[9fans] My view of Plan 9 and it's future

[9fans] My view of Plan 9 and it's future

[Jim]

Hi,

My intention is not to inflame or insite. If you disagree, fine. You're
entitled to your opinion. This is pretty much my final say on this issue
for the time being at least. Thanks for the feedback.

My view is that Lucent/Bell Labs has no intention of any sort of long term
or serious support for Plan 9. I further believe they intentionaly
prohibit commerical use *and* price it out of the general hobby market.
This guaranteeing that it won't be widely encountered.

The technology that is implimented in Plan 9 is more suited for the
Internet than Linux and Win systems. By remaining with these legacy
systems we actualy inhibit technology. That is a bad thing.

When you couple this with ubiquitious computing (a near term reality) the
process/file/i-o model is a wunderkin approach. Just imagine a packet
radio network over regular CB radio for example.

Plan 9's implimentation of crypto at low levels of the network offers
advantages to privacy that unix and Win based systems will never match.

Data havens and other distributed process/file_space applications will
always be inhibited by the unix and Win architectures.

When one looks at the advancement of the Internet with respect to
commercial OS'es and Open Source OS'es it is clear that one of them is a
technology driver, and it ain't the commercial systems. Look at the
increase in general unix skills compared through the 80's and the 90's.
What was the main force? Open Source software caused the entire market to
respond.

So long as Plan 9 is released under a commercial license and the primary
goal is to make money off the OS, instead of as in Open Source where it's
the distribution, training, support, and applications the money is to be
made, it will fail.

When one looks at Plan 9's support of Unicode and the increasing
non-English traffic it offers a solution to a problem that unix and MS
OS'es are only just now beginning to 'get'.

Under the current license Plan 9 will fail. One solution would be to keep
the non-commerical limit and lower the price to something like $99. I
don't believe that will prove an efficient enough strategy. Though it may
increase its use in schools and some more well healed hackers. It could go
with a Open Source license. In which case the development and user
community would most likely work similar to Linux. If Lucent/Bell want to
make money, they could distribute and support their own branded distro.
Would significantly increase the level of support and development over
their current pace.

The distributed process model offers me an enticing capability that
Beowulf just doesn't provide. The concepts of secure distributed BlackNet
data havens, anonymous remailers, public key servers, etc. are all
entirely too realisable with Plan 9 to let the basic ideas go belly up.

In a head to head between Linux and Plan 9, Plan 9 will kick ass and take
names. As users and supporters of Plan 9 our PRIMARY goal should be to
increase the user community of Plan 9. If that means a indy Open Source
initiative then so be it.

    ____________________________________________________________________

            The future is downloading. Can you hear the impact?

                                        O[rphan] D[rift>]
                                        Cyber Positive

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

[9fans] My view of Plan 9 and it's future

[Christopher]

Centuries ago, Nostradamus foresaw a time when Jim Choate would say:
>My intention is not to inflame or insite. If you disagree, fine. You're
>entitled to your opinion. This is pretty much my final say on this issue
>for the time being at least. Thanks for the feedback.

Regardless of intent, your comments _are_ inflammatory and appear
intended to incite flames.

>My view is that Lucent/Bell Labs has no intention of any sort of long
>term or serious support for Plan 9. I further believe they
>intentionaly prohibit commerical use *and* price it out of the
>general hobby market. This guaranteeing that it won't be widely
>encountered.

Fine, that's your view.

I don't think they _care_ all that much about Plan 9, from the
perspective of "Lucent, the umpteen-bazillion dollar company."

As "intellectual assets" go, I would spectulate that Plan 9 weighs in
at the "under $100M mark," which makes its importance in the overall
scheme of Things At Lucent rather small.  (If I were off by a factor
of 10, it would _still_ be pretty small potatos.)

If I'm wrong about that, I expect it is more out of ignorance of the
"true value" on the part of the organization on the part of the PHBs
way up in the organization than any direct intent to "prohibit
commerical use."

Compare to Xerox, whose PARC labs were largely responsible for
inventing such things as Postscript, Ethernet, WIMP GUIs as we know
them, and such.  If you looked at how many Dorado and Star machines
Xerox sold, and pricing, you'd be readily able to make the _same wrong
conclusions_ about Xerox.

