG before there
were any viable (or even not-so-viable!) free UNIX implementations.
TO THIS DAY the first thing I do on a brand-new Solaris system is
replace a few Oracle-supplied utilities like "tar" with their GNU
replacements, to bring their functionality into this decade.
Most highly technical organizations (or organizations with competent
technical staffs) had their own "standard OS load" for a given platform,
that consisted of a commercial vendor-supplied UNIX (SunOS, Ultrix, etc)
with all the bad stuff fixed. Fixing stuff usually meant replacing
"tar", "awk", etc with stuff that had actually progressed since the
1970s...and it was almost always GNU. Further, you had source code, so
you could fix any bugs you ran into without having to deal with talking
to incompetent vendor employees and MAYBE getting a bug fix in a few months.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
6:05 p.m.
New subject: X and GNU software, was Re: GNU and the BSD kernel
First and foremost, X is a protocol, not a piece of
software. The
*reference implementation* of that protocol, and that's ALL it is, is
available for free, and it compiles and runs on many different
platforms, notably Sun and DEC workstations.
I think it's a sample implementation, not a reference implementation.
It certainly used to be. (This is an important difference. With a
reference implementation, the implementation _is_ the spec, by
definition. With a sample implementation, the spec is something else.)
Eventually, most of the free UNIX implementations
started to narrow
their focus to x86, even the still-very-cross-platform ones like
NetBSD. They switched most of their ports over to XFree86 some time
ago, when it became flexible enough to support other hardware, and
that's where most of the development was focused. This wasn't all
that long ago.
This brought its own forms of madness, like X servers that do their own
bus enumeration, even in cases like NetBSD/sparc where there is
absolutely no excuse for it...meaning, eg, that they have to run as
root even when the user starting them owns the framebuffer, keyboard,
and mouse devices.
There's a reason I use the X11R6.4 sample server on machines where I'm
not saddled with peecee-style video-card hell.
And then Xorg came along to muddy the waters further....
> You pretty much only get the GNU userland on Linux
and you pretty
> much only get Linux underneath the GNU userland.
To the extent that this is true (and, as Dave points out, "the GNU
userland" is not an indivisible blob of everything), it's only beacuse
that's what people care to bother doing. A Linux friend of mine tells
me there is, or at least was, a Linux distro that had an almost
entirely BSD userland on top of the Linux kernel - about as much GNU
code in userland as NetBSD, say, has. I've also heard of Linux
userlands[%] running under BSD kernels.)
[%] To the extent that such a thing exists - and, no, I do not mean
"GNU userlands". I mean, approximately, "a userland a Linux distro
might/could/does come with", which typically includes a good deal
of non-GNU code.
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
3 Sep
3 Sep
9:59 a.m.
New subject: X and GNU software, was Re: GNU and the BSD kernel
On 2 September 2012 23:05, Mouse <mouse at rodents-montreal.org> wrote:
I must admit, I thought that was standard.
But my real-Unix experience is minimal. Basically I am a Linux guy who
has also dabbled a bit with AIX and SunOS/Solaris, who happened to
start out on Xenix.
I know of 2:
http://www.debian.org/ports/kfreebsd-gnu/
(Works, not very well, is current)
http://www.debian.org/ports/netbsd/
(Abandoned prototype)
There's also Gentoo/BSD but that uses the BSD userland & just brings
Gentoo package management & build system to BSD. Given that Gentoo's
"portage" was inspired by BSD "ports", this seems oddly circular; if
Portage is so good, why not just merge or update Ports with it or its
ideas or implementation?
(I have played with both Ports and Portage. I much prefer Apt.)
First and
foremost, X is a protocol, not a piece of software. The
*reference implementation* of that protocol, and that's ALL it is, is
available for free, and it compiles and runs on many different
platforms, notably Sun and DEC workstations.
I think it's a sample implementation, not a reference implementation.
It certainly used to be. (This is an important difference. With a
reference implementation, the implementation _is_ the spec, by
definition. With a sample implementation, the spec is something else.)
Eventually, most of the free UNIX implementations
started to narrow
their focus to x86, even the still-very-cross-platform ones like
NetBSD. They switched most of their ports over to XFree86 some time
ago, when it became flexible enough to support other hardware, and
that's where most of the development was focused. This wasn't all
that long ago.
This brought its own forms of madness, like X servers that do their own
bus enumeration, even in cases like NetBSD/sparc where there is
absolutely no excuse for it...meaning, eg, that they have to run as
root even when the user starting them owns the framebuffer, keyboard,
and mouse devices. There's a reason I use the X11R6.4 sample server
on machines where I'm
not saddled with peecee-style video-card hell.
And then Xorg came along to muddy the waters further....
