Re: [9fans] micro vs monolithic kernels

Re: [9fans] micro vs monolithic kernels

[Russ Cox]

 	one question does anyone find bell labs' documentation a poor disgrace to
	software engineering's design process.

if we knew what it was going to look like when
we finished, what would be the point of building it?

more seriously, we're all happy to help you get going,
but setting up a plan 9 file or cpu server is not like
setting up a toaster.  it's not just going to work.
you need to get used to the way the system works so you
can figure out what went wrong when things do go wrong.

you wouldn't expect to set up a full-blown windows nt
file server in a few hours and have it work.  you wouldn't
(or at least shouldn't) expect to sit down with the red hat
box and have a linux system completely ready to go in a
few hours.  because plan 9 is a research system while those
are commercial systems, you should expect even less in terms
of `works right out of the box'.

that's not to say that it actually is harder to install than
windows or linux.  i think installing a terminal is actually
much easier in plan 9 than in the various linux distributions
i've used.  there's no similar program to lead you through
installing a cpu server, and certainly not one to lead you
through installing a file server.

if you go at it with the right frame of mind, plan 9
is a lot of fun and quite pleasant to use.
it _will_ be frustrating at times (especially when
setting up a file server), but in general those
times are few and far between, and it's more
rewarding than frustrating.

russ

Re: [9fans] micro vs monolithic kernels

[Lyndon Nerenberg]

>>>>> "Jim" == Jim Choate <ravage@ssz.com> writes:

    Jim> You guys should work in a 'production' environment, you're
    Jim> getting flabby around your pre-frontals...

No, you should run 'production' software in your 'production' environment.
We aren't running a 'production' environment here. And that's a Good Thing.

--lyndon

Re: [9fans] micro vs monolithic kernels

[William Josephson]

On Mon, Apr 09, 2001 at 04:52:23PM -0500, Jim Choate wrote:
> On average a MS or Linux box takes between 2-3 hours to config once the

> You guys should work in a 'production' environment, you're getting flabby
> around your pre-frontals...

It is amusing to hear Microsoft Windows and Linux labeled
'production' quality software.

 -WJ

Re: [9fans] micro vs monolithic kernels

[Jim Choate]

On Mon, 9 Apr 2001, Russ Cox wrote:

> you wouldn't expect to set up a full-blown windows nt
> file server in a few hours and have it work.  you wouldn't
> (or at least shouldn't) expect to sit down with the red hat
> box and have a linux system completely ready to go in a
> few hours.

I'll have to disagree. My day job is taking GA code for a very large
software/hardware company and testing it on new OS'es as they come out
the door. I manage a group of 5 engineers who spend their week doing about
25-50 OS loads a week and then running the resultant through an automated
testsuite.

On average a MS or Linux box takes between 2-3 hours to config once the
binaries are installed and the system rebooted. I can have a linux box up
and running (sendmail, bind, majordomo, etc.) up and running in under two
hours myself (and have been hitting that target for several years now).
This doesn't include kernel compile time.

So, trying to set the 'base line' standard to install and config a box
outside of 8 hours (a regular work day) is being unreasonable. It should
take x number of hours to setup networking, name resolution, MTA, etc. The
process should be scripted as none of these apps should have ANY hardware
dependency at all.

Saying that your OS won't allow one to configure these base services in a
reasonable and repeatable amount of time is a cop-out.

You guys should work in a 'production' environment, you're getting flabby
around your pre-frontals...

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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

Re: [9fans] micro vs monolithic kernels

[Jim Choate]

On Mon, 9 Apr 2001, Lyndon Nerenberg wrote:

> >>>>> "Jim" == Jim Choate <ravage@ssz.com> writes:
>
>     Jim> You guys should work in a 'production' environment, you're
>     Jim> getting flabby around your pre-frontals...
>
> No, you should run 'production' software in your 'production' environment.
> We aren't running a 'production' environment here. And that's a Good Thing.

I do. I use one of my companies GA products as the test harness of my
automated system (the only team in the entire company to do so I might
add). We use a 'customer like' environment and to make sure it is we
actively work on resolving Crit-Sit's and Sev 1's the rest of the company
has failed to resolve.

In the two years I've managed the group we've hit every time target, come
in under budget, increased the volume of work executed each year (with a 2
person decrease in staff mind you) by a full order of magnitude. If you
look at my groups total performance in the last 24 months we've hit 1000%
improvement (and we ain't done yet).

