Re: [9fans] one reason ideas from Plan 9 didn't catch on

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[anothy]

// Sticking point?  It's not like namespace groups really ``get in the way''.

no, not at all. i was talking strictly from the point of view of trying to
explain the systems to someone. the per-process namespaces are where
most people get stuck in their understanding. you're correct in noting
that not fully understanding that point doesn't prevent someone from
using the system.

// It's definately a feature that, on a CPU server, I can mount my own
// fileserver without giving everyone access to it.  And the way that
// /dev/cons works under rio is just what you want.

couldn't agree more. it makes nearly everything much nicer to do (the
only things it doesn't make nicer are the things it doesn't affect). i
_really_ like how doing a gateway for an internal-only network becomes
a simple import command
ア

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Eyal Lotem]

anothy@cosym.net wrote:

> no, not at all. i was talking strictly from the point of view of trying to
> explain the systems to someone. the per-process namespaces are where
> most people get stuck in their understanding.

Per-process namespaces are very much like per-process capability pools, but
allow processes to express requests they are not authorized to, as well as
requiring a clumsy namespace-access interface.  What advantages do you find
to this, over pure capability systems?  With capabilities, your requests
are in terms of the capabilities you have (aka: Directly in terms of your
authority), without having to 'name' them in a namespace.

In other words, if you go through to process-grained security, with the
more correct approach to security of visibility, and the terms of the
requests themselves, why not go all the way, with pure capability systems?
Getting rid of traditional file systems has other big pluses.

I personally think that if Plan 9 implemented a pure capability system, and
orthogonal persistency as early as it implemented its design, it could have
caught on, and be a lot more secure/efficient.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Andrew Simmons]

>It was one of those rude, self-serving acts meant to promote a hegemony
>instead of foster liberty. Being pushy is hardly 'kind', irrespective of
>ones definition.
>

If there is any rudeness here, it is certainly not on the part of Mr
Presotto.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Boyd Roberts]

Could someone explain to me why linux caught on?

Is it because a bunch of people had nothing better to do that
to endlessly tinker with it?

I had a fun afternoon yesterday installing linux -- ghastly.

I really loved the question:

   Which one of these 5 useless sendmail configurations don't you want?

None of them were right and I've hacked the code to that thing
and fixed more broken config files than I care to remember.

Plan 9 installs in 15 minutes (or so) or _it doesn't install at all_.

This is good because all I stand to lose is 15 minutes.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Douglas A. Gwyn]

Boyd Roberts wrote:
> Could someone explain to me why linux caught on?

Originally, because it was free, so hobbyists had an incentive
to check it out.  Obviously anything resembling UNIX has great
appeal (by comparison with Microsoft systems) for people
motivated enough to bother to choose which OS they use.
At the time, it seemed like the only feasible cheap alternative
to Windows, and when it proved to be a higher-throughput
platform for Web servers, vendors started paying attention to
it and making sure that Linux device drivers were developed.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

boyd@strakt.com (Boyd Roberts) writes:

> Could someone explain to me why linux caught on?

It worked, it was available, people could hack on it and see their
changes hit the world, there was nothing else.

> I had a fun afternoon yesterday installing linux -- ghastly.

Hrm.  It's generally easier to install than Windows, but this is
rarely noticed because Windows is rarely installed: it's usually
already been done for you by someone else.  But yes, sometimes it can
be hard.  On the other hand, complaining on the Plan 9 list about how
hard GNU/Linux can be to install is a bad prospect: the Plan 9 install
system is, in my experience, considerably worse.

> I really loved the question:
>
>    Which one of these 5 useless sendmail configurations don't you want?

An excellent reason not to use sendmail!  The GNU/Linux distribution I
prefer (Debian) doesn't use it.  (Well, it's there as an option for
those who really want it, but the normal MTA is exim.)

> Plan 9 installs in 15 minutes (or so) or _it doesn't install at all_.
>
> This is good because all I stand to lose is 15 minutes.

So far I lost many more hours than that trying to install Plan 9.

Thomas

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 182 bytes --]

possibly, although we've found that people have been subjected to so much
<propaganda /> and soft SOAP that quite a few programmers
don't necessarily understand files any more.


[-- Attachment #2: Type: message/rfc822, Size: 3555 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 131 bytes --]

this may be the best summary of the use of
namespaces in Plan 9 (and, by association,
Inferno) i've yet heard. thanks, rob!



[-- Attachment #2.1.2: Type: message/rfc822, Size: 1743 bytes --]

From: "rob pike" <rob@plan9.bell-labs.com>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Wed, 14 Nov 2001 09:29:12 -0500
Message-ID: <20011114142914.DC9FB199F2@mail.cse.psu.edu>

> What advantages do you find  to this, over pure capability systems?

The most important aspect of Plan 9's design is that its unusual
properties all derive from the use of files, a concept every computer
user and programmer is comfortable with.  For example, make every
resource file-like, add a network file system, and suddenly every
resource can not only be accessed remotely, the users easily grasp how
to do so.  This really is the crux of the system: a combination that
combines ease of composability with familiarity.

-rob

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[anothy]

[-- Attachment #1: Type: text/plain, Size: 134 bytes --]

this may be the best summary of the use of
namespaces in Plan 9 (and, by association,
Inferno) i've yet heard. thanks, rob!
ア


[-- Attachment #2: Type: message/rfc822, Size: 1743 bytes --]

From: "rob pike" <rob@plan9.bell-labs.com>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Wed, 14 Nov 2001 09:29:12 -0500
Message-ID: <20011114142914.DC9FB199F2@mail.cse.psu.edu>

> What advantages do you find  to this, over pure capability systems?

The most important aspect of Plan 9's design is that its unusual
properties all derive from the use of files, a concept every computer
user and programmer is comfortable with.  For example, make every
resource file-like, add a network file system, and suddenly every
resource can not only be accessed remotely, the users easily grasp how
to do so.  This really is the crux of the system: a combination that
combines ease of composability with familiarity.

-rob

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[presotto]

[-- Attachment #1: Type: text/plain, Size: 974 bytes --]

I tried twice with XOS and Demos/MP.  XOS was a traditional capability system,
Demos a less traditional and more usable one.  XOS was mine, Demos/MP from
Los Alamos/Stanford.  Demos is described in an old SOSP (in the 70's),
XOS in a OSR in the 80's.  We had the usual problem with persistence;
garbage collecting the capabilities, revoking old capabilities, building
tools to walk the arbitrary graphs that resulted and make some sense of
them, ...  Even something like tar turned into a major pain in the backside
to build.  When you gave someone a capability, you had to do a transitive
closure of the access of that capability to figure out what you were giving
away or restrict the capability to be intransitive.  The result was
a lot more copying.

On the positive side, access control was nicer.  Unfortunately, that was the
only plus.

I keep believing that capabilities are a good idea but the concept requires
someone better than me to implement.

[-- Attachment #2: Type: message/rfc822, Size: 2626 bytes --]

From: Eyal Lotem <eyal@hyperroll.com>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Wed, 14 Nov 2001 09:52:54 GMT
Message-ID: <3bf1a2de@news.bezeqint.net>

anothy@cosym.net wrote:

> no, not at all. i was talking strictly from the point of view of trying to
> explain the systems to someone. the per-process namespaces are where
> most people get stuck in their understanding.

Per-process namespaces are very much like per-process capability pools, but
allow processes to express requests they are not authorized to, as well as
requiring a clumsy namespace-access interface.  What advantages do you find
to this, over pure capability systems?  With capabilities, your requests
are in terms of the capabilities you have (aka: Directly in terms of your
authority), without having to 'name' them in a namespace.

In other words, if you go through to process-grained security, with the
more correct approach to security of visibility, and the terms of the
requests themselves, why not go all the way, with pure capability systems?
Getting rid of traditional file systems has other big pluses.

I personally think that if Plan 9 implemented a pure capability system, and
orthogonal persistency as early as it implemented its design, it could have
caught on, and be a lot more secure/efficient.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[rob pike]

> What advantages do you find  to this, over pure capability systems?