>Plan 9's implimentation of crypto at low levels of the network offers
>advantages to privacy that unix and Win based systems will never match.

.... Which misses that UNIX has gotten "retrofitted" with a whole lot
of interesting things over the years ....

>So long as Plan 9 is released under a commercial license and the primary
>goal is to make money off the OS, instead of as in Open Source where it's
>the distribution, training, support, and applications the money is to be
>made, it will fail.

If it was Eric Raymond saying this, people might take the comments
_somewhat_ seriously.  (Others of us would hold our noses and hope
he'd shut up.)

>Under the current license Plan 9 will fail. One solution would be to keep
>the non-commerical limit and lower the price to something like $99.

I think you're under the impression that Lucent has a "Plan 9
Marketing Division."  It doesn't.  Plan 9 is a _research_ OS, and they
are really only "pushing" it at researchers, who have rather different
sets of priorities and values than you seem to be projecting on
them.

It's interesting to see that there seems to be some new activity
surrounding Plan 9; I would speculate that this may be another
evidence of us coming out of the Long Dark Night of OS Research
Pessimism.

In the 1990s, Microsoft bought out various OS research groups, and
spent rather a lot of money making it look like there was little point
to OS research.

I would be entirely unshocked if the higher-ups at Lucent that hold
purse-strings looked at the money and research staff flows, and
concluded that this was Not A Good Time To Deploy Another OS.

The growth of Linux has provided some new interest in UNIX, as well as
getting the market used to the idea that there Might Be Alternatives
To The Microsoft Hegemony.  Which opens up the potential for other
OS research to bear fruit.
--
"Purely applicative languages are poorly applicable." -- Alan Perlis
cbbrowne@hex.net- <http://www.ntlug.org/~cbbrowne/oses.html>

[9fans] My view of Plan 9 and it's future

[Bengt]

> From: forsyth@vitanuova.com
> i'm a bit surprised, because i was fairly sure that kfs stored its metadeta in machine-dependent form

sorry. this was from a 5 year old memory. presumably i managed to install directly from cd to sparc then.
there was sparc binaris on the cd in those days :-)

this time i will need a pc with network card (instead of scsi)...

[9fans] My view of Plan 9 and it's future

[forsyth]

>>with the current (hide at the end of the disk) system that has worked fine.

i'm a bit surprised, because i was fairly sure that kfs stored its metadeta
in machine-dependent form, which ought to confound attempts to read
little-endian pc structures on a big-endian sparc.

[9fans] My view of Plan 9 and it's future

[Bengt]

> From: Will Rose <cwr@crash.cts.com>
>  Are disks moved between machines a lot?

With the PC centric distribution I need to
1 borrow a pc
2 install plan9 onto a scsi disk on the pc
3 compile a sparc system on that disk
4 move the disk to my spac
5 return the pc
6 start using plan9

with the current (hide at the end of the disk) system that has worked fine.
will it still with the new (PC-style) partition table ?

[9fans] My view of Plan 9 and it's future

[Will]

Bengt Kleberg <bengt@softwell.se> wrote:
: In article <8g3orv$2m9f$1@thoth.cts.com>, Will Rose <cwr@crash.cts.com> wrote:

:> Bengt Kleberg <bengt@softwell.se> wrote:
: ....deleted
:> : I really appreciated that Plan9 would 'hide' at the end of the disk,
:> : no matter what OS that had formatted/labeled the disk in question.
:> : If the new release uses PC standard partition table
:> : it should mean that I can no longer move a disk from x86 Plan9
:> : to Sparc Plan9. Or am I wrong?
:>
:> It can stay in the same place on the disk; just update the (PC-style)
:> partition table accordingly, so other (PC) OSes know its there.  OSes
:> that don't use the table won't be affected.
:>
: A disk formatted for use on a Sparc system is not formatted for a (PC) OS.
: It has another kind of table in the beginning of the disk.
: Will Plan9 accept that disk on a PC, or will Plan9 on x86 machines now
: require a (PC-style) partition table?

