Re: [9fans] fortune-worthy

Re: [9fans] fortune-worthy

[David Presotto]

[-- Attachment #1: Type: text/plain, Size: 100 bytes --]

I'll tell you in a year how they make out.  It's sure to
be a wild ride, I'm looking forward to it.

[-- Attachment #2: Type: message/rfc822, Size: 2434 bytes --]

From: "Skip Tavakkolian" <fst@centurytel.net>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] fortune-worthy
Date: Fri, 19 Dec 2003 19:29:36 -0800
Message-ID: <90e261a38bba4488665a238d92ec829c@centurytel.net>

> No the flavor of the week is a move to Windows servers with
> users getting thin clients that connect to them.  I'ld rather
> write something that talks their thin client protocol (if only
> I can figure out what it is).
>

I don't know how anyone can show a positive ROI for service/server
development and operation in a Windows environment.  The Windows
market -- because of its high maintenance requirements -- has
generated a huge support business ecosystem that benefits from
its complexity and fragility.

Re: [9fans] fortune-worthy

[Skip Tavakkolian]

> No the flavor of the week is a move to Windows servers with
> users getting thin clients that connect to them.  I'ld rather
> write something that talks their thin client protocol (if only
> I can figure out what it is).
>

I don't know how anyone can show a positive ROI for service/server
development and operation in a Windows environment.  The Windows
market -- because of its high maintenance requirements -- has
generated a huge support business ecosystem that benefits from
its complexity and fragility.

Re: [9fans] fortune-worthy

[Skip Tavakkolian]

> I'll tell you in a year how they make out.  It's sure to
> be a wild ride, I'm looking forward to it.

I'll be more than a year.

Re: [9fans] fortune-worthy

[Skip Tavakkolian]

>> I'll tell you in a year how they make out.  It's sure to
>> be a wild ride, I'm looking forward to it.
>
> I'll be more than a year.
It will be more than a year.

Re: [9fans] fortune-worthy

[boyd, rounin]

> Windows is the sticky tar that killed the dinosaurs.  Once you stick a
> finger in, you're not going to get out.

yup, straight out of _the mythical man month_.

RE: [9fans] fortune-worthy

[cej]

Peter A. Cejchan
Lab of Paleobiology and Paleoecology
Institute of Geology
Academy of Sciences
Rozvojova 135
16502 Prague, Czechia
http://www.gli.cas.cz/home/cejchan



-----Original Message-----
From: ron minnich [mailto:rminnich@lanl.gov] 
Sent: Wednesday, December 17, 2003 5:54 AM
To: 9fans@cse.psu.edu
Subject: Re: [9fans] fortune-worthy


On Wed, 17 Dec 2003 okamoto@granite.cias.osakafu-u.ac.jp wrote:

> > However, now, things have been changed, that is, now most of computer 
> > users are naive users, and they don't want to LEARN computers.

...

> Some things never change, and resistance to new technology is one of them. 
> But, sooner or later, better new stuff will beat out older crap stuff. It 
> just always takes too long, and its easy to get discouraged. Plus, Plan 9 
...


http://www3.interscience.wiley.com/cgi-bin/abstract/106568390/ABSTRACT
How history happens, or why the conventional wisdom is always wrong
John L. Casti
Complexity
Volume 8, Issue 6, 2003. Pages 12-16
Copyright (c) 2003 Wiley Periodicals, Inc.

;-)))
++pac.

Re: [9fans] fortune-worthy

[boyd, rounin]

> Secured properly with third party lockdown
> software, it apparently just "works."

yeah, but it only takes a gazillion hz/bytes cpu/ram to do it.

i'd like to see a vax 750 [comet] port of XP ;)

Re: [9fans] fortune-worthy

[bs]

boyd, rounin wrote:
>>seems to be called RDP.
>
>
> Reliable Datagram Protcol?  never implemented.
>
Remote Desktop Protocol

Re: [9fans] fortune-worthy

[okamoto]

> I can't count the number of times
> someone tells me, `I'll send you a text file!' and I get a Word
> document.  It's stupid, but it's reality.

Indeed, also excel 'documents', too.

Kenji

Re: [9fans] fortune-worthy

[boyd, rounin]

> seems to be called RDP.

Reliable Datagram Protcol?  never implemented.

Re: [9fans] fortune-worthy

[boyd, rounin]

> people *off* of it? windows generally looses the TCO war, ...

tso(6)?

Re: [9fans] fortune-worthy

[Dan Cross]

a@9srv.net writes:
> // In fact, they're on the campaign trail again to move the whole
> // company onto an environment that's cheaper for them to support.
>
> and what does that have to do with windows? unless they're moving
> people *off* of it? windows generally looses the TCO war, unless
> you're willing to undersupport your users and have things in a
> perpetually broken state. in which case it can win, because
> moderatly-bad windows admins are cheaper than moderatly-bad unix
> admins.

Windows is the sticky tar that killed the dinosaurs.  Once you stick a
finger in, you're not going to get out.  And the finger that everyone
got stuck with was Microsoft Word.  I can't count the number of times
someone tells me, `I'll send you a text file!' and I get a Word
document.  It's stupid, but it's reality.

	- Dan C.

