Linux-Development-Sys Digest #241, Volume #6 Sat, 9 Jan 99 02:14:42 EST
Contents:
Linux Sound Engine (Ross Vandegrift)
Re: Open Configuration Storage - was Registry for Linux (George MacDonald)
Re: Why I'm dumping Linux, going back to Windblows (Brett W. McCoy)
Re: Linux Sound Engine (Luke Scharf)
! * Free Long Distance Phone Service * ! (ICBA)
Re: Registry for Linux - Bad idea (Frank Sweetser)
Re: disheartened gnome developer ("Robert J. Hansen")
Re: Registry for Linux - Bad idea (Leslie Mikesell)
Re: lock_kernel in 2.2.0pre4 (Andi Kleen)
Re: Linux Journaling Filesystem (Andi Kleen)
Re: Use of LDT (Andi Kleen)
Xfree86 3.3.3 good, Matrox good, Red Hat, not so good (Re: Matrox Millenium G-200
Drivers) ("Frank T. Lofaro")
Re: GUI, The Next Generation ("Sam Robb")
How to get CPU usage? (JiSook Kim)
Re: accessing user space memory (Bryan Hackney)
UltraIDE Pro PCI(tm) Controller drivers. (Supernews User)
Re: klogd problem with new kernels ("P.K.")
Re: disheartened gnome developer (Frank Sweetser)
----------------------------------------------------------------------------
From: Ross Vandegrift <[EMAIL PROTECTED]>
Subject: Linux Sound Engine
Date: Fri, 08 Jan 1999 23:46:39 -0500
Reply-To: [EMAIL PROTECTED]
I once had an idea, that was kinda trivial, while riding in the car with
a friend: what if we began dealing with sound samples the same way we
deal with 3D polygons? What if we rendered sounds like pixels? What
could be done with that? Tonite, months later, I thought about
something similar again. I envision a wrapper library for the OSS Sound
Modules to provide equal sound access to application, transparently.
The idea sprang after downloading esd, the Enlightened SounDaemon. Esd
runs in the background and any esd aware apps open esd, not the sound
device. The problem here is that this ties up the sound device
completely. So, I was thinking, why not make a real-time mixer
pipeline? It could replace the current kernel interface to provide
transparent support for existing applications, and simply run on top of
the OSS Modules. This engine would have a number of benfits that I can
see:
1) Sound becomes non-exclusive - an arbitrary number of apps can now
play sound through one device.
2) Real-time mixing - these sounds can be given an ``Alpha channel'',
and mixed before sending it to the sound card to create effects.
I am thining of seriously going ahead with this project, though I'd like
to hear opinions - is there a better way of doing this? Should it be
done? How should it be done?
--
Ross Vandegrift | Eric J. Fenderson
alt.binaries.punk: for those of us too
punk to pay money for the music.
------------------------------
From: George MacDonald <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Open Configuration Storage - was Registry for Linux
Date: Sat, 09 Jan 1999 04:50:26 GMT
Leslie Mikesell wrote:
>
> In article <[EMAIL PROTECTED]>,
> George MacDonald <[EMAIL PROTECTED]> wrote:
> >
> >Information should be defined in one place only, except where it
> >is replicated for performance and/or for service discontinuty reasons.
>
> How do you propose to deal with the case where you want to run
> multiple copies of the same program with different options? For
> example you want to test a change to sendmail.cf without breaking
> other instances of sendmail running in production? How do
> NT registry-oriented progams deal with this? Programs with
> initialization files nomally have a command line option to use
> an alternate.
I was planning on supporting named config sets, and yes settable via
command line. Or from a GUI config selector in a GUI app.
Sendmail config is more problematic, how do you handle different
locations on a mobile station? You need to adapt the config setting's
based on your current "venue". I was thinking about defining contexts
that allow switching config evaluation & precedence on the fly!
Apps would need to register for notifications. Daemons would want
to be notified via signal, IPC, message, or perhaps direct kernel
intervention. The later is not to appealing but does have some benefits.
If I were modifying the daemons I would have them use /proc to
get any values that they need that are config sensitive. If performance
were an issue then I would set them up for a notifcation.