I see - you're speaking of a machine using Plan 9 only. I was thinking
of machines running multiple OSes.  It looks like a fundamental
incompatibility then.  Are disks moved between machines a lot?  I tend
to copy stuff, rather than physically move disks.


Will
cwr@cts.com

[9fans] My view of Plan 9 and it's future

[Bengt]

In article <8g3orv$2m9f$1@thoth.cts.com>, Will Rose <cwr@crash.cts.com> wrote:

> Bengt Kleberg <bengt@softwell.se> wrote:
....deleted
> : I really appreciated that Plan9 would 'hide' at the end of the disk,
> : no matter what OS that had formatted/labeled the disk in question.
> : If the new release uses PC standard partition table
> : it should mean that I can no longer move a disk from x86 Plan9
> : to Sparc Plan9. Or am I wrong?
>
> It can stay in the same place on the disk; just update the (PC-style)
> partition table accordingly, so other (PC) OSes know its there.  OSes
> that don't use the table won't be affected.
>
A disk formatted for use on a Sparc system is not formatted for a (PC) OS.
It has another kind of table in the beginning of the disk.
Will Plan9 accept that disk on a PC, or will Plan9 on x86 machines now
require a (PC-style) partition table?

[9fans] My view of Plan 9 and it's future

[Tom]

On May 19,  8:29am, Christopher Browne wrote [edited]:
> I expect it is more out of ignorance of the "true value"
> on the part of the PHBs way up in the organization.
                     ^^^
Inside Bell Labs, the most common sort of Pointy Haired Boss,
the sort that's immune to innovation, used to be known as
`people with Bell-Shaped Heads.'  (I'm off topic?
Very well, then, I'm off topic.  I am large, I contain
multitudes.)

--
Tom Duff.  Bet you I can go a week without saying "synergy" or
"value-added."

[9fans] My view of Plan 9 and it's future

[Will]

Bengt Kleberg <bengt@softwell.se> wrote:
:> > Will Rose wrote:
:> > OTOH, I really, really wish the PC version used the standard partition
:> > table in the conventional way...

:> From: "Douglas A. Gwyn" <DAGwyn@null.net>
:> Jim indicated that that would be the case for the next release.

: That is unfortunate. I really appreciated that Plan9 would 'hide' at the end of the disk,
: no matter what OS that had formatted/labeled the disk in question.
: If the new release uses PC standard partition table
: it should mean that I can no longer move a disk from x86 Plan9
: to Sparc Plan9. Or am I wrong?

It can stay in the same place on the disk; just update the (PC-style)
partition table accordingly, so other (PC) OSes know its there.  OSes
that don't use the table won't be affected.


Will
cwr@cts.com

[9fans] My view of Plan 9 and it's future

[Tom]

On May 19,  8:29am, Christopher Browne wrote [edited]:
> I expect it is more out of ignorance of the "true value"
> on the part of the PHBs way up in the organization.
                     ^^^
Inside Bell Labs, the most common sort of Pointy Haired Boss,
the sort that's immune to innovation, used to be known as
`people with Bell-Shaped Heads.'  (I'm off topic?
Very well, then, I'm off topic.  I am large, I contain
multitudes.)

--
Tom Duff.  Bet you I can go a week without saying "synergy" or
"value-added."

[9fans] My view of Plan 9 and it's future

[Bengt]

> > Will Rose wrote:
> > OTOH, I really, really wish the PC version used the standard partition
> > table in the conventional way...

> From: "Douglas A. Gwyn" <DAGwyn@null.net>
> Jim indicated that that would be the case for the next release.

That is unfortunate. I really appreciated that Plan9 would 'hide' at the end of the disk,
no matter what OS that had formatted/labeled the disk in question.
If the new release uses PC standard partition table
it should mean that I can no longer move a disk from x86 Plan9
to Sparc Plan9. Or am I wrong?


Best Wishes, Bengt
===============================================================
Everything aforementioned should be regarded as totally private
opinions, and nothing else. bengt@softwell.se
``His great strength is that he is uncompromising. It would make
him physically ill to think of programming in C++.''

