Re: [9fans] hangar 18

Re: [9fans] hangar 18

[Russ Cox]

> What I -do- take exception with is some of the holier than
> though attitudes of -particular- individuals, the head-up-the-ass
> attitude of Lucent and their licenses. There is a reason that I have

No one is 100% happy with the licenses.  But you have to admit that
we've come a long way since $350 in 1995.  Lucent is a big company
with more important things to worry about, and Plan 9 is a small effort
by an ever-dwindling team of people, all of whom are making that effort
in their spare time, on top of whatever it is they get paid to do.
No one is getting paid to work on Plan 9 anymore.

> lab machines named 'pike', 'kernighan', 'ritchie', etc. However, I -do-
> take -great- exception to the attitude of these same people that Plan 9
> should be limited to 'research' applications.

Wrong.  No one has said that Plan 9 should be limited to research
applications.  If you want to make a commercial product out of Plan 9,
that would be just great.  Please do.  However, doing so would be a lot
of effort, and what people -have- said is that they're not interested in
making that effort themselves.  If others want to make the effort, great!
In fact, there are some commercial uses of Plan 9 out there.  If you
want to start the Red Hat of Plan 9, go do it.  I'd love to see that
happen.

> > This is exactly my point.  Why is it competition?
>
> If it isn't then why are you complaining instead of helping?

I have a finite amount of spare time to spend on Plan 9, and
I'm already spending more 100% of that time on it.  I work on
the projects I think are worthwhile, just as you do.  Your web page
says you're interested in replacing the DES authentication.  Now that
factotum is out there, that would be easy.  You could say that I (along
with a few other people that you don't like) got you most of the
way there.  I'm complaining because it bothers me to see you assume
untruths about people based on your own speculation.

> You're clearly too complacent from getting your own way all the time.

If I got my own way all the time, there would be a dropdown box on the
Plan 9 download page where you could choose the license you wanted to
apply to your use of Plan 9, or enter your own.  If I got my own way
all the time, there would be large teams of people getting paid to
write drivers, so I would never have to write another.  If I got my own
way all the time, Plan 9 would interoperate seamlessly with every
file system format and network protocol on the planet.

Don't confuse the way things are with the way anyone would like
them to be.  Don't assume that because someone isn't screaming at the
top of his lungs, he's content with the way things are.

Being confrontational and obnoxious doesn't make Plan 9 any better.
Let's get back to behavior that does.

Russ

Re: [9fans] hangar 18

[Jim Choate]

On Mon, 30 Sep 2002, Russ Cox wrote:

> No one is 100% happy with the licenses.  But you have to admit that
> we've come a long way since $350 in 1995.

And I have on -many- occassions.

It's interesting that you don't seem able to recognize that fact or grant
credit for it...

Have a nice night. This discussion is boring and I have better things to
do with my time.

Adios.


 --
    ____________________________________________________________________

    We don't see things as they are,                      ravage@ssz.com
    we see them as we are.                                   www.ssz.com
                                                  jchoate@open-forge.org
    Anais Nin                                         www.open-forge.org

    --------------------------------------------------------------------

[9fans] VMWare and OS Research

[Jack Johnson]

Russ Cox wrote:
> If I got my own way
> all the time, there would be large teams of people getting paid to
> write drivers, so I would never have to write another.

How much do you think VMWare is going to impact OS research?

I look at the work that's done on the BSDs and Linux because the driver
base is there, but it seems that for people doing lower-level work,
there's a lot to be said for using VMWare if your research or interests
lie somewhere other than driver development, or if nothing else as a
proof-of-concept before diving into driver hell but without supporting
decaying hardware along the way.

Does this seem like a plausible future?

-Jack

Re: [9fans] VMWare and OS Research

[Jim Choate]

On Mon, 30 Sep 2002, Russ Cox wrote:

> > How much do you think VMWare is going to impact OS research?
>
> Anyone building a new x86 OS today would be a fool

So much for tact.


 --
    ____________________________________________________________________

    We don't see things as they are,                      ravage@ssz.com
    we see them as we are.                                   www.ssz.com
                                                  jchoate@open-forge.org
    Anais Nin                                         www.open-forge.org

    --------------------------------------------------------------------

Re: [9fans] VMWare and OS Research

[Russ Cox]

> How much do you think VMWare is going to impact OS research?

Anyone building a new x86 OS today would be a fool
not to start by doing it on top of VMware.  It's just
so much easier than working on real boxes.  The only
reason I can think of not to use VMware is if you were
trying to build something like a microkernel where you
care about every little architecture trick and the
associated speed benefits.  (VMware implements the tricks,
they just might not have the usual costs.)

It's still nice to run on the bare hardware, though.
Compiling is an order of magnitude faster that way,
mainly because there are so many context switches to
kfs and back.  (Using a network file server might
help here quite a bit; I haven't measured that.)

Russ