Re: [9fans] fortune-worthy

[boyd, rounin]

> Isn't this an argument for stronger drawterm effort (or something
> more ambitious)?

i thought that was the whole point; no disk, no fan.  put the servers
somewhere and [draw]term [9P] to 'em.

the [windows] drawterm 'postnote' popup is less than obvious, but
it means that your connection has died.  i have a _bad feeling_ that
the windoze TCP/IP stack sends FINs, to each connection, when
some trivial RJ-45 to [A]DSL cable 'fall on the floor' / removal is
seen.

Re: [9fans] fortune-worthy

[George Michaelson]

> I have a friend who has been doing server/user support for
> windows/linux/novell for about five years.  His latest job
> is for a school system that is windows only.  This past
> weekend he claimed that he's "going to the darkside."  After
> a few moments of awkward silence he explained that XP is
> heads and tails above all other MS operating systems in terms
> of support.  Secured properly with third party lockdown
> software, it apparently just "works."

Our finance package, written to work on Windows 2000, died on XP and on XP with
patches required for security.

yes. XP is much better. XP has some rational behaviours. XP will be firewall on
with SP2. ignore the bouncy gui, the lipgloss, looking at the core engineering
its better, its complex, its huge, but within those compellingly non-better
constraints it is better :-)

but just works is over-stating it. there is much which doesn't just work. IPSEC
to unix for instance. wireless roaming can be fraught. Everybody is badly
affected by Centrino, more Intel wierd shit lockin.

Office is getting way way way way way too complex.

I think M$ trying to End-of-life Win98 is going to be very amusing. I doubt if
this will boost sales in China. pirate XP will persist for much time.

-George

Re: [9fans] fortune-worthy

[Sam]

> and what does that have to do with windows? unless they're moving
> people *off* of it? windows generally looses the TCO war, unless
> you're willing to undersupport your users and have things in a
> perpetually broken state. in which case it can win, because
> moderatly-bad windows admins are cheaper than moderatly-bad unix
> admins.

I have a friend who has been doing server/user support for
windows/linux/novell for about five years.  His latest job
is for a school system that is windows only.  This past
weekend he claimed that he's "going to the darkside."  After
a few moments of awkward silence he explained that XP is
heads and tails above all other MS operating systems in terms
of support.  Secured properly with third party lockdown
software, it apparently just "works."

Re: [9fans] fortune-worthy

[David Arnold]

-->"David" == David Presotto <presotto@closedmind.org> writes:

  David> No the flavor of the week is a move to Windows servers with
  David> users getting thin clients that connect to them.  I'ld rather
  David> write something that talks their thin client protocol (if
  David> only I can figure out what it is).

seems to be called RDP.

  David> Vnc is a step in the right direction, though the windows
  David> server for it sucks.

www.rdesktop.org




d

Re: [9fans] fortune-worthy

[David Presotto]

[-- Attachment #1: Type: text/plain, Size: 544 bytes --]

No the flavor of the week is a move to Windows servers with
users getting thin clients that connect to them.  I'ld rather
write something that talks their thin client protocol (if only
I can figure out what it is).

In general though, if your premise is that I always want to use
windows and sometimes plan 9, it is an argument for a stronger
drawterm.  Mine is that I want to live on plan9 and sometimes
use windows.  I'ld rather the inverse of drawterm.  Vnc is
a step in the right direction, though the windows server for
it sucks.

[-- Attachment #2: Type: message/rfc822, Size: 2632 bytes --]

From: "Skip Tavakkolian" <fst@centurytel.net>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] fortune-worthy
Date: Wed, 17 Dec 2003 23:57:56 -0800
Message-ID: <577fd0c726f326714323074c32cead74@centurytel.net>

>> (Hard to believe that the birthplace of Unix will be forced by management
>> to use Windows!)
>
> Research won't be forced to live on it.  However we already are forced to
> use it to communicate with those who don't really care what the solution
> is, which is most of the company.  Also, if we want to solve peoples'
> problems within the company, we have to do it in their context even if
> our initial experiments are on Plan 9.

Isn't this an argument for stronger drawterm effort (or something
more ambitious)?

BTW, I'm assuming this is referring to Windows for desktop environments
and not for running servers. right?

Re: [9fans] fortune-worthy

[a]

// In fact, they're on the campaign trail again to move the whole
// company onto an environment that's cheaper for them to support.

and what does that have to do with windows? unless they're moving
people *off* of it? windows generally looses the TCO war, unless
you're willing to undersupport your users and have things in a
perpetually broken state. in which case it can win, because
moderatly-bad windows admins are cheaper than moderatly-bad unix
admins.
ア

Re: [9fans] fortune-worthy

[Skip Tavakkolian]

>> (Hard to believe that the birthplace of Unix will be forced by management
>> to use Windows!)
>
> Research won't be forced to live on it.  However we already are forced to
> use it to communicate with those who don't really care what the solution
> is, which is most of the company.  Also, if we want to solve peoples'
> problems within the company, we have to do it in their context even if
> our initial experiments are on Plan 9.

Isn't this an argument for stronger drawterm effort (or something
more ambitious)?

BTW, I'm assuming this is referring to Windows for desktop environments
and not for running servers. right?

Re: [9fans] fortune-worthy

[bs]

> Some things never change, and resistance to new technology is one of them.
IMO resistance to something new or different is one of them.