--
We stand on the shoulders of those giants who coded before.
Build a good layer, stand strong, and prepare for the next wave.
Guide those who come after you, give them your shoulder, lend them your code.
Code well and live! - [EMAIL PROTECTED] (7th Coding Battalion)
------------------------------
From: [EMAIL PROTECTED] (Brett W. McCoy)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Why I'm dumping Linux, going back to Windblows
Reply-To: [EMAIL PROTECTED]
Date: Fri, 08 Jan 1999 15:18:17 GMT
On 8 Jan 1999 02:03:14 GMT, Horst von Brand
<[EMAIL PROTECTED]> wrote:
>vi(1) is an excellent text editor: Small footprint, regular command syntax,
>suprisingly powerful. In the hands of somebody who knows it well. emacs(1)
>is also excellent, mostly because of the incredible versatility it gets by
>extensions written in elisp.
>
>None of them both is particularly easy to learn to use, but so what? If you
>want to control raw power, you'll have to invest some effort in taming it.
There are also smaller editors like pico, joe and jed, which are fine for
quick edits. Pico even has a spell checker.
--
Brett W. McCoy
http://www.lan2wan.com/~bmccoy/
=======================================================================
"The number of UNIX installations has grown to 10, with more expected."
-- The UNIX Programmer's Manual, 2nd Edition, June, 1972
=====BEGIN GEEK CODE BLOCK=====
Version: 3.12
GAT dpu s:-- a C++++ UL++++$ P+ L+++ E W++ N+ o K- w--- O@ M@ !V PS+++
PE Y+ PGP- t++ 5- X+ R+@ tv b+++ DI+++ D+ G++ e>++ h+(---) r++ y++++
======END GEEK CODE BLOCK======
------------------------------
From: Luke Scharf <[EMAIL PROTECTED]>
Subject: Re: Linux Sound Engine
Date: Sat, 09 Jan 1999 00:22:07 -0500
Ross Vandegrift wrote:
> 1) Sound becomes non-exclusive - an arbitrary number of apps can now
> play sound through one device.
> 2) Real-time mixing - these sounds can be given an ``Alpha channel'',
> and mixed before sending it to the sound card to create effects.
>
> I am thining of seriously going ahead with this project, though I'd like
> to hear opinions - is there a better way of doing this? Should it be
> done? How should it be done?
Check out http://www.be.com. The BeOS Media server sounds exactly like
what you are suggesting.
I wouldn't mind seeing something similar for Linix; Be, while very cool,
is not open source.
Hope this helps!
-Luke
------------------------------
From: [EMAIL PROTECTED] (ICBA)
Crossposted-To:
alt.cyberpunk,rec.sport.baseball,alt.cyberpunk.cypher,alt.cyberpunk.movement,alt.cyberpunk.tech,alt.kill.spammers,alt.games.duke3d,comp.os.linux.development.apps,rec.skiing.nordic,alt.sex.telephone
Subject: ! * Free Long Distance Phone Service * !
Date: Sat, 09 Jan 1999 05:22:00 GMT
Reply-To: [EMAIL PROTECTED]
INTERNET CONSUMERS BUYERS ASSOCIATION
8170 W. SAHARA AVE. SUIT 202
LAS VEGAS, NEVADA 89117
1-800-586-9648 TOLL FREE
1-702-320-8900 LOCAL
1-702-320-8903 FAX
[EMAIL PROTECTED]
Welcome to the Internet Consumer Buyers Association
Free Long Distance Phone Service:
TV and radio are Free because advertisers pay the bill so you can
watch and listen, but they sure hope you will buy their products.
Now advertisers want to pay your long distance phone charges just for
listening to their ads.
Call anywhere in the United States (including Alaska and Hawaii), even
from pay phones, anytime of the day or night! Listen to new
national advertisements on a toll free 800 number and you will be
given FREE long distance phone service. The more ads you listen to the
more long distance call time you receive. How many ads can you listen
to?