Is this un-disentagleable? Because that's the precise bit I sought
clarification on.
> You pretty
much only get the GNU userland on Linux and you pretty
> much only get Linux underneath the GNU userland.
To the extent that this is true (and, as Dave points out, "the GNU
userland" is not an indivisible blob of everything), it's only beacuse
that's what people care to bother doing. A Linux friend of mine tells
me there is, or at least was, a Linux distro that had an almost
entirely BSD userland on top of the Linux kernel - about as much GNU
code in userland as NetBSD, say, has. I've also heard of Linux
userlands[%] running under BSD kernels.)
[%] To the extent that such a thing exists - and, no, I do not mean
"GNU userlands". I mean, approximately, "a userland a Linux distro
might/could/does come with", which typically includes a good deal
of non-GNU code.
Understood, and an interesting idea. Not encountered this. Any
references much welcomed!
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
2 Sep
2 Sep
6:10 p.m.
New subject: Vendor stuff was crummy - Re: X and GNU software, was Re: GNU and the BSD kernel
On 02/09/12 5:31 PM, Dave McGuire wrote:
...
Most highly technical organizations (or organizations with competent
technical staffs) had their own "standard OS load" for a given platform,
that consisted of a commercial vendor-supplied UNIX (SunOS, Ultrix, etc)
with all the bad stuff fixed. Fixing stuff usually meant replacing
"tar", "awk", etc with stuff that had actually progressed since the
1970s...and it was almost always GNU. Further, you had source code, so
you could fix any bugs you ran into without having to deal with talking
to incompetent vendor employees and MAYBE getting a bug fix in a few months.
Tell the kids these days - they won't believe yer.
Seriously: the great Reddit horde has no idea how bad things were before
GNU shook it up. It was hard to dream in 1985 that gcc would supplant
just about every vendor compiler, and when RMS said he hoped this would
happen, many said he was nuts (we have Usenet archives to prove it).
But it did eventually happen. :)
--Toby
-Dave
6:24 p.m.
New subject: Vendor stuff was crummy - Re: X and GNU software, was Re: GNU and the BSD kernel
On Sep 2, 2012, at 3:10 PM, Toby Thain wrote:
On 02/09/12 5:31 PM, Dave McGuire wrote:
And it's a sorry looking code base. That's why clang started up?to replace the
mess that gcc had become. Free stuff is great, but the problem was no one was willing (or
able) to go back and clean it up. All any of the vendors (who were the ones adding
various CPU support into gcc) wanted to do was to make the smallest changes necessary to
get their CPU supported. The internals to gcc have needed replacing for years.
Unfortunately the internals are poorly documented and it's damn near impossible to
replace the pieces of it and still have the various front ends or code gens still work.
So folks started over from scratch. That's where clang came from (but admittedly to
solve some other issues with gcc).
TTFN - Guy
...
Most highly technical organizations (or organizations with competent
technical staffs) had their own "standard OS load" for a given platform,
that consisted of a commercial vendor-supplied UNIX (SunOS, Ultrix, etc)
with all the bad stuff fixed. Fixing stuff usually meant replacing
"tar", "awk", etc with stuff that had actually progressed since the
1970s...and it was almost always GNU. Further, you had source code, so
you could fix any bugs you ran into without having to deal with talking
to incompetent vendor employees and MAYBE getting a bug fix in a few months.
Tell the kids these days - they won't believe yer.
Seriously: the great Reddit horde has no idea how bad things were before GNU shook it up.
It was hard to dream in 1985 that gcc would supplant just about every vendor compiler, and
when RMS said he hoped this would happen, many said he was nuts (we have Usenet archives
to prove it).
But it did eventually happen. :)
7:03 p.m.
New subject: gcc EOL? - Re: Vendor stuff was crummy - Re: X and GNU software, was Re: GNU and the BSD kernel
On 02/09/12 6:24 PM, Guy Sotomayor wrote:
On Sep 2, 2012, at 3:10 PM, Toby Thain wrote:
Not disputing any of this -- it's a natural cycle.
--Toby
On 02/09/12 5:31 PM, Dave McGuire wrote:
And it's a sorry looking code base. That's why clang started up?to replace the
mess that gcc had become. Free stuff is great, but the problem was no one was willing (or
able) to go back and clean it up. All any of the vendors (who were the ones adding
various CPU support into gcc) wanted to do was to make the smallest changes necessary to
get their CPU supported. The internals to gcc have needed replacing for years.
Unfortunately the internals are poorly documented and it's damn near impossible to
replace the pieces of it and still have the various front ends or code gens still work.
So folks started over from scratch. That's where clang came from (but admittedly to
solve some other issues with gcc).
...