An OS is meant to be used. Any environment is a 'production' environment
from a OS system admin perspective. Making a OS a chinese puzzle to solve
isn't doing anyone a service. It's just hard-headed self-congragulatory
mental mastrubation.

Computers are SUPPOSED to take the drudgery out of ones life...

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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

Re: [9fans] micro vs monolithic kernels

[Jim Choate]

On Mon, 9 Apr 2001, William Josephson wrote:

> On Mon, Apr 09, 2001 at 04:52:23PM -0500, Jim Choate wrote:
> > On average a MS or Linux box takes between 2-3 hours to config once the
>
> > You guys should work in a 'production' environment, you're getting flabby
> > around your pre-frontals...
>
> It is amusing to hear Microsoft Windows and Linux labeled
> 'production' quality software.

I don't believe I used the term 'production quality' once. I said they
were used in a 'production environment' and they are.

Find somebody else to mis-quote to use to grind your personal issues.

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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

Re: [9fans] micro vs monolithic kernels

[Mike Haertel]

>you wouldn't expect to set up a full-blown windows nt
>file server in a few hours and have it work.  you wouldn't
>(or at least shouldn't) expect to sit down with the red hat
>box and have a linux system completely ready to go in a
>few hours.  because plan 9 is a research system while those
>are commercial systems, you should expect even less in terms
>of `works right out of the box'.

I routinely set up fully configured FreeBSD boxes (using the
distributed 4.2 cdrom, and manually editing local configuration
into /etc) in about 25 minutes: 15 minutes to boot off the CD,
partition the disk, make a file system, and extract a pretty complete
set of distribution binaries and sources, and 10 minutes to edit
files in /etc.  It almost takes longer to set up the boxes and hook
up all the cables and the network, than to install the OS.

Plan 9 PC terminals are about as easy, but somewhat slower because
of pathetic file system performance when extracting large bunches
of files.  I haven't tried to set up a full Plan 9 environment with
auth, cpu, and file servers yet.

When I was at Intel, I made a custom FreeBSD boot floppy that Unix-clueless
coworkers could use to set up a full configured system in a completely
automated fashion.  It made intelligent decisions that allowed it to
tolerate a variety of hardware configurations, and installed the bits
over the network.  It typically took 10 minutes from booting the
floppy on an empty machine, to rebooting the fully configured system.

The IT guys were amazed.  My friends and I were amazed that they
were amazed.  For about two years the network of FreeBSD boxes that
we ran ourselves was our favorite software development platform.
A lot of Pentium 4 architecture work was done on that homebrew setup.
Eventually the IT guys moved from NT to Linux and we stopped maintaining
our own environment.

Re: [9fans] micro vs monolithic kernels

[William Josephson]

On Mon, Apr 09, 2001 at 05:10:04PM -0500, Jim Choate wrote:

> > It is amusing to hear Microsoft Windows and Linux labeled
> > 'production' quality software.
>
> I don't believe I used the term 'production quality' once. I said they
> were used in a 'production environment' and they are.
>
> Find somebody else to mis-quote to use to grind your personal issues.

A worthy attempt at a flame.  In any event, the point remains that you
are comparing apples and oranges: of course a Linux install is trivial
if you've done it many times over, but to claim that, for instance,
Red Hat installations are trouble-free is completely bogus.  Red Hat's
in particular have been getting worse with each release.  Even Windows
installations aren't much better -- you are completely out of luck if
anything goes wrong.  I'd love to see your reaction to the
installation process for some other research systems I've worked on.

 -WJ

Re: [9fans] micro vs monolithic kernels

[Lyndon Nerenberg]

>>>>> "Jim" == Jim Choate <ravage@einstein.ssz.com> writes:

    Jim> Computers are SUPPOSED to take the drudgery out of ones
    Jim> life...

Doing 15/30/whatever_speed_thats_faster_than_the_other_guy minute
installations of NT/Linux/Solaris/* by rote pretty much _is_ the
definition of drudgery, in my books. Having to engage my brain to get
Plan9 installed isn't drudgery, it's stimulating.

Catering to the clueless merely attracts them. Why would we want
to do that? (I should probably point out that the preceeding
question was rhetorical ...)

--lyndon

Re: [9fans] micro vs monolithic kernels

[Dan Cross]

In article <Pine.LNX.3.96.1010409164128.14587X-100000@einstein.ssz.com> you write:
>You guys should work in a 'production' environment, you're getting flabby
>around your pre-frontals...

