Re: [9fans] Rant (was Re: Plan9 and Ada95?)

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[okamoto]

What means 'NP', sorry I'm comming from another planet. :-)

Kenji

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Dan Cross]

In article <20011108015655.1AA94199E7@mail.cse.psu.edu> you write:
>What means 'NP', sorry I'm comming from another planet. :-)

P = NP?

Polynomial versus Non-deterministic Polynomial; in other words,
does the algorithm run in a *deterministic* amount of time?  From
theoretical computer science.

	- Dan C.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Boyd Roberts]

It's just not the machine code for windows, it's the DLLs, the ghastly
run-time environment etc.  It is far from a trivial problem.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Boyd Roberts]

    Perl and Python i can see, for sure.

There are many things I like and dislike about Python, but it's as close to
limbo as I'm likely to get, at the moment.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

Andrew Simmons <andrew.simmons@xtra.co.nz> writes:

> It is surely neither polite nor accurate to describe mr pike as stamping his
> foot and saying it isn't fair. I take him to have been expressing a sense of
> frustration and sadness with the doctrinaire and ungenerous attitude of the
> free software ultras. I mean, here you are able to obtain at no cost a
> fascinating and elegant operating system (although admittedly the omission
> of the "-v" flag from "cat" is odd), complete with source code and some of
> the best-written documentation around, and instead of thanking Lucent and
> the Plan 9 team for their astonishing generosity, you complain because the
> system does not satisfy your arcane definition of "free".

Except cost has *nothing to do with it*.  Whether it's hugely
expensive or they pay you to run it (negative cost), it's not relevant
to the definition I was using.  Whether the source is generally
visible is not relevant.  What is relevant is whether people have
certain freedoms.  If Lucent doesn't want people to have those
freedoms, then lots and lots of people aren't going to have interest
in hacking on it.  And why should they?  To improve Lucent's bottom
line?  To enable Lucent to steal software from other people by
ignoring the licenses on it?

I'm sorry that Rob feels miffed; he probably did the level best he
could.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

presotto@closedmind.org writes:

> I wouldn't put GCC at the other end of the spectrum; SUIF owns
> that extremity.

There's no doubt that GCC is not the best C compiler in the world.
But (when user -O), the tradeoffs in GCC are always towards making the
optimizer better at the expense of perhaps taking more compile time.

The non-optimized version is entirely subsidiary; the goal there is to
use parts of the "real" compiler in order to go as fast as possible.
A compiler optimized for fast translation would surely do better.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

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

> I should note here that we tried switching to gcc, but because the
> stupid C++ ABI is so vastly different from compiler to compiler, it
> didn't work with our ODBC libraries.

Yeah, chalk this up to the horrors of C++...  sigh.  It is truly a
disaster that a language has so caught on for which there is no
general ABI around--not even a fairly straightfoward obvious one.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[martin.m.dowie]

"David Gordon Hogan" <dhog@plan9.bell-labs.com> wrote in message
news:20011109221155.BD850199B9@mail.cse.psu.edu...
> > There is the Ada Web Server, which should be trivial to get working on
> > Plan9 once we have an Ada compiler which will target the Plan9 platform.
>
> Yeah, the headlines could read:
>
> Plan 9 Targetted By Military

lol   :-)

....if slightly old fashioned viewpoint - today's Ada users include
the likes of Canal+ (interactive TV systems) to Mondex (electronic
cash systems) to Reuters (news agency).

and with the advent of the GNAT compiler a whole bunch of home
enthusiasts! :-)

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Andrew Simmons]

<RANT>
It is surely neither polite nor accurate to describe mr pike as stamping his
foot and saying it isn't fair. I take him to have been expressing a sense of
frustration and sadness with the doctrinaire and ungenerous attitude of the
free software ultras. I mean, here you are able to obtain at no cost a
fascinating and elegant operating system (although admittedly the omission
of the "-v" flag from "cat" is odd), complete with source code and some of
the best-written documentation around, and instead of thanking Lucent and
the Plan 9 team for their astonishing generosity, you complain because the
system does not satisfy your arcane definition of "free".

I would however mildly disagree with mr pike in his choice of the splits in
the Christian churches as the most apt comparison with the free software
community. To my mind, a better parallel is with the various Judaean
liberation groups in "Monty Python's Life of Brian". The similarity between
the odium heaped on "splitters" by the Judaean People's Liberation Front (or
was it the People's Front for the Liberation of Judaea?), and the odium
heaped on "forkers" by the various free software factions is uncanny.
</RANT>

>
> And the people who might work on Plan 9 but don't, because they are
> committed to free software, are committed to the libre definition, not
> the gratis one.  Stamping your foot and saying it isn't fair won't
> change that.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

forsyth@vitanuova.com writes:

> by `if gnat is structured so as to allow another ... in principle',
> i meant, it has some internal representation of the Ada program that
> is at or near a useful level for code generation, before it produces RTL.
> if it is producing RTL directly from a basic parse tree,
> you'll have your work cut out.

Yes, that's true.  Most GCC front ends (and I'm sure gnat is the same
here) produce nice parse trees and generate code from those.  The
generation of the parse tree is, in principle, not tightly coupled
with anything to do with code generation.

However, the parse trees are usually partial (for example, the idea of
parsing a whole C file at once died a long time ago), and there may be
a lot of niggling little dependencies, since nobody's bothered to try
and separate the parts out that way.

Some people have tried to take the bison parser description files from
GCC and use them for other purposes as well, with varying degrees of
success.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Dan Cross]

In article <15342.13829.659344.232543@nanonic.hilbert.space> you write:
>If the compiler generates fast code, you can always recompile the compiler
>and it will run faster :-).

Yes, but if you believe Proebsting, it'll take 18 years for it to be
twice as fast.  :-)

	- Dan C.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Dan Cross]

In article <200111111120.LAA10681@localhost.localdomain> you write:
>...which you weren't doing. On a more general note, "faster compiler"
>fits in with "faster processor", "bigger disk" and "more memory" as
>excuses for "can't be bothered to design properly." Of course, "compiler
>generates faster/smaller code" also supports it.

True.  Hence the paradoxical result that our code wasn't really any
better for the pain of having to work that way.  You'd think that since
we had to do all this garbage to get it to build correctly, people
would have invested more time in getting things right the first time.
But they didn't.

I think that's a general result of people `growing up' in environments
where development revolves around a edit/compile/test/repeat cycle, but
leaving those environments before they've fully matured.  (Sorry for
the age/maturity metaphor.)  While that model is extremely powerful in
the hands of those with some experience and education in its proper
use, in the wrong hands, it can be disasterous, leading to the, ``just
hack it until it works'' syndrome.

Or maybe the problem was that at said company, doing a build was a good
excuse for going off and doing something else for 45 minutes....

	- Dan C.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[presotto]

There is a trade off between the efficiency of the code and
the speed of the compiler.  Ken went for the latter with
some concessions to the former.  It made sense for us since
we spend most of our time writing code and are somewhat sensitive
to the speed of building.  It also makes it possible to build for
all architectures everytime without really worrying about the time
it takes.  Since we still regularly use multiple architectures,
'mk installall' is the thing I type most whenever I've changed
code.

I wouldn't put GCC at the other end of the spectrum; SUIF owns
that extremity.

However, this shouldn't be held up as a reason for not producing
compilers with better code generation.  There's plenty of room
in the Plan 9 world for a better/more feature rich compiler.
Perhaps dhog will massage GCC enough to make it fill that space.
More likely something else will come along.

The only thing I ask is that the code that comes out correctly
reflects the C that went in.  The rest, to a lesser or greater
extent, is just icing on the cake.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Steve Kilbane]

Dan C:

> ...Or if you have to work on a large C++ project where the compilation
> process, running on a quad processor Sun Ultra Enterprise 450 with 4GB
> of RAM, takes 45 minutes.  And the boneheads working on the project
> have screwed up the build structure so totally that to compile an
> incremental change requires building the entire source, then compile
> time becomes very significant.


<boyd>
But if you're using C++, and you've screwed up the build structure,
then you get what you deserve. No point blaming the speed of the compiler
for those problems...
</boyd>

...which you weren't doing. On a more general note, "faster compiler"
fits in with "faster processor", "bigger disk" and "more memory" as
excuses for "can't be bothered to design properly." Of course, "compiler
generates faster/smaller code" also supports it.

steve

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[paurea]

Steve Kilbane writes:
 > From: Steve Kilbane <steve@whitecrow.demon.co.uk>
 > Subject: Re: [9fans] Rant (was Re: Plan9 and Ada95?)
 > Date: Sun, 11 Nov 2001 01:38:59 +0000
 >
 > In terms of compilation speed versus compiled-code speed, the bias
 > is usually to the latter, 9fans notwithstanding. This is usually
 > the case: extra time spent during the coding phase (whether designing
 > or compiling) pays off in reduced time every time the application is
 > used.