Re: [9fans] fortune-worthy

[mirtchov]

> disks.  I've thought off and on of porting an Atari 2600
> or NES emulator, and that would certainly open up the
> possibilities.  (Assuming, of course, that you have legal
> access to some ROM images.)

too much assembly code in those things... though the idea is good :)

Re: [9fans] fortune-worthy

[Micah Stetson]

> Not to mention that last time I checked we have only one game, snake.
> Even my old Vic20 had a choice of 8.

Somebody had a pipe dream like game.  I've compiled rogue
under ape and run it in vt (I know, cursor addressing
Berkeley software polluting Plan 9, but it is an
entertaining little game).  I've also got most of a Tetris
clone -- it works fine, it's just missing features.  That
would make four, if I ever posted the source.  And I
imagine others here have a toy or two hidden on their
disks.  I've thought off and on of porting an Atari 2600
or NES emulator, and that would certainly open up the
possibilities.  (Assuming, of course, that you have legal
access to some ROM images.)

Micah

Re: [9fans] fortune-worthy

[David Presotto]

[-- Attachment #1: Type: text/plain, Size: 498 bytes --]

That's an oversimplification of reality but close enough, I guess.
There was a lot of chagrine that we'ld never managed to make money off
of Unix while everyone else did; that was just sour grapes.  There was
a feeling that we should hold any other software (not Plan 9
specificly) a lot closer and try to exploit them before they started
making money for everyone else in the hopes it would give us a leg up
in the pocketing of filthy lucre.  Absent open-source, they might have
been right.

[-- Attachment #2: Type: message/rfc822, Size: 2946 bytes --]

From: ron minnich <rminnich@lanl.gov>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] fortune-worthy
Date: Wed, 17 Dec 2003 10:56:01 -0700 (MST)
Message-ID: <Pine.LNX.4.44.0312171055090.1623-100000@maxroach.lanl.gov>

On Wed, 17 Dec 2003, Charles Forsyth wrote:

> >>(Hard to believe that the birthplace of Unix will be forced by management
> >>to use Windows!)
>
> they slipped up once and Unix use got away from them
> but i'm sure they are careful that will never ever happen again.


well, in '91, someone who shall remain nameless told me just that: that
a lot of people at ATT felt that letting Unix out was a huge mistake, and
that was a mistake they would never repeat with Plan 9. Talk about a bad
call!

ron

Re: [9fans] fortune-worthy

[ron minnich]

On Wed, 17 Dec 2003, Charles Forsyth wrote:

> >>(Hard to believe that the birthplace of Unix will be forced by management
> >>to use Windows!)
>
> they slipped up once and Unix use got away from them
> but i'm sure they are careful that will never ever happen again.


well, in '91, someone who shall remain nameless told me just that: that
a lot of people at ATT felt that letting Unix out was a huge mistake, and
that was a mistake they would never repeat with Plan 9. Talk about a bad
call!

ron

Re: [9fans] fortune-worthy

[Charles Forsyth]

>>(Hard to believe that the birthplace of Unix will be forced by management
>>to use Windows!)

they slipped up once and Unix use got away from them
but i'm sure they are careful that will never ever happen again.

Re: [9fans] fortune-worthy

[Dennis D. Jensen]

Spreading a little FUD are we?

OK. I couldn't help it...

--- Anastasopoulos S <anastas@ceid.upatras.gr> wrote:
>
>
> On Tue, 16 Dec 2003, andrey mirtchovski wrote:
>
> > this one comes from a heated slashdot discussion
> someone had with somebody
> > else:
> >
> > 	"Plan9 is just another Hurd -- a theoretically
> great OS that
> > 	only gets used by a few people. The most it can
> hope for is that
> > 	its best principles are adopted by a real OS."
> >
> > just letting you know where you are: andrey
> >
> > ps: i'll gladly accept suggestions as to what this
> OS may be (inferno is
> > honorably excluded) and promise to summarize it in
> the Wiki :)
> >
> >
>
> Obviously it is not yet ready for prime time. It
> lacks a web brownser, a
> better web server, quicktime, C++( I really like it
> ), GUI, a DBMS etc...
> The core OS is however designed very well and all
> these can be integrated
> very smoothly.

What was the quote? "Fashion is such a abomination
that it must change every 6 months." -- cannot
remember the name of the author.

> Nowadays the tendency is to provide web based
> distributed services and use
> the brownser as the terminal. I can't say if that is
> good or bad but this
> is how business are done today. In a plan9 world we
> would have a data base
> server that would serve something like SQL/, we
> would import it in our
> name space and then use our toolchain to manipulate
> it(cat, echo, awk or
> custom ones). In the other world this is done with
> mozilla, PHP, MySQL.

I wouldn't be to sure. Consider Apple's IPod.

James Robertson of Cincom (smalltalk) has an
interesting blog on the subject:

"Years ago, the Smalltalk development team used a tool
called Scopus (long since chewed up by Siebel) for bug
tracking. Engineering was never happy with that, and
built a homebrew system called MARS (Minimal Action
Request System). It was first deployed back around
1995, right after VisualWave was developed - as a web
application. This was nifty - it was easily accessible
to any of the development staff that needed to get to
the bug system, and ran in any browser. The only
problem was speed - things like Query By Example were
none too speedy when the steps included:

Send the query
Server processes the query
Client browser has to render the query results

For a long time, we just put up with this. Then a few
months ago, I hacked together a simple UI client that
hit the back end (query by AR number only) via a
servlet. Interestingly, getting results displayed
outside a browser was much, much simpler. I integrated
the simple client tool as an internal plugin for
BottomFeeder. A couple of our engineers ran with this
idea, and started working on a more capable client
using Opentalk - our Smalltalk-Smalltalk distribution
framework. Now, using an ssh tunnel or VPN connection,
any of the internal users can query and update the
system much more easily.