Most highly technical organizations (or organizations with competent
technical staffs) had their own "standard OS load" for a given platform,
that consisted of a commercial vendor-supplied UNIX (SunOS, Ultrix, etc)
with all the bad stuff fixed. Fixing stuff usually meant replacing
"tar", "awk", etc with stuff that had actually progressed since the
1970s...and it was almost always GNU. Further, you had source code, so
you could fix any bugs you ran into without having to deal with talking
to incompetent vendor employees and MAYBE getting a bug fix in a few months.
Tell the kids these days - they won't believe yer.
Seriously: the great Reddit horde has no idea how bad things were before GNU shook it up.
It was hard to dream in 1985 that gcc would supplant just about every vendor compiler, and
when RMS said he hoped this would happen, many said he was nuts (we have Usenet archives
to prove it).
But it did eventually happen. :) TTFN - Guy
6:29 p.m.
New subject: Vendor stuff was crummy - Re: X and GNU software, was Re: GNU and the BSD kernel
Seriously: the great Reddit horde has no idea how bad
things were
before GNU shook it up.
The great Reddit horde has no idea how bad things are now, either.
Bad in completely different ways, mind you. But anyone who lived
through the net with Jon Postel at the top of its governance pyramid
will have some idea. (They won't all agree with me that today is
worse. But I think they'll all agree that today is drastically
different and that many aspects of today's setup are bad.)
It was hard to dream in 1985 that gcc would supplant
just about every
vendor compiler, and when RMS said he hoped this would happen, many
said he was nuts (we have Usenet archives to prove it).
I'm not sure he wasn't nuts.
But it did eventually happen. :)
It did. And now we're scarily close to having a compiler monoculture.
I don't consider that cure worse than the disease...but I do consider
it relatively close. I dislike - and deeply distrust - monocultures of
any sort.
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
3 Sep
3 Sep
3:03 a.m.
New subject: Vendor stuff was crummy - Re: X and GNU software, was Re: GNU and the BSD kernel
On 09/02/2012 06:10 PM, Toby Thain wrote:
Yeah, I'm not too worried about it. I'm the guy who gets paid to
clean up the messes that those kids make, so I laugh all the way to the
bank.
Most highly
technical organizations (or organizations with competent
technical staffs) had their own "standard OS load" for a given platform,
that consisted of a commercial vendor-supplied UNIX (SunOS, Ultrix, etc)
with all the bad stuff fixed. Fixing stuff usually meant replacing
"tar", "awk", etc with stuff that had actually progressed since the
1970s...and it was almost always GNU. Further, you had source code, so
you could fix any bugs you ran into without having to deal with talking
to incompetent vendor employees and MAYBE getting a bug fix in a few
months.
Tell the kids these days - they won't believe yer. Seriously: the great Reddit horde has no idea how bad
things were before
GNU shook it up. It was hard to dream in 1985 that gcc would supplant
just about every vendor compiler, and when RMS said he hoped this would
happen, many said he was nuts (we have Usenet archives to prove it).
But it did eventually happen. :)
Yes it did, with very few (and only very specialized) exceptions.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
9:49 a.m.
New subject: X and GNU software, was Re: GNU and the BSD kernel
On 2 September 2012 22:31, Dave McGuire <mcguire at neurotica.com> wrote:
Interesting. I did not know this.
Thanks!
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
On 09/02/2012 11:00 AM, Liam Proven wrote:
X wasn't a "BSD world" thing either. It wasn't a "free
UNIX" thing at
all.
First and foremost, X is a protocol, not a piece of software. The
*reference implementation* of that protocol, and that's ALL it is, is
available for free, and it compiles and runs on many different
platforms, notably Sun and DEC workstations.
Yes, I'm aware of that much. I wasn't talking about the notional
nonexistent bit of software that is "X-windows" (yes, I did that
deliberately), but about what used to be XFree86 and is now X.org.
X.org,
nee XFree86, never was a Linux thing, any more than GNU. Like the
GNU environment, it started as a cross-platform Free implementation of
commercial unix tools. As far as I know or can remember, neither X.org or
XFree86 ever existed as a Linux-only implementation.
I'll take your word. I am fairly ignorant of the BSD world - the only
BSD I've ever managed to install and get to boot up to a GUI was
PC-BSD, which is kinda cheating.
I was under the impression that for many years, in the XFree86 era,
the BSDs maintained their own X.11 servers, and Solaris still does,
the last time I looked - it offers the Sun X server for Sun graphics
hardware and X.org for non-Sun kit.
I have no experience with AIX on x86, only on RIOS/ROMP, but AFAIAA,
all the commercial Unices still maintain their own X servers, or did
until they died.