If the compiler generates fast code, you can always recompile the compiler
and it will run faster :-).

--
                 Saludos,
                         Gorka

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Dan Cross]

In article <200111110138.BAA02696@localhost.localdomain> you write:
>But then, this ranking comes from a commercial compiler market, where
>the customer doesn't see anything but the finished product. If you're
>in an environment where you have cause to recompile the entire OS and
>surrounding applications four times a day, I can see how (5) might work
>its way up the list.

...Or if you have to work on a large C++ project where the compilation
process, running on a quad processor Sun Ultra Enterprise 450 with 4GB
of RAM, takes 45 minutes.  And the boneheads working on the project
have screwed up the build structure so totally that to compile an
incremental change requires building the entire source, then compile
time becomes very significant.

I should note here that we tried switching to gcc, but because the
stupid C++ ABI is so vastly different from compiler to compiler, it
didn't work with our ODBC libraries.

Testing on that project was a massive pain.  One would have thought
that that would have made the software have some slightly higher
quality, but oddly enough it didn't.

The really funny thing was that the solution would have been to ditch
C++ in favor of a `scripting' language like python, but management
nixed that idea.  Oh well....

	- Dan C.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Steve Kilbane]

In terms of compilation speed versus compiled-code speed, the bias
is usually to the latter, 9fans notwithstanding. This is usually
the case: extra time spent during the coding phase (whether designing
or compiling) pays off in reduced time every time the application is
used.

If I had to rank the attributes of a compiler, I'd say:

1. Correctness.
2a. Speed of compiled code
2b. Size of compiled code
4. Usefulness of debugging
5. Speed of compilation process.

(2a and 2b vary in order, depending on particular circumstances)

But then, this ranking comes from a commercial compiler market, where
the customer doesn't see anything but the finished product. If you're
in an environment where you have cause to recompile the entire OS and
surrounding applications four times a day, I can see how (5) might work
its way up the list.

As a side-note: the "I don't care about run-time speed if I can run it
soon" approach is one of the reasons why Perl is popular.

steve

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 276 bytes --]

so does lcc, if you're looking for something off Plan 9 that's small,
cleanly written, and portable.  you'll still need to sort out
the assembly and linking phases though.

there, you see: it needn't run on plan 9 for
some of us to like it, it must only be well done.


[-- Attachment #2: Type: message/rfc822, Size: 2020 bytes --]

To: 9fans@cse.psu.edu
Subject: Re: [9fans] Rant (was Re: Plan9 and Ada95?)
Date: Fri, 9 Nov 2001 17:26:24 -0500
Message-ID: <20011109222628.9842219A60@mail.cse.psu.edu>

> There's been a lot of noise about how GCC might be more ugly, or
> poorly constructed, or such.

Translation: some people here have opinions that differ from yours.

> I'm asking whether amidst all that noise
> anyone has bothered to see whether it actually performs its job better
> or worse.  It does seem to me to be an important question in
> evaluating tools which one is actually better at the principal job the
> tool is designed to perform.

GCC is painfully slow.  I really don't care if it produces an executable
that's 5% faster, if you're working in a compile-execute-debug-rewrite
cycle, you want that compile step to complete in a reasonable time.
Plan 9's compiler beats GCC hands down on this one.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Andrey A Mirtchovski]

On Sat, 10 Nov 2001, Matthew Hannigan wrote:

>
> dhog says
> > cycle, you want that compile step to complete in a reasonable time.
> > Plan 9's compiler beats GCC hands down on this one.
>
> numbers?  don't disblieve you, but by how much does it beat it?
> even rought estimates would be nice
>

enough!

p9 compilers are twice as fast as gcc compiling code on same hardware...

gcc code is 5% faster than p9 compiled code on same hardware when it comes
to rendering povray images.


andrey

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Matthew Hannigan]

dhog says
> cycle, you want that compile step to complete in a reasonable time.
> Plan 9's compiler beats GCC hands down on this one.

numbers?  don't disblieve you, but by how much does it beat it?
even rought estimates would be nice

William Josephson wrote:
> True, although my biggest pet peeve with GCC
> is that it generates bogus code for the Alpha
> and mips.

i thought this was fixed the alpha in recent versions
(2.96 and above)
just in time for the alphas demise!

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[William Josephson]

On Fri, Nov 09, 2001 at 05:26:24PM -0500, David Gordon Hogan wrote:
> > I'm asking whether amidst all that noise
> > anyone has bothered to see whether it actually performs its job better
> > or worse.  It does seem to me to be an important question in
> > evaluating tools which one is actually better at the principal job the
> > tool is designed to perform.
>
> GCC is painfully slow.  I really don't care if it produces an executable
> that's 5% faster, if you're working in a compile-execute-debug-rewrite
> cycle, you want that compile step to complete in a reasonable time.
> Plan 9's compiler beats GCC hands down on this one.

True, although my biggest pet peeve with GCC
is that it generates bogus code for the Alpha
and mips.

  -WJ

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[David Gordon Hogan]

> If you take Nt and XP, and remove the last and first letters of
> those names (respectively), you end up with NP.  Hmm, coincidence?
> I think not.

I keep telling everybody that XP is Chi-Rho, but nobody listens...

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[David Gordon Hogan]

> But it's not free software, so I won't.  And there are *many* people
> who share the same ideals.  Those people contribute to systems like
> the GNU/Linux distributions and the various *BSD systems.

And the result is so free, it's largely Design Unencumbered.  <ducks>

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[David Gordon Hogan]

> Let's get off the hobbyhorse of criticising all the alternative
> platforms and actually co-operate with them or show that the Plan
> 9 way is as superior as we make it out to be.

Let's write some code.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[David Gordon Hogan]

> There's been a lot of noise about how GCC might be more ugly, or
> poorly constructed, or such.

Translation: some people here have opinions that differ from yours.

> I'm asking whether amidst all that noise
> anyone has bothered to see whether it actually performs its job better
> or worse.  It does seem to me to be an important question in
> evaluating tools which one is actually better at the principal job the
> tool is designed to perform.

GCC is painfully slow.  I really don't care if it produces an executable
that's 5% faster, if you're working in a compile-execute-debug-rewrite
cycle, you want that compile step to complete in a reasonable time.
Plan 9's compiler beats GCC hands down on this one.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[David Gordon Hogan]

> There is the Ada Web Server, which should be trivial to get working on
> Plan9 once we have an Ada compiler which will target the Plan9 platform.

Yeah, the headlines could read:

	Plan 9 Targetted By Military

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Boyd Roberts]

> ... like mozilla, it carries the leaden weight
> of history around inside it.

Mozilla is a horrible thing.  I've seen how it calls WinSock -- ick.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Boyd Roberts]

> true. but i don't see that as a bad thing. what _would_ be bad would
> Plan 9 developers are not arrogant, rather, they are trying to defend
> themselves and their ideas from the random bunch of people who come here
> and lend their advise and thoughts without being asked for, or without
> really being aware of the system they discuss (it's really mostly a "you're
> out of your element.." issue, to continue throwing quotes from favourite
> movie :)...

Dude:  Yeah man, it really tied the room together -

...

Walter:  Forget it, Donny.  You're out of your element.

:)

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Dan Cross]

In article <20011109074116.B19B219A4D@mail.cse.psu.edu> you write:
>You think Windows runs in polynomial time?

No.  That was the point of the joke.

	- Dan C.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Caffienator]

In article <24569.1005289364@apnic.net>, "George Michaelson"
<ggm@apnic.net> wrote:

> Apache is a bad idea because like mozilla, it carries the leaden weight
> of history around inside it.
>
> It has code inside to do threading. This is presumably irrelevant on
> plan9.
>
> It has code to pre-fork children. ditto.
>
> It has code to work around Microsoft .DLL strangeness.
>
> It has its own shared-library runtime load code.
>
> It has backwards compatibility code for NCSA httpd.
>
> If you want it because it has HTTP 1.1 conformant implementation, or an
> interesting model of how to represent a file directory as web, thats
> surely better done discretely?
>
> If you want a service which understands apache .htaccess formatted data,
> that too is surely better written another way? I can understand wanting
> to un-tar a tree of web, including .htaccess, even parse an httpd.conf
> but to port the daemon..
>
> cheers
> 	-George
> --
> George Michaelson       |  APNIC
> Email: ggm@apnic.net    |  PO Box 2131 Milton QLD 4064 Phone: +61 7 3367
> 0490  |  Australia
>   Fax: +61 7 3367 0482  |  http://www.apnic.net

There is the Ada Web Server, which should be trivial to get working on
Plan9 once we have an Ada compiler which will target the Plan9 platform.