This is interesting because it represents a move away
from a browser based interface and over to a smart
client - back to the future, as it were. Using this
approach, using the bug tool is far less painful,
because the roundtrip communication with the server is
so much faster. MS is moving this way in LongHorn; I
wonder how many other people are doing similar
things?"

Web clients only you say? Not necessarily.

> An a final note, i completely disagree that is can
> survive if the other
> OSes adopt its best features. Because we have one
> architecture(x86) that
> means we must have one OS(linux i suppose). Plan9 is
> another design(better
> or not you decide) and i would like to see more new
> OSes or ideas rather
> than see linux take the world.

Have you looked at the number of small
gnu/linux-distributions lately? You can find a mini
gnu/linux dist just for the purpose of multimedia :)
And what about OpenBeOS and it's cousins? Hurd is
progressing although it's only a plan9-clone on the
surface, if there's any similarity at all. Recently
Squeak SmallTalk was put on compact flash card on top
of a modified linux-kernel, and some lispers (SBCL --
Steel Bank Common Lisp) are trying something similar.
A smaller or more specialized environment, while
general enough for common use, seems to be the goal.

One architecture == one OS? It sounds like nonsens to
me. On my stationary mascine there is four OS'es and I
know people who regularly use five different OS'es.

These days more ordinary people come to me for LiveCDs
filled with a small or "polished" gnu/linux and even
some of my technical friends have shifted to Mac OS/X
in hope of a more consistent environment.

I think plan9 still has a place.

Put another way: It doesn't look anymore dead to me
than five years ago. Claims of death usually comes
from people using different systems. It depends on the
community, and plan9 will attract it's share, side by
side with the many gnu/linux distros.

--
Dennis Decker Jensen


__________________________________
Do you Yahoo!?
Protect your identity with Yahoo! Mail AddressGuard
http://antispam.yahoo.com/whatsnewfree

Re: [9fans] fortune-worthy

[Lucio De Re]

On Wed, Dec 17, 2003 at 06:54:42PM +0200, Lucio De Re wrote:
>
> On Wed, Dec 17, 2003 at 09:29:33AM -0500, David Presotto wrote:
> >
> > [ ... ]
> >
> > It is the authentication server used by both Unix and Plan 9 systems
> > so that people can use their netkeys (+ssh for example) to get into the
> > company from the outside internet.
> >
Oops, meant as private mail, excuse the liberties...

++L

Re: [9fans] fortune-worthy

[Lucio De Re]

On Wed, Dec 17, 2003 at 09:29:33AM -0500, David Presotto wrote:
>
> [ ... ]
>
> It is the authentication server used by both Unix and Plan 9 systems
> so that people can use their netkeys (+ssh for example) to get into the
> company from the outside internet.
>
Am I to understand that it is possible to auth to Plan 9 from Unix?
Are such tools broadly available?  PAM, perhaps?

> A better multiuser version of VNC on windows would go a long way to
> making life bearable on that front.
>
Remote desktop <http://www.rdesktop.org> (I think) is one of the
tools I've been looking at (I mentioned it as a client of ASN.1
earlier today) and whereas my priority is SPAM (hence S/MIME - are
you aware that rsa2x509(1) claims to add a DN, but doesn't,
incidentally?  I'll work on the docs once I have something to show)
I wouldn't mind returning to rdesktop if BL have a real need for
it.  On my side, I want to see Plan 9 serve the protocol, so thin
clients can be used unmodified, but I'm nearly ready to tackle the
client.

In case remote desktop means nothing to you (Ron Minnich wasn't
familiar with the concept) it's also known as Terminal Server Client
and originally was Cytrix.

In fact, if time is of the essence to prove a point, I managed to
compile rdesktop 1.10 (I seem to recall) but the graphics module
(that package seems quite elegantly contructed) was way above my
skills level. Still, it was only one file, and not a very large
one at that.

> There's also the constant friction caused by not using the solution of the
> day for embedded systems.  It used to be only Vxworks.  Now its slowly opening
> up to Linux because its the flavor of the week at IBM and Lucent looks at
> IBM as the model of a reincarnated big company.  The GPL scares management
> somewhat which acts as a damping force.

The BSD licence ought to appeal to management more, despite a recent
clash with UCB.  Go figure.  I actually assumed Lucent would be
reluctant to follow in IBM's footsteps.

What would be the official fate of Plan 9 if Bell Labs migrated to
a new platform?

++L

Re: [9fans] fortune-worthy

[mirtchov]

> Not to mention that last time I checked we have only one game, snake.
> Even my old Vic20 had a choice of 8.

MesaGL is the next port in the pipeline.  I really hope we'll be able
to run the basic demos in software rendering.  May have to wait a bit
for faster processors though...

:)

Re: [9fans] fortune-worthy

[John Stalker]

> I think that the main reason plan9 is behind in development is that most
> members of its community are more interested in the core OS(kernel, venti,
> fossil) than in applications. The problem is that is very diffucult to
> build from scratch applications like apache, MySQL, GIMP, OpenGL etc using
> the native plan9 environment(not ape ports) and then compare the results
> with the UNIX world but that would be the ultimate crash test.