Well, the point of Plan 9 is to be a research OS, not a system for
production.

If the Plan 9 folks at Bell Labs started to spend a lot of time
polishing up distributions and writing installation documentation, then
they'd have little time to push the state of the art.  *That* would be
a real shame....

Just a speculation, but I'd guess that someone like Russ can, given the
correct hardware components, set up a fileserver/CPU server combo plus
a few terminals in a known and repeatable amount of time.

	- Dan C.

(ps- I find the Plan 9 documentation refreshingly good.  Indeed,
sometimes amazingly good.)

Re: [9fans] Re: micro vs monolithic kernels

[Lyndon Nerenberg]

>>>>> "Jim" == Jim Choate <ravage@einstein.ssz.com> writes:

    Jim> What drudgery, we start our scripts and away it goes. There
    Jim> is a reason we have a foosball table in the middle of our
    Jim> work area...;)

Finally! you're starting to make some sense :-)

[9fans] Re: micro vs monolithic kernels

[Jim Choate]

On Mon, 9 Apr 2001, Lyndon Nerenberg wrote:

> >>>>> "Jim" == Jim Choate <ravage@einstein.ssz.com> writes:
>
>     Jim> Computers are SUPPOSED to take the drudgery out of ones
>     Jim> life...
>
> Doing 15/30/whatever_speed_thats_faster_than_the_other_guy minute
> installations of NT/Linux/Solaris/* by rote pretty much _is_ the
> definition of drudgery, in my books. Having to engage my brain to get
> Plan9 installed isn't drudgery, it's stimulating.

What drudgery, we start our scripts and away it goes. There is a reason we
have a foosball table in the middle of our work area...;)

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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

Re: [9fans] micro vs monolithic kernels

[Jim Choate]

On Mon, 9 Apr 2001, Dan Cross wrote:

> Well, the point of Plan 9 is to be a research OS, not a system for
> production.

You mean ...was...

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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

Re: [9fans] micro vs monolithic kernels

[Jim Choate]

On Tue, 10 Apr 2001, Steve Kilbane wrote:

> Jim wrote:
> > An OS is meant to be used. Any environment is a 'production' environment
> > from a OS system admin perspective. Making a OS a chinese puzzle to solve
> > isn't doing anyone a service. It's just hard-headed self-congragulatory
> > mental mastrubation.
>
> I don't understand the problem, here. Jim, you *know* what Plan 9's
> history is. Yes, it was supposed to be used, but to find out how well
> the ideas behind it work. More specifically, use != install, so if the
> effort goes into the day-to-day use at the expense of a bumpy install,
> well, that's the trade-off for a research group.

Silly goose. !install == !use.

It's damn hard to 'use' a system if it isn't 'installed'.

Talk about specious distinctions.

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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

Re: [9fans] micro vs monolithic kernels

[Dan Cross]

In article <Pine.LNX.3.96.1010409181024.14587r-100000@einstein.ssz.com> you write:
>
>On Mon, 9 Apr 2001, Dan Cross wrote:
>
>> Well, the point of Plan 9 is to be a research OS, not a system for
>> production.
>
>You mean ...was...

No, I mean is.  I believe that Bell Labs is still using it for
research, and that's their bread and butter.  Just because they
let other people have it doesn't mean they stop using it for
their own purposes and go into support mode.  Nor should they.

	- Dan C.

Re: [9fans] micro vs monolithic kernels

[Steve Kilbane]

Being on UK time, I'm going to be somewhat disappointed if tomorrow morning,
I find 9fans full of "well, I've installed on a weirder system than you"
posturing.

Jim wrote:
> An OS is meant to be used. Any environment is a 'production' environment
> from a OS system admin perspective. Making a OS a chinese puzzle to solve
> isn't doing anyone a service. It's just hard-headed self-congragulatory
> mental mastrubation.

I don't understand the problem, here. Jim, you *know* what Plan 9's
history is. Yes, it was supposed to be used, but to find out how well
the ideas behind it work. More specifically, use != install, so if the
effort goes into the day-to-day use at the expense of a bumpy install,
well, that's the trade-off for a research group.

steve

Re: [9fans] micro vs monolithic kernels

[Steve Kilbane]

> It's damn hard to 'use' a system if it isn't 'installed'.

True. But in general, a system is installed by a single individual over
a handful of hours, and then used by from one to thousands of individuals,
over periods that may last for years. Plan 9 isn't impossible to install,
and evidently many people have reached the end of the installation process,
and are using it. Therefore, evidently, it's worth paying attention to this
ratio.