You can find it at http://www.adapower.com

Implementing Java through Ada is a piece of cake as well, via JGNAT.

Laters.

Caffienator
chris@dont.spam.me

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 569 bytes --]

sometimes the plan 9 compilers do things that gcc doesn't, partly
because adherence to an often elaborate ABI isn't required.  certainly
that's true on the PowerPC and i think it's also true on the ARM. i'd
say from inspection of gcc and experience of 5?  that the 5l linker
for ARM has an easier time sorting out literal pools and ARM/Thumb
linkage than the gcc system, partly because it's working with an
abstract object program as input, not the encoded instruction forms.
we can then convert the resulting executable into a form that other
systems like.


[-- Attachment #2: Type: message/rfc822, Size: 2747 bytes --]

To: 9fans@cse.psu.edu
Subject: Re: [9fans] Rant (was Re: Plan9 and Ada95?)
Date: Fri, 9 Nov 2001 07:43:52 -0600 (CST)
Message-ID: <Pine.GSO.4.10.10111090740440.8629-100000@ultra5a.usask.ca>

plan9's compilers generate code with comparably equal performance to gcc
2.9.5

no, the code is not faster, no, the code is not noticeably slower for jobs
that do not require 5 days to complete (image rendering is what I have
tested -- povray on identical hardware).

that much I can say.

andrey

On Fri, 9 Nov 2001, Thomas Bushnell, BSG wrote:

> nigel@9fs.org writes:
>
> > >> Has anyone compared the efficiency of the code produced by GCC and the
> > >> Plan 9 compiler?
> >
> > I'm not sure that this is a very important issue, whichever is better.
>
> There's been a lot of noise about how GCC might be more ugly, or
> poorly constructed, or such.  I'm asking whether amidst all that noise
> anyone has bothered to see whether it actually performs its job better
> or worse.  It does seem to me to be an important question in
> evaluating tools which one is actually better at the principal job the
> tool is designed to perform.
>

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 914 bytes --]

sorry, i wasn't clear.  i meant that you'd need to change the gnat code generator
to generate a different representation, not that you'd need another
code generator working from the RTL.  as you say, gcc's RTL
is machine-specific.

by `if gnat is structured so as to allow another ... in principle',
i meant, it has some internal representation of the Ada program that
is at or near a useful level for code generation, before it produces RTL.
if it is producing RTL directly from a basic parse tree,
you'll have your work cut out.

i don't remember dewar's description of gnat's structure,
except that it converts its internal representation of Ada
to RTL at some point, so i don't know the answer
and i can't be much more specific without looking
at Gnat's code (which i've also seen but forgotten as regards this point).
it was a good five years ago and i was curious
rather than interested.


[-- Attachment #2: Type: message/rfc822, Size: 2368 bytes --]

To: 9fans@cse.psu.edu
Subject: Re: [9fans] Rant (was Re: Plan9 and Ada95?)
Date: Fri, 9 Nov 2001 10:17:07 GMT
Message-ID: <874ro5vz8b.fsf@becket.becket.net>

forsyth@vitanuova.com writes:

> the gnat code generator doesn't generate code
> directly (or it didn't), it generates the gcc internal representation,
> so it's a non-trivial modification, even if gnat is structured
> so as to allow another code generator in principle.

The "gcc internal representation" (RTL) is not something a different
code generator could work with.  The definition of "valid RTL" is
entirely machine-specific and so forth.  When a GCC front-end begins
writing RTL, it is tightly coupled with the back-end.  (Indeed, it
generates RTL mostly by asking the back-end to spit out RTL for
various operations.)

The value of RTL in GCC is not to firmly decouple the front end from
the back end, but rather to decouple optimizations from the back end.

Thomas

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Andrey A Mirtchovski]

plan9's compilers generate code with comparably equal performance to gcc
2.9.5

no, the code is not faster, no, the code is not noticeably slower for jobs
that do not require 5 days to complete (image rendering is what I have
tested -- povray on identical hardware).

that much I can say.

andrey

On Fri, 9 Nov 2001, Thomas Bushnell, BSG wrote:

> nigel@9fs.org writes:
>
> > >> Has anyone compared the efficiency of the code produced by GCC and the
> > >> Plan 9 compiler?
> >
> > I'm not sure that this is a very important issue, whichever is better.
>
> There's been a lot of noise about how GCC might be more ugly, or
> poorly constructed, or such.  I'm asking whether amidst all that noise
> anyone has bothered to see whether it actually performs its job better
> or worse.  It does seem to me to be an important question in
> evaluating tools which one is actually better at the principal job the
> tool is designed to perform.
>

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

forsyth@vitanuova.com writes:

> the gnat code generator doesn't generate code
> directly (or it didn't), it generates the gcc internal representation,
> so it's a non-trivial modification, even if gnat is structured
> so as to allow another code generator in principle.

The "gcc internal representation" (RTL) is not something a different
code generator could work with.  The definition of "valid RTL" is
entirely machine-specific and so forth.  When a GCC front-end begins
writing RTL, it is tightly coupled with the back-end.  (Indeed, it
generates RTL mostly by asking the back-end to spit out RTL for
various operations.)

The value of RTL in GCC is not to firmly decouple the front end from
the back end, but rather to decouple optimizations from the back end.

Thomas

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

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

> > But it's not free software...
>
> Yes it is.  It really is.  When I see people say this, which I do from time
> to time, it makes me sad for many reasons.  The parallel with religions,
> with the Catholic schism, with the fissiparous history of the Protestant
> church, is so obvious it's almost embarrassing to point out.  But I must.
>
> The `free' software people have won, but they're so intent on having
> everyone agree with their fundamentalist and faith-based definition
> of `free' that they label many of their allies as enemies.  It's painful to
> watch.  It hurts.

The people that are committed to free software are committed to
software that comes with liberties and rights.  I'll gladly pay for a
copy of something as nice as Plan 9.  But I won't sacrifice my
freedoms for it.

And the people who might work on Plan 9 but don't, because they are
committed to free software, are committed to the libre definition, not
the gratis one.  Stamping your foot and saying it isn't fair won't
change that.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

nigel@9fs.org writes:

> >> Has anyone compared the efficiency of the code produced by GCC and the
> >> Plan 9 compiler?
>
> I'm not sure that this is a very important issue, whichever is better.

There's been a lot of noise about how GCC might be more ugly, or
poorly constructed, or such.  I'm asking whether amidst all that noise
anyone has bothered to see whether it actually performs its job better
or worse.  It does seem to me to be an important question in
evaluating tools which one is actually better at the principal job the
tool is designed to perform.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Taj Khattra]

On Thu, Nov 08, 2001 at 07:30:39AM -0500, bwc@borf.com wrote:
>
> What do these economics say about optimizing compilers?
>

here's a link to proebsting's "law" of compiler optimization

	http://research.microsoft.com/~toddpro/papers/law.htm

and some related observations

	http://www.cs.virginia.edu/~jks6b/on_proebstings_law.pdf
	http://www.cs.umd.edu/~pugh/IsCodeOptimizationRelevant.pdf

-taj

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Russ Cox]

You think Windows runs in polynomial time?

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[George Michaelson]

Apache is a bad idea because like mozilla, it carries the leaden weight
of history around inside it.

It has code inside to do threading. This is presumably irrelevant on plan9.

It has code to pre-fork children. ditto.

It has code to work around Microsoft .DLL strangeness.

It has its own shared-library runtime load code.

It has backwards compatibility code for NCSA httpd.

If you want it because it has HTTP 1.1 conformant implementation, or
an interesting model of how to represent a file directory as web, thats
surely better done discretely?

If you want a service which understands apache .htaccess formatted data,
that too is surely better written another way? I can understand wanting
to un-tar a tree of web, including .htaccess, even parse an httpd.conf
but to port the daemon..