Not to mention that last time I checked we have only one game, snake.
Even my old Vic20 had a choice of 8.
--
John Stalker
Department of Mathematics
Princeton University
(609)258-6469

Re: [9fans] fortune-worthy

[John Stalker]

> I think however that its main design goal---easy distribution of services
> and resources and the shared environment it presents to the users---has
> been 'superseded? replaced? ignored?(pick your favorite)' in the real
> world by the web, and that is another discipline.

I disagree somewhat.  Per machine namespaces are really a great nuisance.
I have three and a half machines at home and one at work and I already
feel the strain of keeping everything in sync.  Web interfaces are not
much good for this, and the standard solutions--NSF, rsync, etc.--feel
like the hacks that they are.  I can only imagine what administering
a large network is like.

The problem is, I think, a bit more subtle.  The per process namespace
idea and largely transparent networking are still as good an idea now
as they were originally.  Plan9 doesn't work as seemlessly as I would
like, however, in a network whose topology is continually changing.
My laptop, for example, may be part of my LAN, the internet, my
employer's LAN, or operating stand-alone.  Other OSes are no better
at this, but it seems that this is the sort of thing to which plan9
should be well suited.  Plan B has some interesting ideas in this
area.

Plan9 also has some problems at small scales.  It really wants a
dedicated file server, though not quite as badly as in the old days.

--
John Stalker
Department of Mathematics
Princeton University
(609)258-6469

Re: [9fans] fortune-worthy

[suspect]

On Dec 17, 2003, at 12:41 AM, Russ Cox wrote:
>
> inferno can't run unix c programs.

That's a point, though many would admit for writing new
applications, many would prefer to write them in Limbo
(and know they will be available on other platforms too),
than to write them from scratch in C. The effort to be put into,
say, a web browser (to pick a recent example) is one case.

For already existing applications, there is the C to Limbo
translator, and then there is :

; date
Wed Dec 17 16:01:10 GMT 2003
; bind -a '#*myrmigki' /dev
; cd /dev/myrmigki/
; cat new
; cd 0
; echo -n srecl /srecs/dhrystone.sr > ctl
; tail -f stdout &
; echo -n run > ctl
;
Dhrystone Benchmark, Version 2.1 (Language: C)
Program compiled without 'register' attribute
Execution starts, 65536 runs through Dhrystone
...

Which lets you run compiled C programs, albeit not as
fast in as JIT compiled Limbo on Dis, over Inferno.

But what do I care, my livelihood depends neither on
Inferno nor Plan 9, and I'm neither a zealot nor a bigot :-)


cheers,



On Dec 17, 2003, at 12:41 AM, Russ Cox wrote:

>> Given how little manpower is left at the labs, unrealistic
>> as it might seem to some, I'd be glad to see the two systems
>> merged, for the sake of both surviving. The silly thing is, this
>> is already happening to a certain extent with Inferno 4e picking
>> up 9P2000, and slivers of  factotum, venti etc., as well as
>> drivers. Why duplicate the effort ? Why not take the best
>> of both worlds ?
>
> inferno can't run unix c programs.
>

Re: [9fans] fortune-worthy

[David Presotto]

[-- Attachment #1: Type: text/plain, Size: 856 bytes --]

>
> (Hard to believe that the birthplace of Unix will be forced by management
> to use Windows!)

Research won't be forced to live on it.  However we already are forced to
use it to communicate with those who don't really care what the solution
is, which is most of the company.  Also, if we want to solve peoples'
problems within the company, we have to do it in their context even if
our initial experiments are on Plan 9.  To boldly go, we can use Plan 9.
If we want anyone in our corporate world to follow, we have (or
someone else has) to figure out how to use our solutions in the
other contexts.

For most people, a keyboard and a manual are both fearful things to be
avoided at all cost.  Windows does a pretty good job of making an interface
that can be more or less guessed at by trial pointing and clicking and
lots of prompting.

[-- Attachment #2: Type: message/rfc822, Size: 3033 bytes --]

From: Brantley Coile <bwc@borf.com>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] fortune-worthy
Date: Wed, 17 Dec 2003 09:52:50 -0500
Message-ID: <05765b085429865f54c42d315b473a94@borf.com>

	There's also the constant friction caused by not using the solution of the
	day for embedded systems.  It used to be only Vxworks.  Now its slowly opening
	up to Linux because its the flavor of the week at IBM and Lucent looks at
	IBM as the model of a reincarnated big company.  The GPL scares management
	somewhat which acts as a damping force.

A problem with using Linux in embedded systems, and this would be true of any
large source-only system, is that while it give the developers ready made tools,
it also limits their solution to those tools.  I watched one company use Linux
pretty closely and it worked, but the solution is couched very high up.
My first Embedded Unix product was shipped in 1988.  It was based on V7 and
was only 70K or so of kernel and a couple of small programs.  The kernel was
adapted to the applicaion.  Worked really well.  I recently looked into
using Linux for an embedded system, but that didn't work out.  It's not
documented.  I found it frustrating.  Dropped it and went back to my
own small kernels and just wrote more code, which I did faster than
I could figure out how to get Linux to do what I wanted.