[9fans] My view of Plan 9 and it's future

[Christopher]

Centuries ago, Nostradamus foresaw a time when Jim Choate would say:
>My intention is not to inflame or insite. If you disagree, fine. You're
>entitled to your opinion. This is pretty much my final say on this issue
>for the time being at least. Thanks for the feedback.

Regardless of intent, your comments _are_ inflammatory and appear
intended to incite flames.

>My view is that Lucent/Bell Labs has no intention of any sort of long
>term or serious support for Plan 9. I further believe they
>intentionaly prohibit commerical use *and* price it out of the
>general hobby market. This guaranteeing that it won't be widely
>encountered.

Fine, that's your view.

I don't think they _care_ all that much about Plan 9, from the
perspective of "Lucent, the umpteen-bazillion dollar company."

As "intellectual assets" go, I would spectulate that Plan 9 weighs in
at the "under $100M mark," which makes its importance in the overall
scheme of Things At Lucent rather small.  (If I were off by a factor
of 10, it would _still_ be pretty small potatos.)

If I'm wrong about that, I expect it is more out of ignorance of the
"true value" on the part of the organization on the part of the PHBs
way up in the organization than any direct intent to "prohibit
commerical use."

Compare to Xerox, whose PARC labs were largely responsible for
inventing such things as Postscript, Ethernet, WIMP GUIs as we know
them, and such.  If you looked at how many Dorado and Star machines
Xerox sold, and pricing, you'd be readily able to make the _same wrong
conclusions_ about Xerox.

>Plan 9's implimentation of crypto at low levels of the network offers
>advantages to privacy that unix and Win based systems will never match.

.... Which misses that UNIX has gotten "retrofitted" with a whole lot
of interesting things over the years ....

>So long as Plan 9 is released under a commercial license and the primary
>goal is to make money off the OS, instead of as in Open Source where it's
>the distribution, training, support, and applications the money is to be
>made, it will fail.

If it was Eric Raymond saying this, people might take the comments
_somewhat_ seriously.  (Others of us would hold our noses and hope
he'd shut up.)

>Under the current license Plan 9 will fail. One solution would be to keep
>the non-commerical limit and lower the price to something like $99.

I think you're under the impression that Lucent has a "Plan 9
Marketing Division."  It doesn't.  Plan 9 is a _research_ OS, and they
are really only "pushing" it at researchers, who have rather different
sets of priorities and values than you seem to be projecting on
them.

It's interesting to see that there seems to be some new activity
surrounding Plan 9; I would speculate that this may be another
evidence of us coming out of the Long Dark Night of OS Research
Pessimism.

In the 1990s, Microsoft bought out various OS research groups, and
spent rather a lot of money making it look like there was little point
to OS research.

I would be entirely unshocked if the higher-ups at Lucent that hold
purse-strings looked at the money and research staff flows, and
concluded that this was Not A Good Time To Deploy Another OS.

The growth of Linux has provided some new interest in UNIX, as well as
getting the market used to the idea that there Might Be Alternatives
To The Microsoft Hegemony.  Which opens up the potential for other
OS research to bear fruit.
--
"Purely applicative languages are poorly applicable." -- Alan Perlis
cbbrowne@hex.net- <http://www.ntlug.org/~cbbrowne/oses.html>

[9fans] My view of Plan 9 and it's future

[Douglas]

Will Rose wrote:
> OTOH, I really, really wish the PC version used the standard partition
> table in the conventional way...

Jim indicated that that would be the case for the next release.

[9fans] My view of Plan 9 and it's future

[Richard]

Will Rose writes:

>I _don't_ think it should be Open Source.  I
>value it because it contains a lot of interesting ideas from experienced
>system designers (and it's also fun to play with).  I don't want it given
>over to a larger, less-skilled, group.

If it was Open-Source licensed, then many of the Bell Labs researchers
would continue to work on it.  otherwise, why would Rob Pike be trying
to convince Lucent management to Open Source it?

so, you could insist on using only the releases made or 'blessed' by
Bell Labs researchers.  If you trust those researchers to produce good
code, then surely you can trust them to refrain from putting bad code
written by less-skilled random hackers into their releases.