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

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Steve Kilbane]

Going back slightly, anothy said:

> // ...suggesting that Mozilla be ported to Plan 9.
>
> this is the same discussion as apache, i think.

..with which I'd disagree. A server doesn't fall into the same
category as a program with a GUI (or even a UI), because its
externally-discernable behaviour is far more clearly defined;
in this case, it's defined within a domain - TCP - that Plan 9
supports.

Putting it another way, interoperating with other programs/systems
is a different class of problem from interacting with users used
to those systems.

steve

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Noah Diewald]

On Thu, Nov 08, 2001 at 12:58:07PM +0000, Matt wrote:
> If the world has got time for transparent menus and java applets then
> it can suffer not unrolling every loop and using registers when
> possible.

I've implemented a transparent menu in this email.  See below.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Matthew Hannigan]

It would be interesting to see ?c
ranked here: http://www.bagley.org/~doug/shootout/

Very interesting site.


"Thomas Bushnell, BSG" wrote:
>
> Has anyone compared the efficiency of the code produced by GCC and the
> Plan 9 compiler?

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Quinn Dunkan]

> is all I can contribute to Plan 9 myself.  It is my impression,
> and it must please be seen as such, that the Plan 9 developer are
> fearful that the GNU/Linux/*BSD etc. developer community will
> "taint" Plan 9 with their indiscriminate bloat.  Maybe such
> contamination is possible (I believe it is called miscegenation in

I think it's a much more practical matter.  gcc is not ported because no one
has ported it.  No one has ported it because no one has the time, which means
that no one wants it enough to set aside the time.  Hey, maybe dhog will
finish whatever he's doing with it, and then everyone who wants gcc can go get
it.

Sure, 9fans is often critical of things like gcc.  But if all it takes to
scare you out of porting something is a few disparaging comments from a
mailing list, you'd never have the stomach to do the port in the first place.

Plan9 was obviously designed, and designed by people who have clear and strong
ideas about what good design is.  But if you port or create something that
violates the design ideas of some other people, they're not going to come to
your house and shoot you.  Maybe someday someone will finish and release a gcc
port.  Then maybe someone will port mozilla.  I wouldn't expect those to go
into the standard distribution, but we do have a wiki, and it's dirt simple to
put up a page and a link.

Plan9 is also plenty free enough for me, and I suspect it is for most other
people as well (naturally you only hear from the vocal dissenters, though).
comp.os.plan9 has its share of silly time wasting arguments, but it would
hardly be Usenet without them.

> I don't expect the Plan 9 team to agree with me, we have a very
> different outlook at the "social" level, but perhaps there are
> others on this mailing list who feel that the "social experiment
> (Che Guevara)" is worth conducting even if the casualties could be
> numerous.

If you were waiting for *my* approval, go ahead and conduct it, whatever that
means.  Do whatever you want, as long as it doesn't involve shooting people.

> Naturally, unlike Che Guevara, I don't propose to snatch the
> leadership position from the Plan 9 team, quite the contrary, I
> very much appreciate their continued contribution.  I would just
> appreciate it even more if they were encouraging rather than critical
> in those areas that apparently offend their sensibilities.

I think they're very encouraging.  When technical questions are raised, they
give helpful and clear answers, even when those questions are answered in the
documentation.  When "philosophical" questions are raised, they actually try
to understand what the question is (which is often most of the work, see the
"link" debate), and then give their honest opinions.  What more do you want, a
lollipop?  Since when does anyone need a unanimous vote of acceptance from
9fans to do anything?


On the subject of "what makes plan9 unique", for me, plan9 completely changed
my views on the mouse.  Forget the networking stuff, I just have a 14.4K
modem, the UI is more fun than anything else.  Once, not long ago, I honestly
believed that if I wanted to replace a word I had typed it was easier and
faster to type "^]2bWcWfoo^]A" than to click on the word and type "foo", and
therefore, it was a necessary evil to embed GNU readline into everything that
read from the terminal.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[anothy]

// Well, let's see if we can agree: XML, GCC, X, C++, JAVA, Perl, off
// the cuff, have all drawn criticism on this list.

true. but i don't see that as a bad thing. what _would_ be bad would
be if Plan 9 did _not_ recieve criticism on this list, or if such criticism
was not listened to or addressed. i believe it is. all these things - your
list and Plan 9 - have problems, and we can only correct them by
addressing them.

// None of the criticism has addressed the social need, as you so
// eloquently explain in your example, to support these entities in
// some fashion or provide alternatives,

i'm not sure that's true. i agree there hasn't been much, if any, talk
about how to support XML or X. but there's been talk of and work on
importing GCC (thus C++) and Perl. further, note your own trailing
clause: "or provide alternatives." so we don't have X, and nobody's
working on getting it. why? because we've got an alternative we like
better, and nobody's shown reason to like X better. there's been a
little talk of getting Java going, but not much, because i think most
people who'd be interested in what it offers on this list consider us
to have an alternative we prefer: Limbo.

i think maybe there's a missing catagory in the discussion here:
interoperability tools. VNC is a great example, i think. what VNC
does for graphics is most definatly _not_ the Plan 9 graphics
model, but it's a very useful tool for talking to a diverse range of
systems. and we've got a vnc viewer. i think that suggests we (as a
community) do look at things from other places, and make the
cost/gain decision i keep harping on. there _are_ things of value
that come from other places, but getting them running locally has
some cost associated with it. each case is a seperate decision.

that being said, not everyone has to come to the _same_ conclusion
when such decisions are made. don't think the correct call was made
in ignoring a local X server? maybe you're right (i still consider that);
go do it. think a Java VM would give you things better than Inferno?
cool, import or build one. think Perl, or Python, or Ada is a good
development language/tool? wonderful, we eagerly anticipate your
results. i really think we do.

i don't think anyone'd be "mad" at you for importing X, or GCC, or
whatever. people might suggest it wasn't the best use of your time,
but that's their decision to make. luckily, you get to determine what
you do with your time, not they.

maybe we do get overly critical from time to time. maybe we do get
too set in our ways. but i still hold that to be the exception here, not
the rule. and i think we do okay at avoiding it.
ア

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[forsyth]

>>Is there anything in GNAT
>>that exclusively requires GCC? Can those issues be dealt with via a
>>little select modification of the GNAT sources?

the gnat code generator doesn't generate code
directly (or it didn't), it generates the gcc internal representation,
so it's a non-trivial modification, even if gnat is structured
so as to allow another code generator in principle.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[forsyth]

[-- Attachment #1: Type: text/plain, Size: 75 bytes --]

if it's that superior(!), perhaps it doesn't think much of your laptop.


[-- Attachment #2: Type: message/rfc822, Size: 1924 bytes --]

To: 9fans@cse.psu.edu
Subject: Re: [9fans] Rant (was Re: Plan9 and Ada95?)
Date: Thu, 8 Nov 2001 10:39:32 GMT
Message-ID: <87n11xyaar.fsf@becket.becket.net>

lucio@proxima.alt.za (Lucio De Re) writes:

> Let's get off the hobbyhorse of criticising all the alternative
> platforms and actually co-operate with them or show that the Plan
> 9 way is as superior as we make it out to be.

And hey, if it's that superior, it would run on my laptop, right? :)

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[presotto]

[-- Attachment #1: Type: text/plain, Size: 80 bytes --]

The lack of a browser pretty much makes it useless
for most peoples' laptops.

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

From: "Thomas Bushnell, BSG" <tb+usenet@becket.net>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] Rant (was Re: Plan9 and Ada95?)
Date: Thu, 8 Nov 2001 10:39:32 GMT
Message-ID: <87n11xyaar.fsf@becket.becket.net>

lucio@proxima.alt.za (Lucio De Re) writes:

> Let's get off the hobbyhorse of criticising all the alternative
> platforms and actually co-operate with them or show that the Plan
> 9 way is as superior as we make it out to be.

And hey, if it's that superior, it would run on my laptop, right? :)

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[rob pike]

> But it's not free software...

Yes it is.  It really is.  When I see people say this, which I do from time
to time, it makes me sad for many reasons.  The parallel with religions,
with the Catholic schism, with the fissiparous history of the Protestant
church, is so obvious it's almost embarrassing to point out.  But I must.

The `free' software people have won, but they're so intent on having
everyone agree with their fundamentalist and faith-based definition
of `free' that they label many of their allies as enemies.  It's painful to
watch.  It hurts.

-rob

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[bwc]

> Has anyone compared the efficiency of the code produced by GCC and the
> Plan 9 compiler?