Observation: A sufficently large amount of source == no source at all.

(Hard to believe that the birthplace of Unix will be forced by management
to use Windows!)

Brantley

Re: [9fans] fortune-worthy

[Brantley Coile]

	There's also the constant friction caused by not using the solution of the
	day for embedded systems.  It used to be only Vxworks.  Now its slowly opening
	up to Linux because its the flavor of the week at IBM and Lucent looks at
	IBM as the model of a reincarnated big company.  The GPL scares management
	somewhat which acts as a damping force.

A problem with using Linux in embedded systems, and this would be true of any
large source-only system, is that while it give the developers ready made tools,
it also limits their solution to those tools.  I watched one company use Linux
pretty closely and it worked, but the solution is couched very high up.
My first Embedded Unix product was shipped in 1988.  It was based on V7 and
was only 70K or so of kernel and a couple of small programs.  The kernel was
adapted to the applicaion.  Worked really well.  I recently looked into
using Linux for an embedded system, but that didn't work out.  It's not
documented.  I found it frustrating.  Dropped it and went back to my
own small kernels and just wrote more code, which I did faster than
I could figure out how to get Linux to do what I wanted.

Observation: A sufficently large amount of source == no source at all.

(Hard to believe that the birthplace of Unix will be forced by management
to use Windows!)

Brantley

Re: [9fans] fortune-worthy

[suspect]

I am saddened by the fact that Inferno did not become
the primary research focus after its initial commercial
demise via the inferno BU.

My reason for saying this is, there is more incentive to
develop applications for Inferno than there is for Plan 9,
since those applications could be used on a multitude
of platforms.

It seems to me that with Inferno, it was said "we took all
the lessons we learned from Plan 9 and built a better system",
and then, it was all just discarded (by the primary researchers,
(politely disregarding for the moment, VN)).

Given how little manpower is left at the labs, unrealistic
as it might seem to some, I'd be glad to see the two systems
merged, for the sake of both surviving. The silly thing is, this
is already happening to a certain extent with Inferno 4e picking
up 9P2000, and slivers of  factotum, venti etc., as well as
drivers. Why duplicate the effort ? Why not take the best
of both worlds ?

cheers,



On Dec 17, 2003, at 9:29 AM, David Presotto wrote:

> While I'm happy about links, it doesn't really solve many of my
> problems.

Re: [9fans] fortune-worthy

[David Presotto]

> continue to be used only for specialized domains(Question: how plan9 is
> used as Bell-Labs and for what type of jobs? There isn't much information
> about its real world usage).

At the labs, its used as a development system for approximately 8 people.
It used to be a bigger crowd but Lucent, and to a lesser extent our lab,
has shrunk a lot over the last 3 years.

To a larger community (50 people) it is their main web server (about
100000 hits a day).

It is a mail server to yet a different subset of our lab, using imap
and pop3, as well as to those of us who use plan9 exclusively.  We depend on it
as our DNS server for cs.bell-labs.com and as the DHCP server for a handful
of networks (100 systems or so).  There are currently 3 large file servers,
about 10 CPU servers of various flavors (about half of them multiprocessors
of 2 to 16 cpus), and a few dozen terminals.

It is the authentication server used by both Unix and Plan 9 systems
so that people can use their netkeys (+ssh for example) to get into the
company from the outside internet.

We're using it as the embedded OS in telephony things we do in the lab.
For example, an encrypted home ethernet bridge, and a wireless base station.

And, of course, its still our music jukebox.

Given all that, if we had to make a business case for Plan 9, the company
would probably force us onto Windows for living and Linux for an embedded
OS.  In fact, they're on the campaign trail again to move the whole
company onto an environment that's cheaper for them to support.  We'll
survive that because we do things they can't (like DNS hacks, IP
protocol stack hacking, etc) but they'll take away some of users that
don't hack kernels.  It gets harder and harder to do business with the
rest of the company unless you live 100% of the time on the same Windows
OS that they use.   We see the ratcheting effect of proprietary tools
and formats, especially when the company decides on a single solution and
feeds it to the majority of employees.

While I'm happy about links, it doesn't really solve many of my problems.
It simply can't deal with the (ever changing) Microsoft based web environment
that Lucent has.  I have windows of time when the latest and greatest
mozilla/netcape can but those windows close over time (Firebird is currently
holding its own but I'm already seeing fraying at the edges).
We have a couple of NT and XP systems in our lab just so we can walk over,
log on, and deal with corporate administrative stuff every now and then.
A better multiuser version of VNC on windows would go a long way to
making life bearable on that front.

There's also the constant friction caused by not using the solution of the
day for embedded systems.  It used to be only Vxworks.  Now its slowly opening
up to Linux because its the flavor of the week at IBM and Lucent looks at
IBM as the model of a reincarnated big company.  The GPL scares management
somewhat which acts as a damping force.

Re: [9fans] fortune-worthy

[Brantley Coile]

Every few years I try something new.  I'm typing this on an XP system
using drawterm.  I tried the cygwin stuff (don't get me started!).  I've done
Oberon (which I really liked, but ...), New Linux, FreeBSD, and a years back
SUN with NCD Xterminals.  I always come back to Plan 9, and I'm thankful
that I can.

I don't just smell this way, I'm spoiled.