The most important aspect of Plan 9's design is that its unusual
properties all derive from the use of files, a concept every computer
user and programmer is comfortable with.  For example, make every
resource file-like, add a network file system, and suddenly every
resource can not only be accessed remotely, the users easily grasp how
to do so.  This really is the crux of the system: a combination that
combines ease of composability with familiarity.

-rob

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

rob@plan9.bell-labs.com (rob pike) writes:

> The most important aspect of Plan 9's design is that its unusual
> properties all derive from the use of files, a concept every computer
> user and programmer is comfortable with.  For example, make every
> resource file-like, add a network file system, and suddenly every
> resource can not only be accessed remotely, the users easily grasp how
> to do so.  This really is the crux of the system: a combination that
> combines ease of composability with familiarity.

Yes, very well said.  This is the sort of thing that I would like to
see in my hypothetical totally-subjective-namespace idea, which (for
example) is not the way Amoeba is built to work.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Fco.J.Ballesteros]

:  at putting releases together as linus and friends or we need a better way.  Perhaps
:  we should keep a machine on the internet with our more or less current
:  sources.

Getting access to current sources together with some mean (9fans?) to
listen for changes would allow some of us to help. At least you could
consider doing this as an experiment to see if it pays. I think that
would be great.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[okamoto]

First of all, I'd like to say "Don't hurry too much to distribute new release".
We are using 3rd ed, and are mostly happy with it.  The most important thing to
us, Plan 9 users, is to enrich it by producing many of applications.   If so, stable
version is desired for such purpose (I don't say the new release will be unstable,
but users are more famillier with current release, and working now for it.  ^_^

If the next release will not fit into a floppy, I'd like to get working version from
net.  However, I suppose this is related to the matter what kind of people you
want to distribute it.   According to my understanding, Plan 9 is not suitable for
personal use, but rather to some organization such as Univ. or company.  I may
be wrong when we consider to run Plan 9 on a single machine is important.
If most users may belong to some organization, getting it from net is suitable, I think.

Kenji

PS. You will see many complaines about the difficulty of Plan 9 installation, and so on.
Don't be afraid it too much.   According to my observations in these more than five years,
users who are contributing Plan 9 community have never complained, but just worked
mainly by him/her-selves... :-)

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[David Gordon Hogan]

[-- Attachment #1: Type: text/plain, Size: 116 bytes --]

He did say `compressed'.

I figure that with 650MB per CD, there ought to be enough
room for GCC 3.0 as well.


[-- Attachment #2: Type: message/rfc822, Size: 1757 bytes --]

From: Dan Cross <cross@math.psu.edu>
To: 9fans@cse.psu.edu
Cc: 
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Wed, 14 Nov 2001 00:12:18 -0500 (EST)
Message-ID: <200111140512.AAA20038@augusta.math.psu.edu>

In article <20011114044205.62B6D199F2@mail.cse.psu.edu> you write:
>base package, maybe 150MB with tex, perl,
				     ^^^^
>cvs, and various other enormous programs.

Perl??  Surely you jest, sir.

	- Dan C.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Russ Cox]

I expect the compressed CD image would
be approximately as large as the compressed
distribution currently is -- 50-70MB for the
base package, maybe 150MB with tex, perl,
cvs, and various other enormous programs.

The format shouldn't change the size significantly.

Russ

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Dan Cross]

In article <20011114044205.62B6D199F2@mail.cse.psu.edu> you write:
>base package, maybe 150MB with tex, perl,
				     ^^^^
>cvs, and various other enormous programs.

Perl??  Surely you jest, sir.

	- Dan C.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 426 bytes --]

most equipment i could buy last year offered cd boot.  there's no need
to upgrade: booting a cd kernel from a floppy or hard disk to run from
the cd is nearly as good.  it was unfortunate i didn't include a 9pccd on
the vita cd last time, but i didn't think of it then.  in fact, it was only
the recent arrival of a floppy-less thinkpad that reminded me of it.
i'll probably put it up on the web site now i've got it.


[-- Attachment #2: Type: message/rfc822, Size: 1947 bytes --]

To: <9fans@cse.psu.edu>
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Tue, 13 Nov 2001 23:27:53 -0000
Message-ID: <006f01c16c9a$d20e4b00$2248dec2@falken>

How much target h/w has a CD-drive and a BIOS
that supports CD-boot?
I know mine doesn't!

Looks like it's getting time to upgrade.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 651 bytes --]

i'd do a bootable cd with a diskette used only if you can't
get your system to boot the CD directly.  i suspect the boot
diskette will then fit 1.44mb.  the main reason the vita cd isn't bootable
now is that i got it to boot but then discovered i'd need to mend
something to complete the process and there wasn't enough time (at the
time).  recently i've done some more to get a floppy-less thinkpad
going, and the next cd should be bootable.  this simplifies the
process because you've got access to the full version of all commands
on the full system (on the cd) rather than trying to cram things in
to a compressed file system on floppy.

[-- Attachment #2: Type: message/rfc822, Size: 1688 bytes --]

To: 9fans@cse.psu.edu
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Tue, 13 Nov 2001 16:46:31 -0500
Message-ID: <200111132146.fADLkWd06580@new-york.lcs.mit.edu>

multiple floppies is a slippery slope.
1.44MB really should be enough to do an installation.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Chris Hollis-Locke]

How much target h/w has a CD-drive and a BIOS
that supports CD-boot?
I know mine doesn't!

Looks like it's getting time to upgrade.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Lucio De Re]

On Tue, Nov 13, 2001 at 11:27:53PM -0000, Chris Hollis-Locke wrote:
>
> How much target h/w has a CD-drive and a BIOS
> that supports CD-boot?
> I know mine doesn't!
>
> Looks like it's getting time to upgrade.
>
Hm,  Uhfrika's a bit behind on the bandwidth bandwagon, too, so a CD
image would be a little tough on us noble savages :-)

But I'll supply the local community if there is demand.  Postal service
sucks too, unfortunately.

++L

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[presotto]

[-- Attachment #1: Type: text/plain, Size: 404 bytes --]

sorry I'm really an idiot, I could have sworn I'ld included a message there.
It was an answer to the suggestion of why not put the distribution on two floppies.

I really didn't even consider it because I hate floppies to begin with and
two scares me.  However, it may come to that.  Dhog is putting in dynamicly
linked drivers so I may go to building floppies with just the drivers needed
first.

[-- Attachment #2: Type: message/rfc822, Size: 2772 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 34 bytes --]

Didn't think of what?

	Sape


[-- Attachment #2.1.2: Type: message/rfc822, Size: 1226 bytes --]

From: presotto@closedmind.org
To: 9fans@cse.psu.edu
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Tue, 13 Nov 2001 16:44:30 -0500
Message-ID: <20011113214431.5C552199E8@mail.cse.psu.edu>

Because I'm an idiot and didn't think of it?

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Dan Cross]

In article <20011113215001.AD7EB1998A@mail.cse.psu.edu> you write:
>I really didn't even consider it because I hate floppies to begin with and
>two scares me.  However, it may come to that.  Dhog is putting in dynamicly
>linked drivers so I may go to building floppies with just the drivers needed
>first.

I hate floppies too, but they fill a gap.  I like the idea of using a
kernel with just the drivers you need, but the idea of dynamicaly linked
drivers scares me.

	- Dan C.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Russ Cox]

multiple floppies is a slippery slope.
1.44MB really should be enough to do an installation.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Sape Mullender]

[-- Attachment #1: Type: text/plain, Size: 34 bytes --]

Didn't think of what?

	Sape


[-- Attachment #2: Type: message/rfc822, Size: 1226 bytes --]

From: presotto@closedmind.org
To: 9fans@cse.psu.edu
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Tue, 13 Nov 2001 16:44:30 -0500
Message-ID: <20011113214431.5C552199E8@mail.cse.psu.edu>

Because I'm an idiot and didn't think of it?

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[presotto]

Because I'm an idiot and didn't think of it?