[9fans] My view of Plan 9 and it's future

[Will]

Douglas A. Gwyn <DAGwyn@null.net> wrote:
: Jim Choate wrote:
:> My view is that Lucent/Bell Labs has no intention of any sort of
:> long term or serious support for Plan 9. ...

: They seem to have been willing to fund its development over many
: years now, with a lot of staff time redirected for a while toward
: the Inferno product development.

:> So long as Plan 9 is released under a commercial license and the
:> primary goal is to make money off the OS, ... it will fail.

: Plan 9 is a *research* platform, not a commercial product!
: I was happy to pay my $350 for the previous snapshot of Plan 9,
: including nice printed manuals, in order to do the "hobby" things
: you claimed it was too pricy for.

:> ... As users and supporters of Plan 9 our PRIMARY goal should be
:> to increase the user community of Plan 9.

: I disagree.  What we want is quality, not quantity.

I would obviously have liked Plan 9 to be cheaper, but I really can't
see it as expensive; and I _don't_ think it should be Open Source.  I
value it because it contains a lot of interesting ideas from experienced
system designers (and it's also fun to play with).  I don't want it given
over to a larger, less-skilled, group.  I'll definitely take quality
above quantity in this case; there are already enough 'quantity' OSes.

OTOH, I really, really wish the PC version used the standard partition
table in the conventional way...


Will
cwr@cts.com

[9fans] My view of Plan 9 and it's future

[Bill]

In article <39220A7F.30C75330@null.net>, "Douglas A. Gwyn" <DAGwyn@null.net> writes:
|> Jim Choate wrote:
|> > So long as Plan 9 is released under a commercial license and the
|> > primary goal is to make money off the OS, ... it will fail.
|>
|> Plan 9 is a *research* platform, not a commercial product!
|> I was happy to pay my $350 for the previous snapshot of Plan 9,
|> including nice printed manuals, in order to do the "hobby" things
|> you claimed it was too pricy for.

And I'll add my Amen right here.  I have already put my boss on alert
that when it is released I want the next one too.  I have serious plans
to get the Plan9 lab up again this summer and a new release might be
just what I need to get students interested in doing projects with it.

bill

--
Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves
bill@cs.scranton.edu     |  and a sheep voting on what's for dinner.
University of Scranton   |
Scranton, Pennsylvania   |         #include <std.disclaimer.h>

[9fans] My view of Plan 9 and it's future

[Douglas]

Jim Choate wrote:
> My view is that Lucent/Bell Labs has no intention of any sort of
> long term or serious support for Plan 9. ...

They seem to have been willing to fund its development over many
years now, with a lot of staff time redirected for a while toward
the Inferno product development.

> So long as Plan 9 is released under a commercial license and the
> primary goal is to make money off the OS, ... it will fail.

Plan 9 is a *research* platform, not a commercial product!
I was happy to pay my $350 for the previous snapshot of Plan 9,
including nice printed manuals, in order to do the "hobby" things
you claimed it was too pricy for.

> ... As users and supporters of Plan 9 our PRIMARY goal should be
> to increase the user community of Plan 9.

I disagree.  What we want is quality, not quantity.

[9fans] My view of Plan 9 and it's future

[Tom]

On May 13,  7:58pm, Jim Choate wrote:
> Subject: [9fans] My view of Plan 9 and it's future
>
> Hi,
>
> My intention is not to inflame or insite. If you disagree, fine. You're
> entitled to your opinion. This is pretty much my final say on this issue
> for the time being at least. Thanks for the feedback.
>
> My view is that Lucent/Bell Labs has no intention of any sort of long term
> or serious support for Plan 9. I further believe they intentionaly
> prohibit commerical use *and* price it out of the general hobby market.

Bah!  Plan 9 pricing is designed to cover costs, nothing more -- there
is no sinister intent.

For $350 you get a site license for `you or an organization of which
your are a member or employee.'  You and all your friends can go
in together, start a Plan 9 club and chip in a few dollars each.