Brantley

Re: [9fans] fortune-worthy

[Matthias Teege]

Fco.J.Ballesteros <nemo@plan9.escet.urjc.es> writes:

> IMHO, what will happen is that some years from now someone will
> re-invent the technology Plan 9 users were using 20 years before,
> give it a name like Java^H^H^H^H KJT9, and pretend that it's brand
> new technology. At that point, people would probably adopt it.

It happens now. One of my customers rebuild his network with SAN
(fileserver), 20 Windows Terminalservers (CPU servers) and diskless
terminals. Oracle announced better support for "grid computing" in
10g.

Bis dann
Matthias

--
Matthias Teege -- http://www.mteege.de
make world not war

RE: [9fans] fortune-worthy

[a]

// Mine's is MS money. sorry. _o_

hey, no need to apologize. i'd love to get my
hands on their money, too. could fund some
*^[[29~serious* plan 9 development that way. if you've
got a share of it, more power to you! :-)
ア

Re: [9fans] fortune-worthy

[Anastasopoulos S]

On Tue, 16 Dec 2003, andrey mirtchovski wrote:

> this one comes from a heated slashdot discussion someone had with somebody
> else:
>
> 	"Plan9 is just another Hurd -- a theoretically great OS that
> 	only gets used by a few people. The most it can hope for is that
> 	its best principles are adopted by a real OS."
>
> just letting you know where you are: andrey
>
> ps: i'll gladly accept suggestions as to what this OS may be (inferno is
> honorably excluded) and promise to summarize it in the Wiki :)
>
>

I think that the main advantage of Plan9 is that it revived OS research
which was more or less irrelevant as Rob Pike pointed out. Its design and
implementation is very simple and you can try new ideas without having to
break the thick wall(COM, CORBA, Heavy APIs, 1G lines of source code...)
the other OSes happily provide.

Obviously it is not yet ready for prime time. It lacks a web brownser, a
better web server, quicktime, C++( I really like it ), GUI, a DBMS etc...
The core OS is however designed very well and all these can be integrated
very smoothly.

I think however that its main design goal---easy distribution of services
and resources and the shared environment it presents to the users---has
been 'superseded? replaced? ignored?(pick your favorite)' in the real
world by the web, and that is another discipline.

Nowadays the tendency is to provide web based distributed services and use
the brownser as the terminal. I can't say if that is good or bad but this
is how business are done today. In a plan9 world we would have a data base
server that would serve something like SQL/, we would import it in our
name space and then use our toolchain to manipulate it(cat, echo, awk or
custom ones). In the other world this is done with mozilla, PHP, MySQL.

It is my opinion that until Plan9 decides how to approach the web it will
continue to be used only for specialized domains(Question: how plan9 is
used as Bell-Labs and for what type of jobs? There isn't much information
about its real world usage).

An a final note, i completely disagree that is can survive if the other
OSes adopt its best features. Because we have one architecture(x86) that
means we must have one OS(linux i suppose). Plan9 is another design(better
or not you decide) and i would like to see more new OSes or ideas rather
than see linux take the world.

I think that the main reason plan9 is behind in development is that most
members of its community are more interested in the core OS(kernel, venti,
fossil) than in applications. The problem is that is very diffucult to
build from scratch applications like apache, MySQL, GIMP, OpenGL etc using
the native plan9 environment(not ape ports) and then compare the results
with the UNIX world but that would be the ultimate crash test.

Spyros

Re: [9fans] fortune-worthy

[Fco.J.Ballesteros]

[-- Attachment #1: Type: text/plain, Size: 343 bytes --]

IMHO, what will happen is that some years from now someone will
re-invent the technology Plan 9 users were using 20 years before,
give it a name like Java^H^H^H^H KJT9, and pretend that it's brand
new technology. At that point, people would probably adopt it.

My argument to support this claim is the history of CS.

Couldn't resist.

[-- Attachment #2: Type: message/rfc822, Size: 2194 bytes --]

From: okamoto@granite.cias.osakafu-u.ac.jp
To: 9fans@cse.psu.edu
Subject: Re: [9fans] fortune-worthy
Date: Wed, 17 Dec 2003 13:47:26 +0900
Message-ID: <ec94c93f9317b1c8700d2a125647ffab@granite.cias.osakafu-u.ac.jp>

> all the same stupid shit people always say. maybe they're right, maybe 
> wrong. 
> 
> But they all said the same thing about Unix, long ago ...

However, now, things have been changed, that is,
now most of computer users are naive users, and they
don't want to LEARN computers.   So, I don't worry about the
number of Plan 9 users, because it's very natural in these days.
So, I use Plan 9 for developping, and anything other Linux, and
Windows at home.  ☺

Kenji

RE: [9fans] fortune-worthy

[Skip Tavakkolian]

> To have Windows at home has one benefits for us.

Same here. MS Flight Simulator.
Correction, two! the other is VMware (to run a term in)

RE: [9fans] fortune-worthy

[okamoto]

>> To have Windows at home has one benefits for us.
>
> Same here. MS Flight Simulator.
> Correction, two! the other is VMware (to run a term in)

Mine's is MS money. sorry. _o_

Kenji

RE: [9fans] fortune-worthy

[cej]

>>> 	"Plan9 is just another Hurd -- a theoretically great OS that
>>> 	only gets used by a few people. The most it can hope for is that
>>> 	its best principles are adopted by a real OS."