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[andrey]

you simply have never had to install freebsd :)

presotto@closedmind.org wrote:

> Because I'm an idiot and didn't think of it?

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[David Gordon Hogan]

ravage@ssz.com writes:
> You just want to bitch.

Are you familiar with the concept of ``projection'' in psychology?

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[David Gordon Hogan]

> > The license thread is much much more tedious and longwinded
> > than the compiler thread.  And pointless; what does anyone hope
> > to accomplish?

> If the fault is in the license, try to get it changed to something more
> reasonable.

Did you even READ what I said?  I notice that you chopped off an
important part of my mail, which was:

<dhog> If you care that much, become a lawyer and get
<dhog> yourself employed by the Lucent legal department...

ie we (or rob at least) have tried all we could, it's out of our
hands.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

On Tue, 13 Nov 2001, David Gordon Hogan wrote:

> Did you even READ what I said?  I notice that you chopped off an
> important part of my mail, which was:

Yes, it isn't relevant to the point.

> <dhog> If you care that much, become a lawyer and get
> <dhog> yourself employed by the Lucent legal department...

And you ask why I chopped it out...;(

> ie we (or rob at least) have tried all we could, it's out of our
> hands.

Which isn't relevent. Especially since nobody said anything about anyone
doing a BAD job. <shrug>

The bottem line is that there ARE issues with the presentation of the
license if not its substance (I'll offer this is an open question as
well). If those issues aren't addressed in some mechanism they will remain
a significant impedement.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[presotto]

In the very recent past, I've installed Windows 98, Red Hat, and Plan 9.  I haven't
tried Debian yet so I can't say anything about it.

I must say, the hardest was Plan 9 although procedurally it was the easiest (less
questions, more ability to screw around if things did go wrong) although that
only helps if you have a basic familiarity with Plan 9.

The main problem was that the pre-patch system that's out there just doesn't go with
any hardware I could find at Avaya.  Everything I had was supported but only after
patches.  I ended up building a new kernel and 9load at bell labs, sticking them
onto the install floppy, along with an edited vgadb and going around the cycle
3 or 4 times till I got it right.  I couldn't build a kernel or 9load that would
work more generally because one with all the drivers is too big (to fit in 64k, to
fit compressed on a 1.4meg floppy, ...).  I know that precious few people could
have done the same.

The startup kernel for next release isn't going to fit on a floppy if we
don't change things.  We're (jmk and I mostly) considering suggestions
for what to do.
- One possibility is to build a kernel on the fly for people
downloading and have it contain the right drivers.  We get enough info from
the little question and answer session to set up plan9.ini.  That could keep
us going for a number of generations since it considerably reduces kernel size.
- Another is to have dynamicly linked drivers though that too would require
a prebuild since not all the drivers would fit on the floppy and drivers
with linkage info are larger.
- We could just give everyone login access to
a machine on the net and have them do their builds there and copy stuff
out as they see fit.
- We could also just start releasing CDROM's again where size wouldn't matter
as much.  We'ld need someone willing to make them.
- ???

Once I got a kernel/9load that worked, things went well till I wanted
to connect out.  Plan 9 is really crappy to customize once you've got a
stand-alone system up and running.  If noone else has done it, I'm about to start
on a GUI based configuration tool, sort of a collision between X86Config,
winipcnfig, and Wavelanconfig, ...  It needs to switch on something
(I'm assuming $menuitem) to pick a configuration; my laptop normally
wakes up tio find itself in any of 2 corporate networks and 3 ISP's.
DHCP/ppp/v6-resource-discovery doesn't provide enough info and often
has to be overriden so just depending on that isn't enough, you have
to be able to override/supplement.

Setting up plan 9 authentication is like giving birth through your eye ball.
That's all my fault.  Rsc, ehg, and I have been working on a new security architecture that
makes ssh, ssl, private passwords, etc. easier to keep track of and use,
sort of sshagent++, in addition to fixing the plan9 authentication.  I want
to get that out but it depends on getting the new system out which depends
on figuring out how to do a release better...

Finally, we just can't get releases put together fast enough.  We're now
a whole protocol behind so that we can't even send diffs out.  Part of that
is that this is a sideline for all of us.  We either need people as decicated
at putting releases together as linus and friends or we need a better way.  Perhaps
we should keep a machine on the internet with our more or less current
sources.  It wouldn't be as consistent as a release (man pages slightly out of
date, some things that might not build) but would be someplace to grab new
stuff from to try out.  I'ld prefer that to CVS since the bulk of the stuff
is still done at the labs without CVS.

All suggestions welcome.  Sorry for the long message.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[William Josephson]

On Tue, Nov 13, 2001 at 02:58:16PM -0500, presotto@closedmind.org wrote:

> - One possibility is to build a kernel on the fly for people
>downloading and have it contain the right drivers.  We get enough info from
>the little question and answer session to set up plan9.ini.  That could keep
>us going for a number of generations since it considerably reduces kernel size

> - We could also just start releasing CDROM's again where size wouldn't matter
> as much.  We'ld need someone willing to make them.

One of the big advantages of the current system is that I can make a
boot floppy off the net and then install the whole system over the
network, too.  Building a kernel on the fly has the advantage of
preserving this ability.  On the other hand, distributing a bootable
ISO image seems to have worked reasonably well with things like
FreeBSD and Linux.  Having a source of burned CDs is nice, too, of
course, but CD-R drives are pretty cheap anymore, so making people
burn their own (or maybe buy a copy from Vita Nuova) doesn't strike me
as too unreasonable.

I have yet to see anyone get installation "right" -- it never just
works in every case.  I have had by far the worst time with Windows on
the one or two occasions I made the mistake of trying to install it
myself.  Ugh.

 -WJ

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Mike Haertel]

>The startup kernel for next release isn't going to fit on a floppy if we
>don't change things.  We're (jmk and I mostly) considering suggestions
>for what to do.
>[possibilities elided]
>- ???

Why not split it into two floppies: one containing the kernel proper,
that prompts you to insert the 2nd floppy containing the file system
and all the user level programs.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[George Michaelson]

> Setting up plan 9 authentication is like giving birth through your eye ball.

This wins my grossest analogy of the year award. EUGHHHHH!

> That's all my fault.  Rsc, ehg, and I have been working on a new security
> architecture that makes ssh, ssl, private passwords, etc. easier to keep
> track of and use, sort of sshagent++, in addition to fixing the plan9
> authentication.

This both intrigues and worries me. I use ssh-agent a lot, but I have huge
lingering worries that the chain of open FD back through process history
and the IPC mechanisms is a gaping yaw of risk.

Wouldn't a kerberos tkt like mechanism pose less risks? Isn't the pain of
occaisional re-authentication purposeful? I exclude systems like the SUNray
where a physical token can be removed and moved to carry the auth info.

Having said which, if Plan9 has a clean abstraction for this (and the little
I understand about the mechanisms suggest this strongly) then its a wonderful
idea.

Doesn't it also pose risks for loss of that parent ssh-agent-like thing?

cheers
	-george
--
George Michaelson       |  APNIC
Email: ggm@apnic.net    |  PO Box 2131 Milton QLD 4064
Phone: +61 7 3367 0490  |  Australia
  Fax: +61 7 3367 0482  |  http://www.apnic.net

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[William Josephson]

On Wed, Nov 14, 2001 at 08:54:53AM +1000, George Michaelson wrote:

> > That's all my fault.  Rsc, ehg, and I have been working on a new security
> > architecture that makes ssh, ssl, private passwords, etc. easier to keep
> > track of and use, sort of sshagent++, in addition to fixing the plan9
> > authentication.
>
> This both intrigues and worries me. I use ssh-agent a lot, but I have huge
> lingering worries that the chain of open FD back through process history
> and the IPC mechanisms is a gaping yaw of risk.

I personally refuse to use ssh-agent and avoid ssh X11 forwarding
because I don't trust either to get this sort of thing right.  I'll
leave it to Presotto to discuss the system in detail, but they've done
a lot of thinking about this sort of problem.