I recently re-read Jim Gray's paper `The 5-minute Rule.'  It's interesting
to note that when he wrote the paper (1985) a meg of memory was $5K and a MIP
was $50K.  Now, by my calulations, a meg is about $0.50 and a mip is about
$0.30.

The economics of CS used to be:
	1) correctness of programs
	2) time efficiency
	3) space efficiency

Considering the changes in speed and memory, I assert that for all but
the most demanding case, only 1) is still a limited resource.

What do these economics say about optimizing compilers?

  Brantley Coile

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[nigel]

[-- Attachment #1: Type: text/plain, Size: 1093 bytes --]

>> Has anyone compared the efficiency of the code produced by GCC and the
>> Plan 9 compiler?

I'm not sure that this is a very important issue, whichever is better.

If the Plan 9 C compiler produced better code, would that immediately
causes the free-nix community to change compiler? Of course not, there
are many considerations other than efficiency.

The answer is that gcc is probably/possibly/allegedly more efficient,
but not by a substantial degree.

So, lets say the code is X% faster, and even X% smaller.  How does
this help?  If the code you want to run is within X% of catastrophe,
then squeezing the code with the aid of a compiler is not the only
solution.  Throwing away a lot of the code is quite a good one too.

And, before I get flamed that this is not a commercially minded
answer, a substantial part of my employ has been spent building small
embedded systems.  When the code didn't fit, we invariably played with
the compiler, decided it didn't help enough, and then started removing
code.

What value of X makes changing compiler worthwhile?


[-- Attachment #2: Type: message/rfc822, Size: 1497 bytes --]

From: "Thomas Bushnell, BSG" <tb+usenet@becket.net>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] Rant (was Re: Plan9 and Ada95?)
Date: Thu, 8 Nov 2001 10:39:58 GMT
Message-ID: <87g07pya0g.fsf@becket.becket.net>


Has anyone compared the efficiency of the code produced by GCC and the
Plan 9 compiler?

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

lucio@proxima.alt.za (Lucio De Re) writes:

> A lot of same criticism seems to point to a desire to see Plan 9
> gain its rightful position in the operating system marketplace and
> the unfairness of it having to compete with obviously inferior
> products with greater market share.

I can't speak for the evils of Redmond.  But a clear reason that Plan
9 is beaten out by GNU/Linux is that the latter is a free operating
system, and always has been, and Plan 9 isn't and never was.

> At the core, I have always believed that we can draw on the broad
> developer community for further development, mainly because that
> is all I can contribute to Plan 9 myself.

The broad developer community you can hope to draw from is, more or
less, committed to free software.  I would turn around tomorrow and
run Plan 9 nearly exclusively if it were free software, and I'd spend
effort making it as good as possible.

But it's not free software, so I won't.  And there are *many* people
who share the same ideals.  Those people contribute to systems like
the GNU/Linux distributions and the various *BSD systems.

Thomas

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

Has anyone compared the efficiency of the code produced by GCC and the
Plan 9 compiler?

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Thomas Bushnell, BSG]

lucio@proxima.alt.za (Lucio De Re) writes:

> Let's get off the hobbyhorse of criticising all the alternative
> platforms and actually co-operate with them or show that the Plan
> 9 way is as superior as we make it out to be.

And hey, if it's that superior, it would run on my laptop, right? :)

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Caffienator]

In article <20011107192058.2C84D199E7@mail.cse.psu.edu>, "forsyth"
<forsyth@caldo.demon.co.uk> wrote:

>>>Let's get off the hobbyhorse of criticising all the alternative
>>>platforms and actually co-operate with them or show that the Plan 9 way
>>>is as superior as we make it out to be.
>
> you're possibly overlooking the possibility that some of us have written
> things such as portable compilers (and other things) ourselves for a
> good few other systems, and take that into account when making
> comparisons -- it's not
> just that this or that isn't Plan 9.


I certainly didnt intend to start a flamewar with my post.

Right now, I develop for Linux. Primarily because it's the simplest to
use in regards to the development tools that are available. Sure, I could
figure out how to jerry rig alot of those tools into Plan9, but I just
dont have the time(for now) to do that sort of thing.
I've been running Plan9 across a few local workstations, just to play
around with it. It has alot of potential, and I would really like to
develop for it, but the tools available on Plan9 are somewhat limited.

With that being said, I do plan on developing some of my own tools and
making them available to the Plan9 community, but the big ones, like an
Ada compiler, just might be a little too much for me to bite off at this
point.

Now, Plan9 does have it's own C compiler. I dont suppose it would be
impossible to integrate GNAT on top of that? Is there anything in GNAT
that exclusively requires GCC? Can those issues be dealt with via a
little select modification of the GNAT sources?

Thoughts?

Caffienator
chris@dont.spam.me

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Christopher Nielsen]

On Thu, Nov 08, 2001 at 10:05:38AM +0200, Lucio De Re wrote:

[snip]
> A CVS repository could be a start, but without the participation of
[snip]

CVS has a lot of problems. If you're thinking of creating a
source repo, take a look at subversion:

http://subversion.tigris.org/

It's not quite ready yet, but it should be in a couple months,
and it's far superior to CVS.

As nifty as the dump fs is, there are some things it doesn't
provide that a large development team needs in the software
development process. Of course, maybe the idea is to rethink
the development process.

My thought is that something like subversion that utilizes
the ideas of plan9 would be a powerful and useful development
tool.

--
Christopher Nielsen - Metal-wielding pyro techie
cnielsen@pobox.com
"Those who are willing to trade freedom for security deserve
 neither freedom nor security." --Benjamin Franklin

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

On Thu, Nov 08, 2001 at 03:51:47AM -0500, Russ Cox wrote:
>
> > of the core entity.  A CVS repository could be a start, but without
>
> I'd be a lot happier using CVS if it didn't require
> having droppings scattered all over my directory trees.
>
> </unhelpful>

Sorry, really don't know how to help that :-)  Maybe you can bind all
the CVS directories from a single, parallel namespace? :-) :-) :-)

++L

PS: what I meant by "CVS repository" was a well maintained public
repository.  What keeps it in shape is of little relevance.  Only
two things are important: it must encourage contributions and it
must draw healthy discussion.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Russ Cox]

> of the core entity.  A CVS repository could be a start, but without

I'd be a lot happier using CVS if it didn't require
having droppings scattered all over my directory trees.

</unhelpful>

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

On Thu, Nov 08, 2001 at 01:45:51AM -0500, anothy@cosym.net wrote:
>
> i don't think this is so, but i'm not sure it's what you intended to
> say, either. elsewhere you seem to argue that it takes a large
> community of developers to produce _lots_ of _different_type_ of good,
> solid products. and i'd agree with that.
>
That's where the _noise_ comes in.  We could wish for an elite of
really cool programmers, of the caliber we're familiar with, but
without Bell Labs' selection criteria, we have to allow natural
selection to perform the discrimination.

> i guess i just don't see the "party line" bit. maybe it's there, but i
> don't see it. i find this to be a much more open forum than most others
> i've spent time in, computer-related or not.
> ?

Well, let's see if we can agree: XML, GCC, X, C++, JAVA, Perl, off
the cuff, have all drawn criticism on this list.  None of the
criticism has addressed the social need, as you so eloquently
explain in your example, to support these entities in some fashion
or provide alternatives, even if only as opening moves towards
superior replacements.

Maybe I expect too much, but I'd like to see the type of comment
that encourages the developer to consider alternatives and eventually
produces them, even if somewhere in the quest for acceptance some
principles have to be compromised.o

For example, C++ is too unwieldy to implement efficiently, but a
few features, such as extensions to structs, operator overloading
might be worth adding to the C compiler for the benefit of compiling
existing code, on the assumption that only some C++ extensions have
really gained popularity.  Purists will frown on such suggestions,
and their opinion should be noted, but not necessarily followed.

I could get myself excommunicated from here by listing all the
heresies I have considered, the above is just a sample.  My feeling
is that one needs a place where heresies are vented in public, draw
appropriate criticism and what valid essence they have becomes part
of the core entity.  A CVS repository could be a start, but without
the participation of the Plan 9 team (which incidentally would
mirror, magnified, Linus Torvald's role in the Linux kernel) it
would soon deteriorate to bulk for the sake of publication.  Maybe
I'm just dreaming, because something along these lines ought to
develop of its own accord, but my fear is that criticism of "foreign"
products discourages such contributions.