 From your other posts, it sounds like you're working on solving the
installation problems, though. Very nice, but it would have been nicer still
if you could have just said so, without attacking. The Bell folks are always
open and forthcoming about where their efforts fall short, be it because of
lack of resources, or lack of vision.

> Talk about specious distinctions.

About as specious as writing code that's coherent enough to be read by
someone else, or as putting coder/decoder complexities into the coder so
that the decoder just needs to stream. "Done once" versus "done lots" is
a pretty damn important difference.

When resources are limited, choices have to be made about where one
puts the effort. Indulge my curiosity, for a moment: if you'd being doing
it, Jim, which part of Plan 9 would you have dropped (i.e. not designed or
implemented) so that the time could be spent on the installation process?

steve

Re: [9fans] micro vs monolithic kernels

[Douglas A. Gwyn]

Jim Choate wrote:
> Computers are SUPPOSED to take the drudgery out of ones life...

The guy has a point, but what he misses is that Plan 9
doesn't have a product development staff the size of
Windows'.  They're doing pretty well considering the
limited resources.

Re: [9fans] Re: micro vs monolithic kernels

[Boyd Roberts]

"David Lukes" <davel@luchie-chowchows.demon.co.uk> a crit dans le message
news: 01041016242401.10507@luchie-chowchows.demon.co.uk...
>
> No, we see the bogus dichotomy between so-called programmers and
> so-called software engineers when fly-by-wire aircraft control systems
> override their pilots and crash airliners.


so that's why they never let us work together again?

did we _know_ something _they_ didn't?

--
Boyd Roberts        http://www.insultant.net        boyd@insultant.net

What do you know about surfing, Major?  You're from goddamn New Jersey.

    -- Lt. Colonel Kilgore
Followup-To:
Distribution:
Organization: University of Bath Computing Services, UK
Keywords:
Cc:


--
Dennis Davis, BUCS, University of Bath, Bath, BA2 7AY, UK
D.H.Davis@bath.ac.uk

Re: [9fans] Re: micro vs monolithic kernels

[David Lukes]

<flame-bait>

> Hence we see the segmentation between programmers and software engineers.

No, we see the bogus dichotomy between so-called programmers and
so-called software engineers when fly-by-wire aircraft control systems
override their pilots and crash airliners.

> one question does anyone find bell labs' documentation a poor disgrace to
> software engineering's design process.

The BTL^H^H^H Lucent documentation is far better than the reams of
vacuous bureacratic drivel which accompanies most so-called
"software engineering" projects.

Also, you use a singular, whereas, like religions,
there are many "software-engineering" "disciplines".

The use of the term "software engineering" to imply a degree
of reliability not present in "programming" is totally bogus:
engineering is not an exact art/science/whatever:
buildings and bridges still fall down,
car manufacturers still need product recalls,
and aircraft, both fly-by-wire and traditional, still fall out of the sky.

</flame-bait>

	Dave.

Re: [9fans] Re: micro vs monolithic kernels

[Douglas A. Gwyn]

jmk@plan9.bell-labs.com wrote:
> give it a rest, the stupid armadillo sig is wearing very thin.

To call an Armadillo stupid is an insult to stupid critters.

Re: [9fans] Re: micro vs monolithic kernels

[Boyd Roberts]

"Dan Cross" <cross@math.psu.edu> a crit dans le message news:
200104100233.WAA03526@augusta.math.psu.edu...
> In article <Pine.LNX.3.96.1010409203706.14587y-100000@einstein.ssz.com>
you write:
>
> >And as to 'it's hard', and you people are supposed to be world class
> >programmers?
>
> This is really rude.  Russ has been nothing but gracious in hearing out
> your thoughts on this matter, and doesn't deserve this sort of
> response.  In fact, none of the Bell Labs folks do....

yeah, russ is something else.  he gets the job done -- politely.

Re: [9fans] Re: micro vs monolithic kernels

[Boyd Roberts]

<jmk@plan9.bell-labs.com> a crit dans le message news:
20010410002812.9B63019AA7@mail.cse.psu.edu...
> give it a rest, the stupid armadillo sig is wearing very thin.

nah, armadildo, surely.