> Wouldn't a kerberos tkt like mechanism pose less risks? Isn't the pain of
> occaisional re-authentication purposeful? I exclude systems like the SUNray
> where a physical token can be removed and moved to carry the auth info.

It isn't tied to a particular protocol.

> Having said which, if Plan9 has a clean abstraction for this (and the little
> I understand about the mechanisms suggest this strongly) then its a wonderful
> idea.

> Doesn't it also pose risks for loss of that parent ssh-agent-like thing?

It serves a similar purpose and ssh-agent is probably the closest
thing under Unix.  Unlike Unix, Plan 9 has per-process namespaces (and
single-user terminals).  You end up having to trust the kernel either
way, of course.

 -WJ

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[David Gordon Hogan]

> No, it shouldn't. It's a hell of a lot more interesting than a long winded
> discussion about 'compiler efficiency' which is a moot point for most
> users. Where's the 'compiler' list?...

The license thread is much much more tedious and longwinded
than the compiler thread.  And pointless; what does anyone hope
to accomplish?  If you care that much, become a lawyer and get
yourself employed by the Lucent legal department...

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

On Mon, 12 Nov 2001, David Gordon Hogan wrote:

> > No, it shouldn't. It's a hell of a lot more interesting than a long winded
> > discussion about 'compiler efficiency' which is a moot point for most
> > users. Where's the 'compiler' list?...
>
> The license thread is much much more tedious and longwinded
> than the compiler thread.  And pointless; what does anyone hope
> to accomplish?

If the fault is in the license, try to get it changed to something more
reasonable. If it's a common misconception then it's important to address
it (perhaps at some point through a FAQ - though the discussion has to
take place for the FAQ to be birthed...).

My only real sadness is that the discussion is seen as a problem and not
an opportunity to address the emotional aspects of peoples computer usage.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Dan Cross]

In article <Pine.LNX.3.96.1011112181546.1019K-100000@einstein.ssz.com> you write:
>My only real sadness is that the discussion is seen as a problem and not
>an opportunity to address the emotional aspects of peoples computer usage.

That's because probably 95% of the people on this list are interested in
the technology and not debates surrounding the merits of this license or
that.

What does this have to do with Plan 9?

	- Dan C.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

On Mon, 12 Nov 2001, Dan Cross wrote:

> What does this have to do with Plan 9?

I think that is its own worst answer.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[William Josephson]

On Mon, Nov 12, 2001 at 08:16:22PM -0600, Jim Choate wrote:
> On Mon, 12 Nov 2001, Dan Cross wrote:
> > What does this have to do with Plan 9?
> I think that is its own worst answer.

YM "nothing" HTH

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

cross@math.psu.edu (Dan Cross) writes:

> In article <Pine.LNX.3.96.1011112181546.1019K-100000@einstein.ssz.com> you write:
> >My only real sadness is that the discussion is seen as a problem and not
> >an opportunity to address the emotional aspects of peoples computer usage.
>
> That's because probably 95% of the people on this list are interested in
> the technology and not debates surrounding the merits of this license or
> that.
>
> What does this have to do with Plan 9?

Have you forgotten the history of the thread?  Go back and look at the
first message posted under this subject (which was mine).  You'll see
that it addresses a relevant question that was being addressed at the
time I posted it.

It's pretty obvious to me that "How Plan 9 is licensed" is a relevant
topic for the Plan 9 list.  It's also pretty obvious to me that
"reasons why Plan 9 ideas haven't caught on" is a relevant topic: it's
not one that I started, by the way.

It's also clear that some people think that this list is only for
people who think Plan 9 is the bee's knees; that all things Plan 9 are
the best in the world, and that all other things are a waste of time
and effort.  But not everyone here has that kind of attitude.

Thomas

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[David Gordon Hogan]

> but by far the majority of people i've explained it to get stuck,
> at least for a while, on private namespaces. the largest single sticking
> point in the system.

Sticking point?  It's not like namespace groups really ``get in the way''.
(occaisional inter-window mishaps notwithstanding).

It's definately a feature that, on a CPU server, I can mount my own
fileserver without giving everyone access to it.  And the way that
/dev/cons works under rio is just what you want.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[anothy]

// It's hard to make people "get" private name spaces for some reason.
// At least it's hard for me to make people get it.

hey, don't feel bad. by now i've probably tried to explain Plan 9 and/or
Inferno directly to hundreds of people, and i find that to be the toughest
single concept, as well. virtual machines, Styx/9P, Limbo, device
control via read/write (no ioctl), and others people seem to get with no
problem, but by far the majority of people i've explained it to get stuck,
at least for a while, on private namespaces. the largest single sticking
point in the system.

well, that and how the arrow keys work.
ア

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[presotto]

Actually, there were requests for it, go look at the archive.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[nigel]

>> Which none of us asked for, nor apparently find value in.

I'd prefer it if you didn't volunteer you own opinions as mine.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

On Mon, 12 Nov 2001 nigel@9fs.org wrote:

> >> Which none of us asked for, nor apparently find value in.
>
> I'd prefer it if you didn't volunteer you own opinions as mine.

I didn't. I simply observed that those of us who were having this
discussion didn't jump en masse. That DOES imply a lack of interest/value
irrespective if your INDIVIDUAL views.

You just want to bitch.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Skip Tavakkolian]

Wrong assumption; I my case, it only implies that I ignore the the
irrelevant drivel. Everyone here has had to read the license and agree to
it. Why would we need a rehash? If my needs or plans for Plan9 change then
I'll revisit the license and get on the other list and talk about it.

I respect Stallman, but when it starts looking like he is issuing decrees,
it is too much like the Ayatolla's Fatwa for my liking.  I am certain this
is not his intent, but there are those that out of to respect for Stallman,
 lack of experience, or lack of time will take his word for gospel rather
than verify for themselves. This is not healthy; It starts looking like a
GNU cult, rather than a free association of people that believe keeping the
source code locked up is wrong.

(oh shit, now I'm starting to sound like a windbag)

At 06:03 PM 11/12/2001 -0600, Jim Choate wrote:
>
>On Mon, 12 Nov 2001 nigel@9fs.org wrote:
>
>> >> Which none of us asked for, nor apparently find value in.
>>
>> I'd prefer it if you didn't volunteer you own opinions as mine.
>
>I didn't. I simply observed that those of us who were having this
>discussion didn't jump en masse. That DOES imply a lack of interest/value
>irrespective if your INDIVIDUAL views.
>
>You just want to bitch.
>
>
> --
>    ____________________________________________________________________
>
>             Day by day the Penguins are making me lose my mind.
>
>                                             Bumper Sticker
>
>       The Armadillo Group       ,::////;::-.          James Choate
>       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
>       www.ssz.com            .',  ||||    `/( e\      512-451-7087
>                           -====~~mm-'`-```-mm --'-
>    --------------------------------------------------------------------
>
>
>

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

skipt@real.com (Skip Tavakkolian) writes:

> I respect Stallman, but when it starts looking like he is issuing decrees,
> it is too much like the Ayatolla's Fatwa for my liking.

Heavens!  He hasn't posted here, and nobody has issued any decrees.

Thomas

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[geoff]

> Shoudln't this thread be sent to the plan 9 licensing list that
> Presotto so kindly created?

Yes it should.  It's been over two days since the mailing list
plan9-license-discussions@plan9.bell-labs.com was created and all
licensing discussion should now move there.  Otherwise we may ask
Scott to automatically divert all such messages there.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

geoff@collyer.net writes:

> > Shoudln't this thread be sent to the plan 9 licensing list that
> > Presotto so kindly created?
>
> Yes it should.  It's been over two days since the mailing list
> plan9-license-discussions@plan9.bell-labs.com was created and all
> licensing discussion should now move there.  Otherwise we may ask
> Scott to automatically divert all such messages there.