>all the same stupid shit people always say. maybe they're right, maybe 
wrong. 

>But they all said the same thing about Unix, long ago ...

>ron

from the same discussion on slashdot:
I look at a system with plan9 on my desk (currently turned off), I read your answer, and I remember this quote: "The IQ of a crowd is inversely proportional to it's size".

;-)

++pac.

RE: [9fans] fortune-worthy

[okamoto]

> I got rid of Windoze some 3 years ago, still tolerating Linux for tasks I cannot (yet) do
> on Plan 9 (which is my main environment, both at work, and at home. 

To have Windows at home has one benefits for us.
We see how it comes too difficult to deal with when we met unexpected
( not foreseen) problem. ☺

Kenji

Re: [9fans] fortune-worthy

[Russ Cox]

> Given how little manpower is left at the labs, unrealistic
> as it might seem to some, I'd be glad to see the two systems
> merged, for the sake of both surviving. The silly thing is, this
> is already happening to a certain extent with Inferno 4e picking
> up 9P2000, and slivers of  factotum, venti etc., as well as
> drivers. Why duplicate the effort ? Why not take the best
> of both worlds ?

inferno can't run unix c programs.

Re: [9fans] fortune-worthy

[okamoto]

> In the great days of the 70s, I was a card punch/tape jockey. People were 
> pretty naive users then, believe me. 

Now, people who have interest to something intelligent toys can select
Linux or BSDs, where too many of sources are there to play with.
Why thet do so, because more imformations are available for those unices
than Plan 9.   Just image how hard to make Plan 9 network for them.

> Some things never change, and resistance to new technology is one of them. 
> But, sooner or later, better new stuff will beat out older crap stuff. It 
> just always takes too long, and its easy to get discouraged. 

Yes, I agree.

>Plus, Plan 9 
> has this artificial and unnecessary 10-year delay thrown in -- it would 
> have been easier had Plan 9 made it out in '91, as we all hoped it would.

probably most parts of the delay came from License matter of Plan 9...
It has been too expensive to those who has much time but money.☺

Kenji

RE: [9fans] fortune-worthy

[cej]

> However, now, things have been changed, that is,
> now most of computer users are naive users, and they
> don't want to LEARN computers.   So, I don't worry about the
> number of Plan 9 users, because it's very natural in these days. So, I use Plan 9 for
> developping, and anything other Linux, and Windows at home.  ☺

I got rid of Windoze some 3 years ago, still tolerating Linux for tasks I cannot (yet) do
on Plan 9 (which is my main environment, both at work, and at home. Don't care what
mainstream people say...

++pac.

RE: [9fans] fortune-worthy

[cej]

> > 	"Plan9 is just another Hurd -- a theoretically great OS that
> > 	only gets used by a few people. The most it can hope for is that
> > 	its best principles are adopted by a real OS."

> all the same stupid shit people always say. maybe they're right, maybe 
wrong. 

> But they all said the same thing about Unix, long ago ...

... oh, those poor believers in that the "invisible hand of market" always selects "the best standards" -- you'll never convince them. Agree with Ron.
++pac.

Re: [9fans] fortune-worthy

[ron minnich]

On Wed, 17 Dec 2003 okamoto@granite.cias.osakafu-u.ac.jp wrote:

> However, now, things have been changed, that is, now most of computer
> users are naive users, and they don't want to LEARN computers.

In the great days of the 70s, I was a card punch/tape jockey. People were
pretty naive users then, believe me. The expectation then was that unix
was a nice toy, but the real work would always be done on OS/370. Unix
even had to have a JCL command for a while (according to an old V5 man set
I once had) for submitting batch jobs to "real" computers.

Some things never change, and resistance to new technology is one of them.
But, sooner or later, better new stuff will beat out older crap stuff. It
just always takes too long, and its easy to get discouraged. Plus, Plan 9
has this artificial and unnecessary 10-year delay thrown in -- it would
have been easier had Plan 9 made it out in '91, as we all hoped it would.

ron

Re: [9fans] fortune-worthy

[okamoto]

> all the same stupid shit people always say. maybe they're right, maybe 
> wrong. 
> 
> But they all said the same thing about Unix, long ago ...

However, now, things have been changed, that is,
now most of computer users are naive users, and they
don't want to LEARN computers.   So, I don't worry about the
number of Plan 9 users, because it's very natural in these days.
So, I use Plan 9 for developping, and anything other Linux, and
Windows at home.  ☺

Kenji

Re: [9fans] fortune-worthy

[ron minnich]

On Tue, 16 Dec 2003, andrey mirtchovski wrote:

> 	"Plan9 is just another Hurd -- a theoretically great OS that
> 	only gets used by a few people. The most it can hope for is that
> 	its best principles are adopted by a real OS."

all the same stupid shit people always say. maybe they're right, maybe
wrong.

But they all said the same thing about Unix, long ago ...

ron

[9fans] fortune-worthy

[andrey mirtchovski]

this one comes from a heated slashdot discussion someone had with somebody
else:

	"Plan9 is just another Hurd -- a theoretically great OS that
	only gets used by a few people. The most it can hope for is that
	its best principles are adopted by a real OS."

just letting you know where you are: andrey

ps: i'll gladly accept suggestions as to what this OS may be (inferno is
honorably excluded) and promise to summarize it in the Wiki :)