Re: [9fans] Re: micro vs monolithic kernels

[Boyd Roberts]

"Jim Choate" <ravage@einstein.ssz.com> a crit dans le message news:
Pine.LNX.3.96.1010409190541.14587u-100000@einstein.ssz.com...
> I started using Linux in the .12 days. If memory serves the first time I
> ever used Linux I loaded it on a CompuAdd 316NX laptop (a Samsung based
> 316SX machine) and we had it running in less time than a 8 hour shift (we
> were working the 4-midnite shift). There was no reliable networking at
> that point admittedly. Of course I'd already been using computers for more
> than 20 years at the time. I believe it took me longer to get gcc working
> than the OS.


get a grip.  you have forgotten the history.

i hacked ctrl-s/q flow control into the 6th ed on a an '11/34 with a
spare RK05 pack, while unix not much has _really_ changed since 7th
ed, except for the bloat.

linsux: how many ptrace or /proc style interfaces has it got?

where did /proc come from?  8th ed, ~1984.

how many 8th ed licences where issued and where?

'we' had an 8th ed port up in a few weeks in 1985, working at
night 'cos during the day the basser staff/students/us
had to do get some work done during the day.

--
Boyd Roberts        http://www.insultant.net        boyd@insultant.net

What do you know about surfing, Major?  You're from goddamn New Jersey.

    -- Lt. Colonel Kilgore

Re: [9fans] Re: micro vs monolithic kernels

[Andrey A Mirtchovski]

On Mon, 9 Apr 2001, Dan Cross wrote:

> Hmm, there's been a lot said on this.  I'll spare 9fans and myself
> further replies.
>

worst topic drift ever?

when i put the subject in the original message i definitely didn't have this
ending in mind...

Re: [9fans] Re: micro vs monolithic kernels

[Dan Cross]

In article <Pine.LNX.3.96.1010409203706.14587y-100000@einstein.ssz.com> you write:
>The reality is that simply because you don't know of it, doesn't mean it
>isn't.

Right.  See below.

>And as to 'it's hard', and you people are supposed to be world class
>programmers?

This is really rude.  Russ has been nothing but gracious in hearing out
your thoughts on this matter, and doesn't deserve this sort of
response.  In fact, none of the Bell Labs folks do....

There's no mandate which says that they HAVE to share their work with
the rest of the world, and certainly none that says that they have to
spend time packaging it into an easy to swallow pill before doing so.

If you perceive a problem, then continue with pursuing a solution, by
all means.  But don't act as if it should be the rest of the world's
number one priority to help you do so.  Sometimes the problems you see
aren't the ones that matter to other people, and vice versa, as you
yourself have pointed out.

Hmm, there's been a lot said on this.  I'll spare 9fans and myself
further replies.

	- Dan C.

(ps- The Bell Labs folks *are* world class programmers, but that's
evident from their code.)

Re: [9fans] Re: micro vs monolithic kernels

[Dave Iafrate - CSCI/F1997]

Hence we see the segmentation between programmers and software engineers.


On Mon, 9 Apr 2001, William Josephson wrote:

> Date: Mon, 9 Apr 2001 21:45:20 -0400
> From: William Josephson <wkj@eecs.harvard.edu>
> Reply-To: 9fans@cse.psu.edu
> To: 9fans@cse.psu.edu
> Subject: Re: [9fans] Re: micro vs monolithic kernels
>
> On Mon, Apr 09, 2001 at 08:44:09PM -0500, Jim Choate wrote:
>
> > And as to 'it's hard', and you people are supposed to be world class
> > programmers?
>
> I'm sure Russ is too professional to rise to such petty bait, but yes
> he is a world class programmer.  He is also a full-time undergraduate
> with all that that entails.  Why not stop while you are ahead?
>
>  -WJ
>

Re: [9fans] Re: micro vs monolithic kernels

[William Josephson]

On Mon, Apr 09, 2001 at 08:44:09PM -0500, Jim Choate wrote:

> And as to 'it's hard', and you people are supposed to be world class
> programmers?

I'm sure Russ is too professional to rise to such petty bait, but yes
he is a world class programmer.  He is also a full-time undergraduate
with all that that entails.  Why not stop while you are ahead?

 -WJ

Re: [9fans] Re: micro vs monolithic kernels

[Jim Choate]

To be accurate Russ, I wasn't complaining. Simply stating the reality of
the situation.

And, as for 'complaining'. I don't complain, I either remove the irritant
or fix it.