That's not appropriate.  Or rather, doing so would violate the
principles of Usenet, and the gatewaying of 9fans to comp.os.plan9 to
the 9fans mailing list should cease at such a point.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Fco.J.Ballesteros]

Shoudln't this thread be sent to the plan 9 licensing list
that Presotto so kindly created?

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

Which none of us asked for, nor apparently find value in.

It was one of those rude, self-serving acts meant to promote a hegemony
instead of foster liberty. Being pushy is hardly 'kind', irrespective of
ones definition.

No, it shouldn't. It's a hell of a lot more interesting than a long winded
discussion about 'compiler efficiency' which is a moot point for most
users. Where's the 'compiler' list?...

On Mon, 12 Nov 2001, Fco.J.Ballesteros wrote:

> Shoudln't this thread be sent to the plan 9 licensing list
> that Presotto so kindly created?


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

nemo@plan9.escet.urjc.es (Fco.J.Ballesteros) writes:

> Shoudln't this thread be sent to the plan 9 licensing list
> that Presotto so kindly created?

Well, I'm reading comp.os.plan9.  Perhaps if comp.os.plan9.licensing
exists, I'll post relevant articles there instead.  Until then, this
is the place.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Scott Schwartz]

Thomas writes:
| Well, I'm reading comp.os.plan9.  Perhaps if comp.os.plan9.licensing
| exists, I'll post relevant articles there instead.  Until then, this
| is the place.

gnu.misc.discuss is an appropriate place for such discussions.

The problem with having them on the 9fans list is that the regulars have
already heard it all, and resurrecting the topic just gets people annoyed.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[okamoto]

>I'm starting to get tired of this thread.

I've been bored.  :-)

Kenji

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Fco.J.Ballesteros]

I'm starting to get tired of this thread.

1. Plan 9 is free software, you may like it or not, but it's free
software. The fact that It's not [L]GPL is not an issue.

2. The small user base is mainly due to the fact that it is really
hard to convince people to loose their applications in favor of
a new way of doing things; but this have been the same in
operating systems as long as I remember.

3. Nevertheless, the ideas are spreading. Looked at /proc on Linux
these years? At the attempts to use userfs on Linux? Remember that
some of the ideas are too radical for a unix system and may never
be imported to it.

4. Regarding your suggestions for improvements, I'm still waiting
for you to write something like:

	"Here is this piece of code of a plan 9 application;
	now, by changing plan 9 in this way, this application
	could be now as I show below (i.e. more simple,
	or more whatever)..."

Could we get back to tech topics and stop religious affairs now?
I'm sorry to be so rude, but I wouldn't like this list to look
like the linux ones.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[pac]

>> From 9fans-admin@cse.psu.edu Fri Nov 09 10:34:02 2001
>> Envelope-to: cej@cejchan.gli.cas.cz
>> Delivered-To: 9fans@cse.psu.edu
>> From: Fco.J.Ballesteros <nemo@plan9.escet.urjc.es>
>> To: 9fans@cse.psu.edu
>> Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
>> MIME-Version: 1.0
>> Content-Type: text/plain; charset="US-ASCII"
>> Content-Transfer-Encoding: 7bit
>> Sender: 9fans-admin@cse.psu.edu
>> Errors-To: 9fans-admin@cse.psu.edu
>> X-BeenThere: 9fans@cse.psu.edu
>> X-Mailman-Version: 2.0.6
>> Precedence: bulk
>> Reply-To: 9fans@cse.psu.edu
>> List-Id: Fans of the OS Plan 9 from Bell Labs <9fans.cse.psu.edu>
>> List-Archive: <http://lists.cse.psu.edu/archives/9fans/>
>> Date: Fri, 9 Nov 2001 10:21:23 +0100
>>
>> I'm starting to get tired of this thread.
>>
>> 1. Plan 9 is free software, you may like it or not, but it's free
>> software. The fact that It's not [L]GPL is not an issue.
>>
>> 2. The small user base is mainly due to the fact that it is really
>> hard to convince people to loose their applications in favor of
>> a new way of doing things; but this have been the same in
>> operating systems as long as I remember.
>>
>> 3. Nevertheless, the ideas are spreading. Looked at /proc on Linux
>> these years? At the attempts to use userfs on Linux? Remember that
>> some of the ideas are too radical for a unix system and may never
>> be imported to it.
>>
>> 4. Regarding your suggestions for improvements, I'm still waiting
>> for you to write something like:
>>
>> 	"Here is this piece of code of a plan 9 application;
>> 	now, by changing plan 9 in this way, this application
>> 	could be now as I show below (i.e. more simple,
>> 	or more whatever)..."
>>
>> Could we get back to tech topics and stop religious affairs now?
>> I'm sorry to be so rude, but I wouldn't like this list to look
>> like the linux ones.
>>
>>

I agree 100%.
-Peter.


--
Peter A Cejchan
biologist
Acad. Sci., Prague, CZ
<cej at cejchan dot gli dot cas dot cz>

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

nemo@plan9.escet.urjc.es (Fco.J.Ballesteros) writes:

> 1. Plan 9 is free software, you may like it or not, but it's free
> software. The fact that It's not [L]GPL is not an issue.

There is lots of free software that isn't GPL'd.  The problem is that
Plan 9's license does contain important liberty restrictions, and
those restrictions are, in fact, important to the very large free
software development community.

Plan 9 is available at low cost, but that's a separate issue from
whether it is free software in the libre sense, which is the sense
that matters to all those developers.

> 4. Regarding your suggestions for improvements, I'm still waiting
> for you to write something like:
>
> 	"Here is this piece of code of a plan 9 application;
> 	now, by changing plan 9 in this way, this application
> 	could be now as I show below (i.e. more simple,
> 	or more whatever)..."

I don't think I've made any suggestions for improvements.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[David Rubin]

"Fco.J.Ballesteros" wrote:

> 3. Nevertheless, the ideas are spreading. Looked at /proc on Linux
> these years? At the attempts to use userfs on Linux? Remember that
> some of the ideas are too radical for a unix system and may never
> be imported to it.

For example, look at the December issue of DDJ (Operatig Systems). Someone at
LANL has written about how they implemented process namespaces in Linux.

	david

--
If 91 were prime, it would be a counterexample to your conjecture.
    -- Bruce Wheeler

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Ronald G Minnich]

On Mon, 12 Nov 2001, David Rubin wrote:

> For example, look at the December issue of DDJ (Operatig Systems). Someone at
> LANL has written about how they implemented process namespaces in Linux.

yeah, that was me. I just put that code on my web site at
www.acl.lanl.gov/~rminnich.

I actually wrote that a few years back while at Sarnoff, with DARPA
funding.

One reason it never made it out to publication earlier was Usenix
reviewers who did not seem to understand that concept. It's hard to make
people "get" private name spaces for some reason. At least it's hard for
me to make people get it.

ron

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Russ Cox]

[-- Attachment #1: Type: text/plain, Size: 578 bytes --]

> > >>    modules...there are also several newer GUI's out there in the
> > >>    Open Source landscape that might bring a better interface to
> > >>    Plan 9.
> >
> > what are those?
>
> Source Forge and Freshmeat are good starting places.

You might as well say ``google is a good starting place.''
There's so much stuff on Source Forge at least that I certainly
can't separate wheat from chaff.  Google might actually be
a better starting place, since PageRank will probably give
me more interesting projects first.

What GUIs did you have in mind?

Russ


[-- Attachment #2: Type: message/rfc822, Size: 2164 bytes --]

From: Jim Choate <ravage@ssz.com>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] one reason ideas from Plan 9 didn't catch on
Date: Thu, 8 Nov 2001 18:51:31 -0600 (CST)
Message-ID: <Pine.LNX.3.96.1011108184616.364H-100000@einstein.ssz.com>


On Thu, 8 Nov 2001 forsyth@vitanuova.com wrote:

> >>    modules...there are also several newer GUI's out there in the
> >>    Open Source landscape that might bring a better interface to
> >>    Plan 9.
>
> what are those?