--
Tom Duff.  That's music in a nutshell.

[9fans] My view of Plan 9 and it's future

[Tom]

On May 13,  7:58pm, Jim Choate wrote:
> Subject: [9fans] My view of Plan 9 and it's future
>
> Hi,
>
> My intention is not to inflame or insite. If you disagree, fine. You're
> entitled to your opinion. This is pretty much my final say on this issue
> for the time being at least. Thanks for the feedback.
>
> My view is that Lucent/Bell Labs has no intention of any sort of long term
> or serious support for Plan 9. I further believe they intentionaly
> prohibit commerical use *and* price it out of the general hobby market.

Bah!  Plan 9 pricing is designed to cover costs, nothing more -- there
is no sinister intent.

For $350 you get a site license for `you or an organization of which
your are a member or employee.'  You and all your friends can go
in together, start a Plan 9 club and chip in a few dollars each.

--
Tom Duff.  That's music in a nutshell.

[9fans] My view of Plan 9 and it's future

[Jim]

On Sat, 13 May 2000 /usr/rsc/lib/from.eecs@plan9.bell-labs.com wrote:

> You are railing against a license that was written
> in 1995.

And could have been changed during any year since.

> The Internet hadn't taken off,

You must have been living in a different city than I was.

> Eric Raymond hadn't even started to think about writing
> fetchmail, Linux was not much more than a toy,

I've used Linux since 1992 (0.42). In 1995 I'd already been using Linux
for business apps for over a year. Linux was far more than a toy in 1995.

> and companies giving large pieces of software away
> for free was not as socially blest as it is today.

Tell that to FSF and the host of GNU tools that were available.

> In 1995, AT&T was simply not ready or willing to give
> things away for free.

That was then, this is now.

[Remaining rant deleted]

    ____________________________________________________________________

            The future is downloading. Can you hear the impact?

                                        O[rphan] D[rift>]
                                        Cyber Positive

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

[9fans] My view of Plan 9 and it's future

[/usr/rsc/lib/from.eecs]

[Disclaimer: despite the occasional email
from Plan 9 servers, I don't speak for anyone but
myself; in particular, I'm not one of the day-to-day
Bell Labs guys.  I just play there when I'm not in school.]

Perhaps I have been too subtle.  Step back for a second,
and look at this discussion.

You are railing against a license that was written
in 1995.  The Internet hadn't taken off,
Eric Raymond hadn't even started to think about writing
fetchmail, Linux was not much more than a toy,
and companies giving large pieces of software away
for free was not as socially blest as it is today.

In 1995, AT&T was simply not ready or willing to give
things away for free.  $350 is orders of magnitude
cheaper than what it cost to get a UNIX license.  You
observe correctly that such a price does put it
out of the reach of most hobbyists, but it was
a big step forward, and better than not releasing
anything.

There was never any intention of making money off
the distribution; there are enormously bigger cash
sources than the handful of people who bought the
Plan 9 CD before it went out of print.  If there
was an intention of making money off it, it wouldn't
have been given away to universities in 1992.
It was and is a research system, and while commercial
systems have been derived from it, the 1995 distribution
and the upcoming one are about releasing a research
system, not selling an operating system.

The primary goal of the Plan 9 developers over the
last decade has been to construct a system that is
useful and pleasant to use.  They've done a good job.
Making money off of the sale of Plan 9 is just not
a goal.  If the developers had their way, I'd bet
Plan 9 would have been given away for free in 1995.
They want to see the system used outside Bell Labs
as much as, if not more than, anyone else.

I don't know what's in the works as far as a license,
but you'd be very wrong to assume that the plan is
to reuse the 1995 license and price tag.  Two years
ago Rob wrote here that while details were sketchy
(and might still be sketchy for all I know) a release
would likely be over the web and free for non-commercial
use, and he hasn't reneged on that vision.

Continued haranguing against the 1995 license terms
is silly, as is debating what the next license
``should be''.  If, when the distribution happens,
you find the license unacceptable, then by all means
go off and start your own open source OS with the
Plan 9 ideas.  But until then, let's have a little
patience and just wait and see what happens.

Russ