I am working on it via 'Hangar 18' and the half dozen laptops I've now
gone through trying to get it to work (on hardware it's supposed to work
on mind you). Now I'm down to trying to recover some of the 486's I've got
laying around here because the other dozen machines are all busy doing
something constructive (or isn't supported, not that I'd replace my
AmigaDos machines in the first place). ISA EIDE controllers are a bitch to
find in any sort of quantity. Hangar 18 has about 18 subscribers and at
least three of them have systems up and running. I put up the 'Hangar 18
Policies and Process' (tentative) last nite for review. If all goes well
we should have some sort of public 'community' (for lack of a better word)
by the end of the month.

And we don't have Bell Labs paying for our time and resources (like
$400 a month for a dedicated ISDN w/ a C-class address space to use)...

The reality is that simply because you don't know of it, doesn't mean it
isn't.

And as to 'it's hard', and you people are supposed to be world class
programmers?

On Mon, 9 Apr 2001, Russ Cox wrote:

> i think we all agree that it'd be great
> if plan 9 were easier to install, and if
> the vga support were better.
>
> we're not disagreeing with you, but writing an os
> installation program is difficult and time-consuming.
> i spent last may doing almost nothing else.
> given another month of work, i bet an even better
> one could be written, one that handles cpu servers
> and file servers and so on.  but i haven't had time, and
> apparently neither has anyone else, inside bell labs
> or out.
>
> if you're going to do something about it,
> great.  if not, then we're not much better off
> than we were before you complained.

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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

Re: [9fans] Re: micro vs monolithic kernels

[Russ Cox]

i think we all agree that it'd be great
if plan 9 were easier to install, and if
the vga support were better.

we're not disagreeing with you, but writing an os
installation program is difficult and time-consuming.
i spent last may doing almost nothing else.
given another month of work, i bet an even better
one could be written, one that handles cpu servers
and file servers and so on.  but i haven't had time, and
apparently neither has anyone else, inside bell labs
or out.

if you're going to do something about it,
great.  if not, then we're not much better off
than we were before you complained.

russ

Re: [9fans] Re: micro vs monolithic kernels

[jmk]

give it a rest, the stupid armadillo sig is wearing very thin.

[9fans] Re: micro vs monolithic kernels

[Jim Choate]

On Tue, 10 Apr 2001, Matt wrote:

> >What drudgery, we start our scripts and away it goes. There is a reason we
> >have a foosball table in the middle of our work area...;)
>
> is it not your years of experience as a computer administrator that wrote
> those scripts

And others don't have that experience to pass along to new users? Excuse
me but Plan 9 is what 12+ years old and there aren't any significant
automation efforts? Hmmmmm.....

I thank whichever of the Moirai that it's Open Source now and we'll see it
grow instead of stagnate in a 'research' environment.

My motivation had nothing to do with experience. It had to do with
psychology, doing computer maintenance is boring, dull, pointless work
and absolutely in no way, shape, or fashion, what I'm interested in
computers for. I absolutely hate to wash my car or mow my yard too.

> I bet the first time you installed Linux it didn't all happen in 30 minutes

I don't believe I said anything about anything taking '30 minutes'. It
takes somewhere between 2-4 hours depending on which OS we're speaking of.

I started using Linux in the .12 days. If memory serves the first time I
ever used Linux I loaded it on a CompuAdd 316NX laptop (a Samsung based
316SX machine) and we had it running in less time than a 8 hour shift (we
were working the 4-midnite shift). There was no reliable networking at
that point admittedly. Of course I'd already been using computers for more
than 20 years at the time. I believe it took me longer to get gcc working
than the OS.

If you've got a 1'st or 2'nd ed. "Running Linux" laying around look in the
appendix under d/l sites and look in the 512 area code for a system called
'Solar Soyuz Zaibatsu' (ie SSZ)...

> I'm not trying to negate your argument but it seems to me the benchmarks
> you are using are spurious.

What 'benchmarks'? I've simply made the statement that whether an OS is
'research' or 'production' (Don't know about 'spurious' but that
distinction is certainly specious from the sys admin POV - ALL systems
need sys admin attention) anybody with half a lick of sense will share
that experience with others coming down the road later. Well, they will if
they're not lone-gun's or wanting to use it as an IQ test (a very
popular past time apparently)...

Plan 9 is absolutely the most kick-ass OS I've seen and if I were going
to write one it would be quite similar to Plan 9. The only thing standing
in its way now is Plan 9 designers and users.

    ____________________________________________________________________

       To speak algebraically, Mr. M. is execrable, but Mr. G. is
       (x+1)-ecrable.
                                         Edgar Allan Poe

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