There are NO LIMITS, NO BILLS, NEVER PAY FOR A LONG DISTANCE CALL
AGAIN! Also listeners, while listening to ads, will be chosen at
random to receive Free Gifts and Merchandise. Coming soon for our
members will be FREE INTERNET ACCESS, FREE CELL PHONE SERVICE,
FREE SATELLITE TV (including equipment) and much much more!
BECOME A FREE CALLER TODAY!
If you are a member of the I.C.B.A., e-mail us ( [EMAIL PROTECTED] )
for your registration number to get into the Free Long Distance
Program. Or call us at the number above.
NON MEMBERS must fill out below application and call 1-800-586-9684
receive a registration number. Due to the administration and mailing
cost of membership packages a small processing fee will be asked of
all non-members.
========================================================================
-
MEMBERSHIP PRE APPLICATION
Registration #___________________________
1. Name:___________________________________________________________
2. Address:________________________________________________________
3. City/State/Zip:_________________________________________________
4. Home Phone:________________ Work:_________________ Fax:_________
5. E-Mail Address:_________________________________________________
6. Member#:_______________________ Refereed By#:____________________
7. Visa _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Exp. DATE _________
8. MC _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Exp. DATE _________
9. _____________________________________
__________________________________ _________________
SIGNATURE DATE
NON MEMBERS BEFORE SENDING IN PRE APPLICATION MUST CALL
1-800-586-9648 8am TO 8pm TO RECEIVE A REGISTRATION#. NO APPLICATION
WILL BE PROCESSED WITHOUT SIGNATURE AND REGISTRATION#!
------------------------------
From: Frank Sweetser <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Registry for Linux - Bad idea
Date: 09 Jan 1999 00:47:58 -0500
[EMAIL PROTECTED] (Leslie Mikesell) writes:
> In article <[EMAIL PROTECTED]>,
> George MacDonald <[EMAIL PROTECTED]> wrote:
> >
> >Information should be defined in one place only, except where it
> >is replicated for performance and/or for service discontinuty reasons.
>
> How do you propose to deal with the case where you want to run
> multiple copies of the same program with different options? For
> example you want to test a change to sendmail.cf without breaking
> other instances of sendmail running in production? How do
> NT registry-oriented progams deal with this? Programs with
> initialization files nomally have a command line option to use
> an alternate.
no idea about NT, but this design allows values to be overriden relativly
easily.
--
Frank Sweetser rasmusin at wpi.edu fsweetser at blee.net | PGP key available
paramount.ind.wpi.edu RedHat 5.2 kernel 2.2.0pre5ac1 i586 | at public servers
It's not really a rule--it's more like a trend.
-- Larry Wall in <[EMAIL PROTECTED]>
------------------------------
From: "Robert J. Hansen" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.x
Subject: Re: disheartened gnome developer
Date: Fri, 08 Jan 1999 23:53:49 -0600
> he's saying that troll tech is a company. how do you get he's
> anti-capitalist from that?
First off, this is more for the original poster than for you, Frank: I
think you already understand this. :) The original poster seems to be
in the dark as to how economics works in the free software market.
>> if eveything is free, companies will close, and there will be no
>> jobs.
Well, yeah, Jedi's right here -- except that the product *isn't* free.
What he thinks is the product is really just the advertising for the
product.
Free software can be just that -- free. He's right in that you can't
make money off free (free beer) software. So don't deal in free
software. Red Hat doesn't deal in free software: they deal in
SUPPORTING free software and DEVELOPING new free software. People will
pay money for support, and they'll pay money to have access to better
software.
Red Hat distributes their software free of charge, not as a product but
as a *promotional offer*. I've got Red Hat and Debian CDs on my desk
right now: I could install Debian, then call Red Hat and ask them for a
support contract *and they'd probably sell me one*. (They might hem and
haw a little bit at supporting non-Red Hat stuff, but that's only
because their support staff know RH a lot better than Debian.) That's
Red Hat's business: support. I've got money, which they want, and
they've got the technical know-how, which I want. So we exchange goods
and everybody's happy.
So why do they distribute Red Hat for free? -- As a promotional offer.