++L

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

On Thu, Nov 08, 2001 at 03:43:53PM +1000, George Michaelson wrote:
>
> Folks, its a research tool. Its not a bet-the-farm-lucent-will-rise-again
> product, its not a tool for the masses, If Tom Duff can make movies on it
> thats way cool, but please, don't turn it into the kind of packaged crap
> its "finger-quotes" competing against.
>
I'm not sure where this discussion is going.  I started by questioning
the wisdom of criticising competing products in a detrimental rather
than constructive fashion, a habit I believe detracts from the
otherwise impeccable record of the Plan 9 developers and contributors.

Naturally, I also wanted to know where the attitude originated and
whether it was conceivable that it could be better channeled.  Not
that I can express myself so well as to be understood without much
gesticulation and verbal diarrhoea.

> For me, half the pleasure of reading about this thing is that its obviously
> fanatically small, and reductionist. Making it any bigger will ruin it.
>
A lot of same criticism seems to point to a desire to see Plan 9
gain its rightful position in the operating system marketplace and
the unfairness of it having to compete with obviously inferior
products with greater market share.

I certainly can't speak for the Plan 9 developers, but in my opinion
the Plan 9 concepts, across the board, deserve much greater
acceptance.  Where I believe my opinions disagree with the Plan 9
developers' is how such broader acceptance should occur: I'm almost
Microsoftish with an "embrace and extend" philosophy, while "they"
seem to have more of an "educate and conquer" approach.

At the core, I have always believed that we can draw on the broad
developer community for further development, mainly because that
is all I can contribute to Plan 9 myself.  It is my impression,
and it must please be seen as such, that the Plan 9 developer are
fearful that the GNU/Linux/*BSD etc. developer community will
"taint" Plan 9 with their indiscriminate bloat.  Maybe such
contamination is possible (I believe it is called miscegenation in
the Old Testament) and likely, but I would argue that we have not
yet seen the results that are so greatly feared, and that we should
encourage the experimentation, not prevent it by taking a critical
approach to it.

I don't expect the Plan 9 team to agree with me, we have a very
different outlook at the "social" level, but perhaps there are
others on this mailing list who feel that the "social experiment
(Che Guevara)" is worth conducting even if the casualties could be
numerous.

Naturally, unlike Che Guevara, I don't propose to snatch the
leadership position from the Plan 9 team, quite the contrary, I
very much appreciate their continued contribution.  I would just
appreciate it even more if they were encouraging rather than critical
in those areas that apparently offend their sensibilities.

++L

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Steve Kilbane]

I don't see tools like Perl or Python bringing more people into Plan 9,
particularly when faced with the mass of Windows users. Plan 9 is
about doing things a different way, with the implied assumption that the
difference arises from an attempt to improve things. If potential users
need their comfy feature set to attract them, then they're the wrong user
set. Plan 9 is implicitly targeted at people who are willing to try changing
how they work.

This is also one reason why concepts from Plan 9 (which includes manner of
thought as well as software models) has less impact on the outside world:
both the *nix worlds and the Windows worlds are significantly larger user
bases that are more into evolutionary change than revolutionary change.
Windows XP has only just dropped DOS; Linux and GNU are basically copies of
30-year-old systems. Such large user bases are inherently resistant to
change.

This is also one reason why gcc is the crufty behemoth it is, and ?c
isn't: gcc had to accept and deal with problems on many systems, while
?c could assume they'd go away, by being fixed elsewhere. As forsyth says,
portable compilers aren't inherently that nasty, but gcc itself has become
a large system, dealing with many scenarios, and thus encouraging evolution
rather than revolution. That's why it's a platypus.

steve

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Jim Choate]

On Thu, 8 Nov 2001, George Michaelson wrote:

> Folks, its a research tool. Its not a bet-the-farm-lucent-will-rise-again
> product, its not a tool for the masses, If Tom Duff can make movies on it
> thats way cool, but please, don't turn it into the kind of packaged crap
> its "finger-quotes" competing against.
>
> For me, half the pleasure of reading about this thing is that its obviously
> fanatically small, and reductionist. Making it any bigger will ruin it.

Then get ready to have it 'ruined'.

http://einstein.ssz.com/hangar18


 --
    ____________________________________________________________________

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

                                             Bumper Sticker

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

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[anothy]

// I have never used plan 9's httpd, but from the features list it seem
// very good.  Certainly can't fault it from the client's perspective.

i can't say i recomend it. apache was a reasonable example; it really
_might_ make sense to port it (i don't think so, but it might). i
certainly believe something more robust than what we've got now would be
very usefull.

// ...suggesting that Mozilla be ported to Plan 9.

this is the same discussion as apache, i think. there are features that
we want: good html rendering, good javascript support, good set of
plugins (where good in those three cases, unfortunatly, really should be
read as "what web developers expect"). this is very hard to do. but so
is porting Mozilla (i looked at it. breifly). someone interested in
exerting effort to get these features needs to decide where to spend her
effort. a cost/gain decision.

// ...it takes a large community of developers to produce (amongst the
// noise) good, solid products.

i don't think this is so, but i'm not sure it's what you intended to
say, either. elsewhere you seem to argue that it takes a large
community of developers to produce _lots_ of _different_type_ of good,
solid products. and i'd agree with that.

there is a balance to be struck between the usefulness of importing
foreign code and the danger of doing so - diluting the system's
benefits, turning it into "just another unix". as such, i'm still not
sure what i think of the GCC port. GCC is ugly and awful. but it can
give me things that i want. like helping me get rid of the one
remaining Solaris box i run, when i can build the two things we use
it for that're in c++ on Plan 9. then i can get the web developer
who writes code for those two bits to be writing code on Plan 9, and
i have a chance at migrating him to better things.

i guess i just don't see the "party line" bit. maybe it's there, but i
don't see it. i find this to be a much more open forum than most others
i've spent time in, computer-related or not.
ア

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Andrey A Mirtchovski]

On Thu, 8 Nov 2001, Lucio De Re wrote:

> So, to re-iterate my point, Plan 9 is the better platform, but should
> not present itself from a position of arrogance, but rather endeavour
> to provide a better platform for development (which is easy) with a
> more familiar environment.
>

I have a different opinion, much based on what I see on this list and what I
hear from prople with attitudes similar to this:

Plan 9 developers are not arrogant, rather, they are trying to defend
themselves and their ideas from the random bunch of people who come here
and lend their advise and thoughts without being asked for, or without
really being aware of the system they discuss (it's really mostly a "you're
out of your element.." issue, to continue throwing quotes from favourite
movie :)...

I do believe that Plan 9 is _the_ OS with the clearest philosophy behind it
-- one that has the system build around it and one that can be continuously
defended without running into infinite loops (examples of the opposite would
be -- Simple OS (Linux) and Optimized for i386 (FreeBSD))... Should we break
this just to lure more people in, without making sure they appreciate P9 at
a level different than just very neatly written code?

Well, come to think of it, after a few glasses of wine, P9 might as well be
called 'the dude'... Then everyone else starts to sound like they ask 'vere
is ze money libouski' :)

This not a flame, simply musings I've wanted to share but haven't had the
guts to up until now...

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[George Michaelson]

God, time cycles get shorter. At least for Jerry Cornelius the loop
is the 64,000 year 'small' circuit.

As one who has never run it, but often pressed for wider deployment of Plan9,
suggesting it could be bedded in as DNS servers, or critical infrastructure,
I can only repeat the Al Haig line once again:

	In order to sell Plan 9, it was neccessary to destroy it.

Folks, its a research tool. Its not a bet-the-farm-lucent-will-rise-again
product, its not a tool for the masses, If Tom Duff can make movies on it
thats way cool, but please, don't turn it into the kind of packaged crap
its "finger-quotes" competing against.

For me, half the pleasure of reading about this thing is that its obviously
fanatically small, and reductionist. Making it any bigger will ruin it.

No irony intended.

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

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

On Wed, Nov 07, 2001 at 04:34:47PM -0500, anothy@cosym.net wrote:
>
> Apache's a harder sell. do people want "Apache" or a web server with a
> certain feature set? if the later, one has a decision to make: port (and
> maybe improve) Apache, improve Plan 9's httpd, or build something
> new. there's a legitamate cost/gain analysis here; the fact that we didn't
> build Apache shouldn't enter into it.
>
I was using examples that came to mind immediately, not a short
wish list.  I have never used plan 9's httpd, but from the features
list it seems very good.  Certainly can't fault it from the client's
perspective.