Source Forge and Freshmeat are good starting places.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[presotto]

On Thu Nov  8 05:45:28 EST 2001, tb+usenet@becket.net wrote:
>
> There are some pretty big reasons that Plan 9's very good ideas are
> sitting in an eddy of the stream of OS design: because the political
> shenanigans of those who hold the keys have done their best to keep
> those ideas out of the mainstream.

This is hatefully unfair.  The shenanigans involve Rob spending months
fighting with lawyers to get a license as close to possible to the
model we originally gave them, i.e., ``do with it what you want, just don't
sue us if it breaks''.  It's amazing to me how that became what it
did.  However, its through no fault of Rob's, he got dragged kicking
and screaming all the way.

I just reread the GPL.  The main differences are indeed our 2 clauses

1) our license is one sided.  We demand that, on request, modifications
  are made available to Lucent if the modifications are otherwise
  distributed.  The GPL requires them to be made available to anyone.

2) our license limits lawsuits in too general a way.  You can't sue
  Lucent over intellectual property and keep the license.  There's
  nothing like this in the GPL. Instead it says that you can't
  include anything in the code that might have IP implications.
  Should you do so, you can't redistribute.

The first was intended to make sure Lucent got something back after
paying our salaries for years.  I beleive the more general form would
work for that too and the lawyers might be persuaded.

The second I doubt we can ever do anything about.  Lucent's lawyers don't
want Lucent being sued over stuff in the code.  Even though they didn't
make it more specific I expect that unless the IP involves the code that Lucent
released, it's not defensable.  I have to talk to an Avaya lawyer later
this week so maybe I can get an opinion.  Any of you lawyers out there?

I think lucent has been very upright about (1), i.e., if you give it
to them, it gets redistributed.  I have no idea about (2) since its
never come up when someone sued Lucent that I know of.

The question is, what does free mean?

It clearly doesn't mean the freedom of the contributor since any
licence fetters him in some way.  GPL doesn't let him distribute
changed binary without making the source available to everyone for
example.  It also seems to mean that you lose exclusive rights to
anything you embody in the code in the case of the GPL and that
you lose the ability to sue Lucent in the Plan 9 one.

In both cases the license is transitive.  Does that have to
do with free or freedom?

It does seem to mean that there is no monetary cost to obtaining,
using, and modifying the code other than the cost of copying it.  That
pertains to both licenses.

In these senses both licenses seem free to me, though the Plan 9
one seems lopsided in its fetters.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

presotto@closedmind.org writes:

> On Thu Nov  8 05:45:28 EST 2001, tb+usenet@becket.net wrote:
> >
> > There are some pretty big reasons that Plan 9's very good ideas are
> > sitting in an eddy of the stream of OS design: because the political
> > shenanigans of those who hold the keys have done their best to keep
> > those ideas out of the mainstream.
>
> This is hatefully unfair.  The shenanigans involve Rob spending months
> fighting with lawyers to get a license as close to possible to the
> model we originally gave them, i.e., ``do with it what you want, just don't
> sue us if it breaks''.  It's amazing to me how that became what it
> did.  However, its through no fault of Rob's, he got dragged kicking
> and screaming all the way.

You're pointing out that Rob Pike doesn't hold the keys.  I'm speaking
of the shenanigans of those who hold the keys.  It may be unfair that
the people who hold the keys won't let Plan 9 out, but it's still a
fact that they do so.

> I just reread the GPL.  The main differences are indeed our 2 clauses
>
> 1) our license is one sided.  We demand that, on request, modifications
>   are made available to Lucent if the modifications are otherwise
>   distributed.  The GPL requires them to be made available to anyone.

No, the GPL does not require them to be made available to anyone, with
the sole exception of people who want to make binary-only
distributions.  For people who distribute source, you can distribute
your changes to as few or as many people as you like

> 2) our license limits lawsuits in too general a way.  You can't sue
>   Lucent over intellectual property and keep the license.  There's
>   nothing like this in the GPL. Instead it says that you can't
>   include anything in the code that might have IP implications.
>   Should you do so, you can't redistribute.

The problem with #2 is a total disaster for free software.

It amounts to "Lucent wants the right to violate any license on
anything I write."

> I think lucent has been very upright about (1), i.e., if you give it
> to them, it gets redistributed.  I have no idea about (2) since its
> never come up when someone sued Lucent that I know of.

Certainly if I do send a change back to Lucent, and it's a free
software product, Lucent should have the right to distribute the
change to other people.  That's what freedom is.

But the problem is that I shouldn't have to send the change back to
Lucent.  It might be very expensive for me to do so, for example.  Now
this clause is much better than those which mandate sending all
changes back: it is only operative if Lucent actually asks for the
change to be sent back.

> In both cases the license is transitive.  Does that have to
> do with free or freedom?

"free software" as I use it, and as all those people who are committed
to it use it, is a fairly well-defined notion.  The boundaries are
sometimes fuzzy, but the Plan 9 license fails all of the tests out
there (the FSF defn, the Debian Free Software Guidelines, the *BSD
people, and so forth).

> It does seem to mean that there is no monetary cost to obtaining,
> using, and modifying the code other than the cost of copying it.  That
> pertains to both licenses.

You've greatly misunderstood the GPL then.  There is nothing about the
GPL which prevents charging more than the cost of copying for getting
a copy.  The FSF itself charges significantly more than the cost of
copying for a CD-ROM, and makes a tidy income from that trade.

Thomas

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[John S. Dyson]

presotto@closedmind.org wrote in message news:<20011108145513.B0E9B199F2@mail.cse.psu.edu>...
> On Thu Nov  8 05:45:28 EST 2001, tb+usenet@becket.net wrote:
> >
> > There are some pretty big reasons that Plan 9's very good ideas are
> > sitting in an eddy of the stream of OS design: because the political
> > shenanigans of those who hold the keys have done their best to keep
> > those ideas out of the mainstream.
>
> This is hatefully unfair.  The shenanigans involve Rob spending months
> fighting with lawyers to get a license as close to possible to the
> model we originally gave them, i.e., ``do with it what you want, just don't
> sue us if it breaks''.  It's amazing to me how that became what it
> did.  However, its through no fault of Rob's, he got dragged kicking
> and screaming all the way.
>
> I just reread the GPL.  The main differences are indeed our 2 clauses
>
> 1) our license is one sided.  We demand that, on request, modifications
>   are made available to Lucent if the modifications are otherwise
>   distributed.  The GPL requires them to be made available to anyone.
>
Firstly, I am NOT a fan of general purpose use of the GPL, but I must
correct your impression under subtopic 1 above.   You only have to
give source code to those who you give binaries.  There are limited
uses (IMO) of the GPL, but it is NOT a license of free software in a
traditional sense (e.g. give a friend the binaries that you built, but
have lost the source, or cannot represent a reasonable pointer to the
source.)

John

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[forsyth]

>>    modules...there are also several newer GUI's out there in the
>>    Open Source landscape that might bring a better interface to
>>    Plan 9.

what are those?

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

On Thu, 8 Nov 2001 forsyth@vitanuova.com wrote:

> >>    modules...there are also several newer GUI's out there in the
> >>    Open Source landscape that might bring a better interface to
> >>    Plan 9.
>
> what are those?

Source Forge and Freshmeat are good starting places.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

[9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

One reason that ideas from Plan 9 didn't catch on is that Rob Pike
et. al. filed for patents on some of the ideas in Plan 9.  When he
came to MIT's AI Lab, and gave a nice presentation on Plan 9, I asked
him which of the ideas he had talked about we were allowed to use in
our own software projects.  He said "as far as I'm concerned, all of
them".  I asked if there were any patents that might matter as far as
AT&T was concerned, and he said there were some, but that he didn't
even understand the patent applications.

I know that his talk made an impression: the innovation of the ideas,
the impressiveness of the system built on them, and that not only
didn't we know if we would be sued for using similar ideas in our own
systems, but Rob wasn't going to tell us if that was possible or not.

And then, years later, after Plan 9 failed to capture a big audience,
it gets released for more public consumption, but for some
incomprehensible reason, is still not free software.