I could get a SuSE distro and ask Red Hat for support, but how would I
know to go to Red Hat? Compare this to if I get a $1.50 Red Hat CD-ROM
from CheapBytes and find a piece of documentation on it that says "For
$39.95 you can buy Official RH Linux which includes support", I'm more
likely to eventually buy it -- the support, **not** the free software
which is packaged along with the support as a freebie.
It's just like Microsoft and WinNT 5.0 betas. Microsoft isn't
distributing them just to fix bugs, but to entice consumers to purchase
a different, *more profitable* product -- the final of WinNT 5.0.
Free Linux CDs are cheap advertising for the *real* product a company
offers, which is usually support and/or development.
------------------------------
From: [EMAIL PROTECTED] (Leslie Mikesell)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Registry for Linux - Bad idea
Date: 8 Jan 1999 22:27:26 -0600
In article <[EMAIL PROTECTED]>,
George MacDonald <[EMAIL PROTECTED]> wrote:
>
>Information should be defined in one place only, except where it
>is replicated for performance and/or for service discontinuty reasons.
How do you propose to deal with the case where you want to run
multiple copies of the same program with different options? For
example you want to test a change to sendmail.cf without breaking
other instances of sendmail running in production? How do
NT registry-oriented progams deal with this? Programs with
initialization files nomally have a command line option to use
an alternate.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: lock_kernel in 2.2.0pre4
Date: 08 Jan 1999 16:11:29 +0100
In article <[EMAIL PROTECTED]>,
David Grothe <[EMAIL PROTECTED]> writes:
> With SMP set I notice that the lock_kernel function in
> include/asm/smplock.h actually performs a lock acquisition. I also
> notice that every system call that I have looked at has a lock_kernel at
> the beginning.
> This is the so-called "bit kernel lock".
> I thought that in the 2.2 kernels there was a finer grain of locking.
> This locking implies that there is only one CPU allowed to execute in
> the kernel at any one time.
> I am I misreading this? If so, how so? If not, what is the story?
Most of 2.2 still runs inside the giant lock. There are still
big SMP improvements over 2.0: the scheduler and the interrupt/bottom half
system are fine grained locked, interrupts may be processed on another
CPU if the current CPU is busy in the kernel (and it'll even be forwarded
to this CPU if it is a hardware interrupt). Bottom halves only still run
one at a time though, because that is a fundamental property of them ("
bhs are atomic to themselves")
A lot of code paths are much more SMP friendly in 2.2 compared to 2.0
because instead of global cli/sti - which is rather costly on SMP -
they just disable the local interrupts and use a spinlock to guard against
interrupts on the other CPUs.
If you profile a busy SMP 2.2 system you'll find that is spends a lot of
time spinning in the kernel spinlock, but overall SMP Linux doesn't look
to bad compared with other more finegrained locked systems like Solaris,
because in Linux most code paths are shorter ("less fat") so in practice
less time is still spend in kernel. It all depends on your application of
course, but never come to a conclusion without profiling/benchmarking first!
-Andi
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: Linux Journaling Filesystem
Date: 08 Jan 1999 15:58:36 +0100
In article <[EMAIL PROTECTED]>,
Michael Pronath <[EMAIL PROTECTED]> writes:
> Hi,
> Windows NT has a journaling filesystem, as well as the new
> OS/2 Warp Server. Is there anything like that for Linux yet
> or a project to create some?
> Is there a webpage with an introduction into the concepts of
> journaling filesystems?
Why do you want something if you don't know what it is?
-Andi
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: Use of LDT
Date: 04 Jan 1999 01:26:51 +0100
In article <[EMAIL PROTECTED]>,
Simon Berg <[EMAIL PROTECTED]> writes:
> The problem is if I get a context switch while in the BIOS code, will
> the scheduler handle this and will I get the same LDT back when the
> process gets the CPU back?
Linux supports private LDTs, but only when associated with a thread/process.