A better point would have been made (but I was really trying not
to fan the flames) by suggesting that Mozilla be ported to Plan 9.

It's not the product, the developers, as much as the usefulness.
My point is that there is a developers' community out there and if
only they knew about acme and ?c, they might embrace them.  But
they need attracting, which can only be done with the familiar.
In my opinion, as a NetBSD user amongst Linux developers, what
caught the imagination there was Doom/(oops, can't remember the
name now).  From the large group of game aficionados (not to call
them addicts) grew a smaller community of contributors.

I have no difficulty concurring with everyone here that quality is
not high on these developer's agenda, but bulk cannot be avoided.
Amongst all the chaff, there are a few gems, of which Apache would
certainly be one.

So, to re-iterate my point, Plan 9 is the better platform, but should
not present itself from a position of arrogance, but rather endeavour
to provide a better platform for development (which is easy) with a
more familiar environment.

> i agree everyone could benefit by more active exchange between Plan 9
> and other systems. but i think it's a big leap to go from there to saying
> we should spend more time improving GCC or porting Apache.
> ?

It's a matter of perspective, I think.  My platform needs are closer
to SQL-oriented databases, volume processing, event-oriented diary
systems.  Some of this I can find in the public domain (postgreSQL
or mySQL would be options) while the rest (ERP stuff like SAP) is
unnecessarily complex and expensive, there isn't enough fat in my
client's profit margins to afford SAP or Great Plains, nor is there
enough money to invest in the doubtful expertise of an MCSE as
server administrator.

Plan 9 has all the right attribute as the platform of choice, is
just missing the applications.  And I have little doubt that
developing the applications would be easier and for the same
complexity one may get a much richer feature set.  But it takes a
large community of developers to produce (amongst the noise) good,
solid products.

And one of Plan 9's greatest assets is what my mind has latched on
as the "Bell Labs Giants".  I find it hard to appreciate all their
contributions, though, when their attitude is unnecessarily (not
"unjustifiably", I admit) disparaging.  And I see some churn on
this list as a result, the person I believe should be in there with
the Giants but has been sadly quiet for a long time is G David
Butler, I really thought he had already added good value to Plan
9 and was going to provide more.  There are others I should recall
but it's probably best that I can't who are brushed aside because
they don't quite tow the party line.  I guess that's what I find
hard to swallow.

Not that the party line isn't good, quite the contrary, but it also
benefits from occasional divergence, while using it as a soap box to
denigrate the opposition seems to me to be very short sighted.  Which
is heavy accusation to level at the Giants, but it appears to me to be
a truthful one.

++L

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Dan Cross]

In article <20011107185604.2664B199F3@mail.cse.psu.edu> you write:
>I don't know about Nt/XP yet, but I'm guessing that they'll
>be harder.

If you take Nt and XP, and remove the last and first letters of
those names (respectively), you end up with NP.  Hmm, coincidence?
I think not.

	- Dan C.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[anothy]

// ...continuous criticism... ...of everything beyond the boundaries
// of Plan 9/Inferno, no matter how justified, isn't healthy.

i'd agree with the implication but disagree with the statement.

i think constant criticism is a very good thing, provided it's done in a
productive manner, and the criticism is somewhat more concrete than
"not invented here." it is this ongoing criticism that will help all these
systems change, correct their flaws or failures of vision, and improve.

which brings me to my point of agreement. you say "everything beyond
the boundaries of Plan 9/Inferno" and i think that's a good observation.
Plan 9 and Inferno are by no means perfect. as someone noted some
time ago, the alt.sysadmin.recovery FAQ gets it right: no systems don't
suck, plan 9 simply sucks less than others. i think it's important that
_everybody_ needs to be occasionally reminded that they suck in some
fashion or other. but with that needs to come info on _how_ one sucks,
and how to suck less.

take the current compiler discussion. i would say the contention here is
not that GCC is worthless and ?c/?l are perfect, but rather that one can
be more productive improving ?c/?l than GCC. you correctly note that
the Plan 9 tools don't deal with cross-OS compiling (except in very
limited cases), whereas GCC does (to some degree, anyway). i don't
believe anyone is disputing this, nor claiming it doesn't matter. but i bet
most people here would say it'd require less overall man-hours to get
8c/8l to build Linux binaries than to get GCC to build Plan 9 ones, and
that the results would enable people on whatever platform to develop
things better and more quickly.

// Perl would open another [door], Python a third, Apache a fourth, etc.

Perl and Python i can see, for sure. they're languages, with apps
written in them, that people want to be able to use. and for good reason.
Perl and Python each have benefits one cannot get with C, rc, or Limbo.
i'd love to see them supported better on Plan 9.

Apache's a harder sell. do people want "Apache" or a web server with a
certain feature set? if the later, one has a decision to make: port (and
maybe improve) Apache, improve Plan 9's httpd, or build something
new. there's a legitamate cost/gain analysis here; the fact that we didn't
build Apache shouldn't enter into it.

i agree everyone could benefit by more active exchange between Plan 9
and other systems. but i think it's a big leap to go from there to saying
we should spend more time improving GCC or porting Apache.
ア

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

On Wed, Nov 07, 2001 at 07:58:32PM +0000, forsyth@caldo.demon.co.uk wrote:
>
> >>And the fact that at the present value of our currency, the Inferno
> >>sources (I presume I need that licence to help develop the Inferno
> >>tools) would cost me one month's income :-(
>
> no, that bit is free.  i always intended it to be but it didn't end up
> on the web site until recently for some reason.

Thank you, that is very encouraging.  I'd love to contribute to
Vita Nuova's revenues, I really think you deserve it, but we're on
an economic downwards spiral here, with no end to it in sight.  It
isn't even clear what the causes for it could be.

++L

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

On Wed, Nov 07, 2001 at 07:25:32PM +0000, forsyth@caldo.demon.co.uk wrote:
>
> >>Let's get off the hobbyhorse of criticising all the alternative
> >>platforms and actually co-operate with them or show that the Plan
> >>9 way is as superior as we make it out to be.
>
> you're possibly overlooking the possibility that some of us have
> written things such as portable compilers (and other things)
> ourselves for a good few other systems, and take that into account
> when making comparisons -- it's not
> just that this or that isn't Plan 9.

My point isn't in the achievements, one would have to be very blind
or inexperienced to fault the contributions made by the Plan 9 team
(in its broadest sense).

On the other hand, continuous criticism (it's an attitude that's
been bothering me for years now) of everything beyond the boundaries
of Plan 9/Inferno, no matter how justified, isn't healthy.

Whether it's GCC, USB, X or Emacs, nobody needs to be reminded that
they suck in some fashion or other.  What we all need are replacement
tools that provide equivalent value with all the advantages of the
Plan 9 platform.  And even more importantly, one needs the platform
to support the user base.  Take my bread-and-butter client:  they
run their computer system under SCO Unix (too shy of Solaris, but
they need multiprocessing power for the COBOL application they are
continually developing) and most of the users are effectively data
capture clerks with terminal emulators on their desks.

The cheapest terminal today is a Win'9x workstation, specially if
e-mail, a web browser, a word processor and a spreadsheet are
complementary, productivity enhancing tools.  They needn't be, but
if the perception is they are, one may as well learn to live with
it.  Then there's compatibility with Word and Excel that is more
critical than the compatibility of the terminal emulator - there's
some control in making the COBOL program not push the terminal
emulator too far, the same cannot be said of e-mail messages and the
spread of office documents.

I could go further, but I'm sure the scenario is familiar enough.
We'd all love to put Plan 9 terminals on these desks, reduce the
cost of onwership, support, need to upgrade etc. by one or two
orders of magnitudes.  But this isn't possible, clearly.  Next
best, coordinate the Win'9x workstation using SAMBA to provide file
services (mostly for backup and information exchange) rather than
the more expensive, more difficult to administer NT.  Can we use
Plan 9 there?  No, again.  More's the pity.

We can learn lessons from Plan 9, centralise file services and
authentication, set up roaming profiles, minimise hardware differences,
lots of hints from the Plan 9 philosophy, but the real thing runs
exclusively on my desk and then mostly in the guise of VNCviewer
(I also use Charon as the browser, wherever I can).

So the reality is that Plan 9 _is_ the superior platform, but there
isn't the appeal in it that draws the software developers to produce
applications that users actually want.  Why is this so?  Because
the bridges to the more popular development platforms are not being
built.  Much as I was horrified that dhog was prepared to port GCC
3.0 to Plan 9 (I was, honestly), it would open a door; Perl would
open another, Python a third, Apache a fourth, etc.  One can improve
on each of these, replace them with more suitable implementations,
later.