There are some pretty big reasons that Plan 9's very good ideas are
sitting in an eddy of the stream of OS design: because the political
shenanigans of those who hold the keys have done their best to keep
those ideas out of the mainstream.

Thomas

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

[SSZ: Replies including hangar18 will bounce. I'll forward any relevant
      replies. Sorry, but we insist on a members only submission policy.]

On Thu, 8 Nov 2001, Thomas Bushnell, BSG wrote:

> There are some pretty big reasons that Plan 9's very good ideas are
> sitting in an eddy of the stream of OS design: because the political
> shenanigans of those who hold the keys have done their best to keep
> those ideas out of the mainstream.

While I don't necessarily like some of the 'head-in-the-sand ivory-tower
how-dare-you-question-me' attitude of some aspects of the Plan 9
developers and their hanger-ons views (but I can live with it w/o any
major issues - ignore -em - don't need 'em) I fail to see your problem
with the current Plan 9 license. Yes, the original license (ie $300/use,
no commercial usage) was problematic, and yes the original 'Open Source'
license release was worthless as written, they DID re-write it several
times into its current state. What are your specific views on the current
license shortfalls?

You're about the third or fourth person who has made some complaint on the
Plan 9 license in the last couple of months that I've run across. Yet, I
read it and don't see anything that I'd consider limiting. And they've not
been able to point to a specific sentence, or set, as problematic. I don't
need their permission to create my own Plan 9 branch, and I don't need
their permission to distribute it.

Where's you're beef? Where in the license do you feel it limits your
choices?

ps. check out the 'unununium' OS, no kernel, all run-time swappable
    modules...there are also several newer GUI's out there in the
    Open Source landscape that might bring a better interface to
    Plan 9.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

ravage@einstein.ssz.com (Jim Choate) writes:

> While I don't necessarily like some of the 'head-in-the-sand ivory-tower
> how-dare-you-question-me' attitude of some aspects of the Plan 9
> developers and their hanger-ons views (but I can live with it w/o any
> major issues - ignore -em - don't need 'em) I fail to see your problem
> with the current Plan 9 license. Yes, the original license (ie $300/use,
> no commercial usage) was problematic, and yes the original 'Open Source'
> license release was worthless as written, they DID re-write it several
> times into its current state. What are your specific views on the current
> license shortfalls?

A full description of the problems is at
http://www.gnu.org/philosophy/plan-nine.html.  Here is a precis:

  You agree to provide the Original Contributor, at its request, with a
  copy of the complete Source Code version, Object Code version and
  related documentation for Modifications created or contributed to by
  You if used for any purpose.

This prohibits people from making private modifications.

  and may, at Your option, include a reasonable charge for the cost of
  any media.

This seems to limit the price that people may charge for an initial
distribution, and might prohibit selling copies for a profit.

  Distribution of Licensed Software to third parties pursuant to this
  grant shall be subject to the same terms and conditions as set forth
  in this Agreement,

When you download the software, Lucent demands that you explicitly
consent to the license.  If that's one of the "terms and conditions",
then it's a problem, because it would mean that before I can send a
copy of the software to my friend Joe, I have to get Joe's explicit
accession to the license.

  1. The licenses and rights granted under this Agreement shall
     terminate automatically if (i) You fail to comply with all of the
     terms and conditions herein; or (ii) You initiate or participate
     in any intellectual property action against Original Contributor
     and/or another Contributor.

This is a huge disaster.  It means that if I want to use Plan 9, I
have to promise never to sue Lucent for any IP violation.  By my using
Plan 9, I therefore would be granting to Lucent the right to use my
own copyright works in complete contravention of the license I have
assigned to them.  That would mean that Lucent could ignore the GPL on
anything I'd authored!  Hardly acceptible.

   You agree that, if you export or re-export the Licensed Software or
   any modifications to it, You are responsible for compliance with
   the United States Export Administration Regulations and hereby
   indemnify the Original Contributor and all other Contributors for
   any liability incurred as a result.

This clause is also a problem.  Laws are automatically in force:
whether mentioned by a license or not--for the people those laws
affect.  But by incorporating the law into the license, you extend the
reach of the law to people that would otherwise not be affected.
Since the laws in question work to limit the right of people to freely
copy the software, they infringe freedom.  You aren't responsible for
what the US government enacts, but by incorporating this into the
license, you force people who would not otherwise have to comply with
the USEAR to start complying with them.

 2.2 No right is granted to Licensee to create derivative works of or
   to redistribute (other than with the Original Software or a
   derivative thereof) the screen imprinter fonts identified in
   subdirectory /lib/font/bit/lucida and printer fonts (Lucida Sans
   Unicode, Lucida Sans Italic, Lucida Sans Demibold, Lucida
   Typewriter, Lucida Sans Typewriter83), identified in subdirectory
   /sys/lib/postscript/font.

These fonts include some Ghostscript fonts which are free, but the
rest doesn't even come close.

Thomas

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[T. Kurt Bond]

"Thomas Bushnell, BSG" <tb+usenet@becket.net> writes:
> ravage@einstein.ssz.com (Jim Choate) writes:
> > While I don't necessarily like some of the 'head-in-the-sand ivory-tower
> > how-dare-you-question-me' attitude of some aspects of the Plan 9
> > developers and their hanger-ons views (but I can live with it w/o any
> > major issues - ignore -em - don't need 'em) I fail to see your problem
> > with the current Plan 9 license. Yes, the original license (ie $300/use,
> > no commercial usage) was problematic, and yes the original 'Open Source'
> > license release was worthless as written, they DID re-write it several
> > times into its current state. What are your specific views on the current
> > license shortfalls?
>
> A full description of the problems is at
> http://www.gnu.org/philosophy/plan-nine.html.  Here is a precis:
>
>   You agree to provide the Original Contributor, at its request, with a
>   copy of the complete Source Code version, Object Code version and
>   related documentation for Modifications created or contributed to by
>   You if used for any purpose.

I'd just like to emphasize that this part of the Plan 9 license seems
to have changed since Richard Stallman first wrote the article.  Now
the Plan 9 license says:

    4.0 MODIFICATIONS

    You agree to provide the Original Contributor, at its request, with a
    copy of the complete Source Code version, Object Code version and
    related documentation for Modifications created or contributed to by
    You if distributed in any form, e.g., binary or source.

http://plan9.bell-labs.com/plan9dist/license.html

Note that now it says the Original Contributor may request the
modifications only if you have distributed the modifications in some
form.

The other four "flaws" listed in Stallman's paper seem to remain.
--
T. Kurt Bond, tkb@tkb.mpl.com

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

On Fri, 9 Nov 2001, T. Kurt Bond wrote:

> >   You agree to provide the Original Contributor, at its request, with a
> >   copy of the complete Source Code version, Object Code version and
> >   related documentation for Modifications created or contributed to by
> >   You if used for any purpose.

Which is effectively different from Open Source results, how again?

Open Source says if you take my code and use it, your code is Open Source
and if you distribute it you have to leave my headers and such in.

The Plan 9 license says if you distribute Plan 9 code and the Plan 9
license holder finds something interesting, then the developer agrees to
provide binary, source, and documentation.

Exactly what is the bitch? Either way the authors code is protected AND
any modifications are protected from close source development. The
implication being that if you were to develop close source the license
holder reserves the right to see what you're doing with their work and
that they get a copy of it. This means they could then release it
publicly if they found it interesting (or even as a matter of course),
thus protecting their investment and yours. It does of course allow them
to develop close source, but since they are the license holder that is a
GOOD thing.

After all, it is the goal not the path we each take that is important in
this context. Free, public code libraries.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

ravage@ssz.com (Jim Choate) writes:

> On Fri, 9 Nov 2001, T. Kurt Bond wrote:
>
> > >   You agree to provide the Original Contributor, at its request, with a
> > >   copy of the complete Source Code version, Object Code version and
> > >   related documentation for Modifications created or contributed to by
> > >   You if used for any purpose.
>
> Which is effectively different from Open Source results, how again?
>
> Open Source says if you take my code and use it, your code is Open Source
> and if you distribute it you have to leave my headers and such in.