So you have to either create your own kernel thread for your driver, install
the LDT there and then do all operations from it, or do the LDT switching
by hand. The driver might get called from multiple processes that could
have their own LDT. Inside the kernel there is cooperative multitasking so
doing it by hand would work too (but the thread is probably more elegant)
-Andi
------------------------------
From: "Frank T. Lofaro" <[EMAIL PROTECTED]>
Subject: Xfree86 3.3.3 good, Matrox good, Red Hat, not so good (Re: Matrox Millenium
G-200 Drivers)
Date: 9 Jan 1999 04:35:47 GMT
In <[EMAIL PROTECTED]> on Mon, 04 Jan 1999 01:34:31 -0500, James A.
Cleland <[EMAIL PROTECTED]> wrote in comp.os.linux.development.system:
>JS wrote:
>> I have Red Hat v. 5.1. I would like to run Xwindows but I can't figure out
>> how to configure my video card correctly. It is an 8MB Matrox Millenium
>> G-200 (AGP bus).
>The XFree86 3.3.2 version that ships with RH5.1 doesn't support the G200
>chipset AGP. What you can do is update your XFree to 3.3.3. I'm pretty sure it
>supports the G200 AGP. Check out the xfree86.org website and look at the
>release notes for Matrox cards. I'll bet you find your card is supported in
>3.3.3. I would also bet that someone else will respond to this post and
>confirm this in short order.
Matrox support in XFree86 3.3.3 is much better than in 3.3.2. 3.3.2 is
what Red Hat 5.2 also ships with, unfortunately. Red Hat 5.2 is still
the newest Red Hat, unfortunately. If you upgrade XFree86 to 3.3.3
using an install of XFree86 you lose all the Red Hat X defaults and
setup. I don't know how to avoid that. Other than JUST installing the
new X _server_ and having a hybrid 3.3.2/3.3.3 system. (seems to
work. I think running the new xf86config program is a good idea, if I
remember correctly)
Red Hat doesn't separate the config stuff they add from the base X
stuff like they should (this applies in general also; e.g. install a new
kernel, and lose, install a new version of a package the rest of the
system uses, but for which you don't have an rpm, and lose, do much
sysadmin stuff, and lose).
I like a lot of Red Hat stuff, but this annoys me no end.
------------------------------
From: "Sam Robb" <[EMAIL PROTECTED]>
Subject: Re: GUI, The Next Generation
Date: Fri, 8 Jan 1999 13:12:05 -0500
Ken Sorensen wrote in message ...
>What I want to throw onto the table is this. What kind of UI would be the
>"right" UI for the Next Generation UI? If we came up with the next
>greatest GUI, then Microsoft, et. al. would be the ones playing catch-up.
I want an accepted, standard, completely usable way for developers to build
tools that operate under any interface - command line, GUI, voice activated,
whatever; some kind of input manager.
An example:
appname -abcd filename
appname --gui -abcd filename
appname --voice -abcd filename
The first instance is a default command-line app, the second a gui-based
version of the same app, the third a voice-driven version of the same app. In
the first case, it just runs, doing whatever it does to filename, using
options abcd. In the second, it opens up filename in whatever the gui is,
with options abcd set, ready to do something. In the third, it runs the app
under a voice-driven interface, again with filename open and options abcd set,
ready to do something. The only real difference between the command line
version and the gui/voice versions is that for the command-line version, some
action is taken immediately; in the other two, the app is put into the
specified state, and then waits for user input.
Comments? Thoughts?
-Samrobb
------------------------------
From: JiSook Kim <[EMAIL PROTECTED]>
Subject: How to get CPU usage?
Date: Sat, 09 Jan 1999 15:34:49 +0900
How to get CPU usage?
------------------------------
From: Bryan Hackney <[EMAIL PROTECTED]>
Subject: Re: accessing user space memory
Date: Fri, 08 Jan 1999 12:39:52 -0600
Reply-To: [EMAIL PROTECTED]
I don't think using a big kernel buffer is a bad idea, especially if this is
a custom application and memory is plenty. Use kernel virtual memory if you
wish. Using real kernel memory might simplify the page fault handler.
I would write a very simple module that only allocates memory and handles the
mmap calls.
I don't have experience with userfs, but I did something simliar recently.
Make the kernel buffer as big as you need it. I assume performance is the
main requirement here.
Once all the pages are mapped from user space to your kernel buffer, you
should see a tremendous increase in performance.