So let's not spend our efforts knocking what's familiar, let's
encourage it and then improve it.  There's plenty of effort in
that.

OK, I'd better shut up now, I know this is a boring subject, sorry
to have let off steam at everyone's expense.

++L

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[forsyth]

>>And the fact that at the present value of our currency, the Inferno
>>sources (I presume I need that licence to help develop the Inferno
>>tools) would cost me one month's income :-(

no, that bit is free.  i always intended it to be but it didn't end up
on the web site until recently for some reason.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

On Wed, Nov 07, 2001 at 01:56:02PM -0500, David Gordon Hogan wrote:
>
> No, it's a symptom of ``I wasted weeks of my life struggling
> with this awful code''.
>
Which I'm sure it is.  But a lot of effort has gone into it, and even
though they may not be giants, the GCC developers are hard working
people with good intentions and a legacy that cannot summarily be
discarded.

> > In the meantime, if I want to cross-develop for Windows or Linux,
> > or any other established platforms, Plan 9 is just no use to me,
> > while GCC and its offspring are.  I know what my choice would be,
> > but it's no choice, is it?  Oh, I forget the Inferno tools, those
>
> GCC (Cygnus) is useless for Windows.  There's no way to link
> a VXD!  So if you're doing any _serious_ development, are you
> going to use two compilers, just so you can have the dubious
> pleasure of using GCC?  I don't think so.
>
Unless I'm much mistaken, there's been a flurry of activity in that
direction (I'm afraid I'm only a very superficial follower of the
binutils efforts, I could not follow the GCC mailing list too, it
would be wasted on me) and I got the impression that good results were
obtained.

> I'm sorry to be so ascerbic here, but I am sick of hearing people
> defending GCC.
>
Surely it is more the case that people attack GCC?  It doesn't need
defending, it is by far the most common Unix language compiler, unless
I'm missing something.  I'm frightened to bits of what will happen
when it grows too large to be manageable, but the army of ants that
are still holding it together deserve admiration, not insults.

GCC is like a very large city.  None of it makes sense, but its
citizens cannot escape from it.  Nor can a more logical, more user
friendly version be built to replace it, it will just not succeed.
Think Brazilia.

> Why don't you help us improve the Inferno tools instead of
> complaining about them?
>
Huh?  Quite the contrary, the only flaw I found in the Inferno tools
was that I totally forgot about them - maybe because when I tried to
use them, they complained about a missing rcmain.  As soon as I can
figure out how to use them, I'll be only too pleased to do so.
Specially that old favourite of mine, the rc shell, which I presume is
what rcsh.exe is (lack of documentation, while excusable, is a bit of
a problem).

And the fact that at the present value of our currency, the Inferno
sources (I presume I need that licence to help develop the Inferno
tools) would cost me one month's income :-(

> You want suggestions?  We could force all those GNU people to read
> Rob's essay on programming style, for starters.
>
I guess software bloat is like being overweight is like late software
projects: one bit at the time.  Adn by the time you take stock the
effort involved in undoing the damage may be far too much.

> I have a cunning plan to use 8c to generate files that will
> run under Windows 9x/Me.  Stay tuned...
>
And MS-DOS?  I still use Zortech C to produce .COM files - shouldn't
be exactly a tall order.  CYGWIN isn't quite as slick, but it's
more consistent with what I find familiar.  Sorry, didn't meant to
be tangential, please let me know as you progress, I think the idea
is excellent.

Now that I think about it, of course 8c is already being used for
that, I just need to figure out how.

> Yes, you'll have to use MS compilers for the VXD and EXE
> that get the show started, but that's it.  Unlike GCC, the
> pleasure of using 8c is real.
>
> I don't know about Nt/XP yet, but I'm guessing that they'll
> be harder.

I wouldn't touch 9x/Me if Nt/2000/XP (I'm guessing at the last two,
I'm time-warped with NT4.0SP6a) is available.  Too flimsy.  NT at
least stays up when a task fails, with the MS-DOS based OSes I can't
resist rebooting whenever something falls over.

++L

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[forsyth]

>>Let's get off the hobbyhorse of criticising all the alternative
>>platforms and actually co-operate with them or show that the Plan
>>9 way is as superior as we make it out to be.

you're possibly overlooking the possibility that some of us have
written things such as portable compilers (and other things)
ourselves for a good few other systems, and take that into account
when making comparisons -- it's not
just that this or that isn't Plan 9.

Re: [9fans] Rant (was Re: Plan9 and Ada95?)

[David Gordon Hogan]

> On Wed, Nov 07, 2001 at 12:54:15PM -0500, David Gordon Hogan wrote:
> >
> > It's not.  If anything, it's worse.
>
> That's cheap.  Much as I can easily agree that GNU bloat can be
> improved on, I don't see anyone getting it right.
>
> Maybe the approach is flawed, but the results are out there.  Unlike
> Plan 9's ?c, GCC has to deal not only with different architectures,
> but also different operating systems.
>
> The above comment is a symptom of a Plan 9 syndrome I, for one, am
> not proud of: "we didn't contribute to it, it can't be any good".

No, it's a symptom of ``I wasted weeks of my life struggling
with this awful code''.

> In the meantime, if I want to cross-develop for Windows or Linux,
> or any other established platforms, Plan 9 is just no use to me,
> while GCC and its offspring are.  I know what my choice would be,
> but it's no choice, is it?  Oh, I forget the Inferno tools, those

GCC (Cygnus) is useless for Windows.  There's no way to link
a VXD!  So if you're doing any _serious_ development, are you
going to use two compilers, just so you can have the dubious
pleasure of using GCC?  I don't think so.

I'm sorry to be so ascerbic here, but I am sick of hearing people
defending GCC.

Why don't you help us improve the Inferno tools instead of
complaining about them?

> Sorry for the rant, I really don't mean it in a personal fashion,
> but I also fail to see the benefit of just criticising without
> providing any suggestions on _how_ to improve the things that are
> being criticised.  Like, you can't exactly remove GCC from today's
> development environment, what will you put in its place?  Across
> the board?

You want suggestions?  We could force all those GNU people to read
Rob's essay on programming style, for starters.

I have a cunning plan to use 8c to generate files that will
run under Windows 9x/Me.  Stay tuned...

Yes, you'll have to use MS compilers for the VXD and EXE
that get the show started, but that's it.  Unlike GCC, the
pleasure of using 8c is real.

I don't know about Nt/XP yet, but I'm guessing that they'll
be harder.

[9fans] Rant (was Re: Plan9 and Ada95?)

[Lucio De Re]

I'm not inviting flames, here, I'm just voicing an opinion no one else
seems to share :-(

On Wed, Nov 07, 2001 at 12:54:15PM -0500, David Gordon Hogan wrote:
>
> It's not.  If anything, it's worse.

That's cheap.  Much as I can easily agree that GNU bloat can be
improved on, I don't see anyone getting it right.

Maybe the approach is flawed, but the results are out there.  Unlike
Plan 9's ?c, GCC has to deal not only with different architectures,
but also different operating systems.

The above comment is a symptom of a Plan 9 syndrome I, for one, am
not proud of: "we didn't contribute to it, it can't be any good".

In the meantime, if I want to cross-develop for Windows or Linux,
or any other established platforms, Plan 9 is just no use to me,
while GCC and its offspring are.  I know what my choice would be,
but it's no choice, is it?  Oh, I forget the Inferno tools, those
are extremely useful, but they just don't reach far enough,
unfortunately.  I think that is a good direction, but still falls
short of present developers' needs.

Russ considered using ?c to produce Linux executables, but how much
work is involved?  OK, so dynamic libraries are a bad idea, but
even producing static binaries is near as damn out of the question.

Then forsyth mentions in passing he resorted to redefining thread
code for FreeBSD because Posix threads wouldn't cut it, so why is
the new design not submitted for inclusion in *BSD code?

Let's get off the hobbyhorse of criticising all the alternative
platforms and actually co-operate with them or show that the Plan
9 way is as superior as we make it out to be.

Sorry for the rant, I really don't mean it in a personal fashion,
but I also fail to see the benefit of just criticising without
providing any suggestions on _how_ to improve the things that are
being criticised.  Like, you can't exactly remove GCC from today's
development environment, what will you put in its place?  Across
the board?

++L