Well, you're not quite right here; I'll clear up the subtleties.
Using open source software doesn't impose any obligations on you to
make your own code open source; I assume you must mean something like
copying parts of it into your own code.  (This might seem obvious [as
indeed it should be] but some people seem to think that merely running
the program supposedly incurs some kind of obligations.)

Also, your description seems to be about copylefted software
specifically.  Open source (or its synonym, free software) includes
things which aren't copylefted at all (for example, the X Consortium
or BSD licenses): you can copy parts of that software into your own
program and it doesn't impose any obligations on you to make the
resulting program open source/free software.

So, with the understanding we're talking about the copyleft, and
about actual copying and not just use of the program, you are roughly
right.  If you take part of a GPL'd program, and put it in yours, you
must distribute the combination under the GPL.  That means that if you
give a copy to Fred, you must also give Fred all the rights you had;
a copyleft means you aren't allowed to restrict Fred.

But it imposes no obligation of any kind on you to distribute your
software to John, Mary, or Alice, whether on request or otherwise.
You and Fred are perfectly entitled, under the GPL, to keep your
modifications entirely private if you should so choose.

> The Plan 9 license says if you distribute Plan 9 code and the Plan 9
> license holder finds something interesting, then the developer agrees to
> provide binary, source, and documentation.

That's the difference.  Under a copyleft (like, for example, the GPL),
you have no obligation of any kind to send your changes back to
anyone, whether they ask or not.  Most authors request people to send
back changes, but there is absolutely no obligation to comply.

This confusion is partly caused by the term "open source".  That term,
by avoiding talk of freedom, makes people think the real issue is
whether something is "open" or "visible".  But actually it's perfectly
fine to have secret modifications to GPL'd software that you carefully
guard and only show a special few.  This is true for both "open
source" and "free software"; the two terms define the same set of
programs.  But the former term leads to some confusions because of its
terminology.

Lest you think this is irrelevant, consider that the availability of
free software has been important for those living in repressive
governments, and who have a desire to keep their activities private.
Such a person might well want to modify the software to have some
special samizdat feature, say.  But they need to be allowed to keep
their changes private and not be forced to broadcast them to the
world.  Under the GPL, this right is guaranteed, but under the Plan 9
license, it is not.

This would be enough to keep Plan 9 from counting as free software (or
"open source").  But let me stress, the rule "you can never sue us no
matter what for our IP violations" is a far worse disaster than the
issue of being required to send changes back.

Thomas

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Ralph Corderoy]

Hi Thomas,

> But it imposes no obligation of any kind on you to distribute your
> software to John, Mary, or Alice, whether on request or otherwise.
> You and Fred are perfectly entitled, under the GPL, to keep your
> modifications entirely private if you should so choose.

But Fred is also entitled, if he so chooses, to distribute the program
and the source as far and as wide as he likes without any further
agreement from me.  Am I correct?

Cheers,


Ralph.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

Ralph Corderoy <ralph@inputplus.demon.co.uk> writes:

> > But it imposes no obligation of any kind on you to distribute your
> > software to John, Mary, or Alice, whether on request or otherwise.
> > You and Fred are perfectly entitled, under the GPL, to keep your
> > modifications entirely private if you should so choose.
>
> But Fred is also entitled, if he so chooses, to distribute the program
> and the source as far and as wide as he likes without any further
> agreement from me.  Am I correct?

Certainly.  If you choose to give a copy to Fred, and it's copylefted,
then Fred gets the right to send copies where he wants, when he
wants.  But neither he nor you have any obligation to do so.

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[T. Kurt Bond]

ravage@ssz.com (Jim Choate) wrote in message news:<Pine.LNX.3.96.1011109195430.364R-100000@einstein.ssz.com>...
> On Fri, 9 Nov 2001, T. Kurt Bond wrote:
>
> > >   You agree to provide the Original Contributor, at its request, with a
> > >   copy of the complete Source Code version, Object Code version and
> > >   related documentation for Modifications created or contributed to by
> > >   You if used for any purpose.

I'm not sure how your article ended up with double levels of quoting
for the section from my article, since as far as I can see you were
directly replying to my article.

> Which is effectively different from Open Source results, how again?

I was pointing out that since Stallman originally wrote his article about
the problems he saw with the Plan 9 license that the Plan 9 license had
*changed* slightly, eliminating *that* problem and leaving the others.

As for bitching, *I* wasn't.  The Plan 9 license holders can release (or not)
Plan 9 under any license they want, and I appreciate the fact that they
choose to release it under a license that lets me run the executables and
read the source code.

On the other hand, I don't find it surprising that the GNU Project is
unwilling to use Plan 9 under the condition that they give up their right
to sue the license holders of Plan 9 if those license holders begin using
and distributing GNU software in ways that the GNU software's licenses
prohibit.

So the GNU Project can't use Plan 9, and other people can.  I don't see
that as a serious problem.  An unfortunate minor accident of history,
perhaps, but not a serious problem.
--
T. Kurt Bond, tkb@tkb.mpl.com

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Steve Kilbane]

I've a sneaking suspicion - entirely without evidence - that the
actual contents of the licence are less significant than the impact
of zealots raving about it. If people read the licence for themselves,
and then decide it's not for them, then fine. Though I'd suggest that
the available papers are an even more rewarding read.

steve

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Jim Choate]

On Mon, 12 Nov 2001, Steve Kilbane wrote:

> I've a sneaking suspicion - entirely without evidence - that the
> actual contents of the licence are less significant than the impact
> of zealots raving about it. If people read the licence for themselves,
> and then decide it's not for them, then fine. Though I'd suggest that
> the available papers are an even more rewarding read.

It's more important that you make it out. The reality is that even the
mass of moderates out there are interested in the license. Almost every CD
that I've given away has been accompanied by a 5-10 minute chat about the
license and who makes money off it. Addressing these issues, as unpopular
as they may be, in some manner other than chunking them into some dead-end
mailing list is necessary. Either the objections need to be addressed and
shown to be misunderstandings or else the license may need changing to
make it more acceptable.

That people have questions about license should have a bigger reaction
than making one sad.


 --
    ____________________________________________________________________

             Day by day the Penguins are making me lose my mind.

                                             Bumper Sticker

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------

Re: [9fans] one reason ideas from Plan 9 didn't catch on

[Thomas Bushnell, BSG]

"T. Kurt Bond" <tkb@tkb.mpl.com> writes:

> I'd just like to emphasize that this part of the Plan 9 license seems
> to have changed since Richard Stallman first wrote the article.  Now
> the Plan 9 license says:

This is better, but unfortunately still not quite good enough, I
think.  The problem again is that people might want to distribute
privately, and not have a positive obligation to send copies to
Lucent.  (Whether the recipients of those copies choose to send
something out is their business.)

There seems to be a great fear that private modifications will somehow
take off, and Lucent would get left out of the loop, and not get to
benefit from all the nifty work that would be done.  That's a
reasonable fear; it's exactly the reason the GPL exists.  A copyleft
has the practical effect of allowing you to get changes back almost
always, without actually infringing on freedoms to do it.

A license like the BSD or X Consortium license, by contrast, does
indeed tend to encourage private modifications that never get rolled
back into the original source base, because it's possible with those
licenses to create non-free private modifications.

If the clause were changed into a request instead of a requirement,
it would be no problem at all.

The real killer, of course, is the "you promise never to sue Lucent
for any IP thing" clause.  That's so awful by itself that it dwarfs
the other problems in the license.

I know that Rob Pike and the others who pressured Lucent to make the
code as free as it is were trying to make it more available and to
make it free software.  But the result seems as if they weren't really
aware of what the community actually wants in free sofware, and so
they failed to extract from the lawyers what would actually work.

If there's another go round with the Lucent lawyers (and I certainly
hope there will be), it might be useful to invite Eric Raymond or
someone else of that ilk to help with the process to avoid this kind
of problem.