The following is a page fault handler that works with a large contiguous
kernel buffer.
BH
unsigned long zz_vm_nopage( struct vm_area_struct * vma, unsigned long addr, int
w_access ) {
unsigned long offset = addr - vma->vm_start + vma->vm_offset;
void * pageptr = 0;
int minor = MINOR( vma->vm_inode->i_rdev );
if ( offset >= zzdev[minor].mem.sz ) return 0;
printk( "<7>zz: nopage() %x %x %x\n", (int)addr,(int)w_access, (int)offset );
return (unsigned long)( zzdev[minor].mem.aligned_addr + offset );
}
Andreas Buschmann wrote:
>
> [this message is also send to the linux-userfs mailing list.]
>
> Hello,
>
> I have written a filesystem handling video files using userfs-0.9.4.2 on
> linux kernel v2.0.35.
...
--
Bryan Hackney / BHC / bhackneyatexpress-news.net
*
* Failure teaches only what not to do next time.
*
* Would you trust your mission-critical computing to a company
* that sells stuffed toys?
------------------------------
From: [EMAIL PROTECTED] (Supernews User)
Subject: UltraIDE Pro PCI(tm) Controller drivers.
Date: Fri, 08 Jan 1999 11:06:08 -0800
Hi.
I use an UltraIDE Controller from SIIG for my computer because my older bios
doesn't take
hard drives bigger than 2gigs. The problem is that my RedHat Linux 4
doesn't see that hard
drive because I don't have the drivers for that card. I was just wondering
whether anybody
knows of any drivers for Linux for that card and where I could get them.
Thanks.
Stan.
-**** Posted from Supernews, Discussions Start Here(tm) ****-
http://www.supernews.com/ - Host to the the World's Discussions & Usenet
------------------------------
From: "P.K." <[EMAIL PROTECTED]>
Subject: Re: klogd problem with new kernels
Date: Sat, 09 Jan 1999 05:38:23 +0100
"Michael E. Guo" wrote:
> I began to use new kernels (2.2.0-pre?) my klogd gives strange messages
> like "Could not find map file" or "Error seeking in /dev/kmem" something
> like that, I've download the source of sysklogd-1.3 from tsx11 but
> cannot compile it with new kernel,
> because it's too old (1996) and many structs has changed in
> linux/module.h. It's the reason that klogd's strange behaviour.
> Is there a updated sysklogd? or any patch to solve this problem?
Hi!
You should probably use a source package from your distribution.
Or, if unknown, take the Debian one.
Bye, Peter.
------------------------------
From: Frank Sweetser <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.x
Subject: Re: disheartened gnome developer
Date: 08 Jan 1999 13:05:12 -0500
[EMAIL PROTECTED] writes:
> In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] (jedi) wrote:
>
> > Troll Tech, whatever you or I might think of them, are
> > still just a company. Idealism and wishful thinking
> > won't alter that or change the constraints that companies
> > operate under.
>
> Troll Tech, whatever you or I might think of them, is 7 persons,
> some of which I know and consider my friends. It is not some sort
> of faceless multinational corporation.
>
> I have seen them do a lot more that what was needed as a company, just
> because they have ideals. They have always gone the extra mile just
> out of kindness. Dammit, if I were them, and had taken as much sillyness
> as I know they have, I would have kicked the board and gone home long ago.
>
> Please notice I have no commercial relationship with TT, they owe me
> nothing, and I owe them just appreciation for giving things away.
indeed, i was very impressed by their change of license. when they changed
a license that was already much more open than most, simply because of
people requesting it - not monetary reasons - they went up several notches
in my book.
on top of that, Qt appears to be a *very* nice library =)
--
Frank Sweetser rasmusin at wpi.edu fsweetser at blee.net | PGP key available
paramount.ind.wpi.edu RedHat 5.2 kernel 2.2.0pre3 i586 | at public servers
Sam: What's going on, Normie?
Norm: My birthday, Sammy. Give me a beer, stick a candle in
it, and I'll blow out my liver.
-- Cheers, Where Have All the Floorboards Gone
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