Was XFree86 /never/ just a Linux-on-x86 project? XFree86 came along to support more of the whiz-bang
graphics hardware
in the PC world, all of which tended to be very inconsistent and fiddly
even back then. Who in their right mind changes friggin' sync rates in
order to switch in and out of graphics mode? PC video cards, that's
who. Madness.
Well, PC video cards are, generally, at root, VGA cards, which means
enhanced & extended EGA cards, which means enhanced & extended CGA
cards, right? And that implies a base video mode of 640?something at
50 or 60Hz, I think. CGA = 640?200, EGA = 640?350, VGA = 640?480.
All the rest is layers of enhancement over the top.
Anyway, XFree86 was put together to address all that
crap, primarily under Linux.
Well, that's one way of looking at it, but OK, fair enough.
Eventually, most of the free UNIX implementations
started to narrow
their focus to x86, even the still-very-cross-platform ones like NetBSD.
They switched most of their ports over to XFree86 some time ago, when
it became flexible enough to support other hardware, and that's where
most of the development was focused. This wasn't all that long ago.
You'll see these two implementations of the X protocol referred to as
"MIT X11" and "XFree". It's important to note that
"XFree" is no more
"free" than MIT's freely-available X server; that's just what they
chose
to name it.
OK, understood. But underneath that still means 2 different binary
programs, 2 different packages of umpteen drivers, libraries and bits
of code, right?
One implemented by $Programmer at $Project at $Time and another which
is derived from the XFree86 codebase... yes?
That's what I was getting at. I think we may be looking at the same
thing from 2 radically different POVs, but I am not sure.
The various
incarnations of the HURD are very minority offerings
indeed and most of the other Unix-like kernels have their own
userland. You pretty much only get the GNU userland on Linux and you
pretty much only get Linux underneath the GNU userland.
Well not quite. Most commercial UNIX sysadmins, myself included, had
their first introduction to GNU software as improved replacements for
vendor-supplied utilities. This started happening LONG before there
were any viable (or even not-so-viable!) free UNIX implementations.
TO THIS DAY the first thing I do on a brand-new Solaris system is
replace a few Oracle-supplied utilities like "tar" with their GNU
replacements, to bring their functionality into this decade.
Most highly technical organizations (or organizations with competent
technical staffs) had their own "standard OS load" for a given platform,
that consisted of a commercial vendor-supplied UNIX (SunOS, Ultrix, etc)
with all the bad stuff fixed. Fixing stuff usually meant replacing
"tar", "awk", etc with stuff that had actually progressed since the
1970s...and it was almost always GNU. Further, you had source code, so
you could fix any bugs you ran into without having to deal with talking
to incompetent vendor employees and MAYBE getting a bug fix in a few months.
2 Sep
2 Sep
11:38 a.m.
New subject: RMS warned everyone, the rest is up to us -- Re: GNU and the BSD kernel
On 01/09/12 10:50 PM, Liam Proven wrote:
...
But the thing is, the threat we all face to our freedoms from the big
proprietary vendors is immense. ...
And there is, at the end of the day, only one man who warned about
this and who led a project for many decades to ensure that there /is/
an alternative. That man is not Eric Raymond, Linus Torvalds or Mark
Shuttleworth. It's Richard Stallman. ... at the end of
the day, the world could not thank him enough for that. People have
won Nobel prizes for a lot less.
+1000
Thankyou.
--Toby
4:44 p.m.
New subject: RMS warned everyone, the rest is up to us -- Re: GNU and the BSD kernel
On 09/02/2012 11:38 AM, Toby Thain wrote:
On 01/09/12 10:50 PM, Liam Proven wrote:
Yep. There's really no excuse for not being self-reliant. And it's a
hell of a lot safer. I don't trust these suits with my personal data
and work.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
...
But the thing is, the threat we all face to our freedoms from the big
proprietary vendors is immense. ...
And there is, at the end of the day, only one man who warned about
this and who led a project for many decades to ensure that there /is/
an alternative. That man is not Eric Raymond, Linus Torvalds or Mark
Shuttleworth. It's Richard Stallman. ... at the end of
the day, the world could not thank him enough for that. People have
won Nobel prizes for a lot less.
+1000
Thankyou.
4339
days inactive
4342
days old
28 comments
12 participants
participants (12)
-
aek＠bitsavers.org -
als＠thangorodrim.de -
derschjo＠mail.msu.edu -
doc＠vaxen.net -
ggs＠shiresoft.com -
jon＠jonworld.com -
lproven＠gmail.com -
mcguire＠neurotica.com -
mouse＠Rodents-Montreal.ORG -
spectre＠floodgap.com -
toby＠telegraphics.com.au -
wackyvorlon＠me.com