Linux-Advocacy Digest #674, Volume #32 Tue, 6 Mar 01 14:13:03 EST
Contents:
Re: GPL Like patents. (Roberto Alsina)
Re: GPL Like patents. (Roberto Alsina)
Re: KDE or GNOME? (Bob Hauck)
Re: Sometimes, when I run Windows... (Pete Goodwin)
Re: Sometimes, when I run Windows... (Bob Hauck)
Re: Sometimes, when I run Windows... (Karel Jansens)
Re: Sometimes, when I run Windows... (Karel Jansens)
Re: It's here! IBM's new Linux ad! (Henry_Barta)
New Source for Open Source Applications - Easy Install (Open Country)
Re: GPL Like patents. (Craig Kelley)
Re: GPL Like patents. (Craig Kelley)
Re: GPL Like patents. (Roberto Alsina)
Re: KDE or GNOME? (Craig Kelley)
Re: Sometimes, when I run Windows... (Donovan Rebbechi)
Re: KDE or GNOME? (Donovan Rebbechi)
Re: definition of "free" for N-millionth time (Donovan Rebbechi)
----------------------------------------------------------------------------
From: Roberto Alsina <[EMAIL PROTECTED]>
Subject: Re: GPL Like patents.
Date: Tue, 6 Mar 2001 13:35:10 -0300
mlw wrote:
> Roberto Alsina wrote:
>> http://mail.gnu.org/pipermail/gnu-misc-discuss/1998-May/013912.html
>
> Clearly his statements are NOT covered by the current GPL, he is even
> clear about how this scenario is not described in the document.
He surely seems to be speaking about the current GPL. After all, this was
in 1998, and the document starts with "People have been talking about how
the GNU GPL applies to plug-ins". He is certainly referring to the GPL v2,
as we all know and love.
> Legally, if it isn't explicitly limited, it isn't limited. His conclusions
> are not supported by the GPL which he authored. The program vendor 'B'
> argument is in no way disallowed or even provided by the current GPL.
What Program vendor B? You mean "Party Two", author of "Plugin B"?
Anyway, I can't parse your sentence.
> I hadn't seen that last statement, it is chilling.
>
> I can see a "GPL Classic" movement.
>
> I also take issue with the statement:
>
> "The key to answering this is to realize that there is no legal
> distinction between dynamic linking and static linking."
>
> This is clearly incorrect, there is sufficient copyright law which states
> that I can reference copyrighted material without being in violation of
> copyright, but if I incorporate copyrighted material I am in violation."
>
> This is very analogous to dynamic vs static linking.
Amateur lawyering also claimed deCSS would never be ruled against in court,
you know.
--
Roberto Alsina
------------------------------
From: Roberto Alsina <[EMAIL PROTECTED]>
Subject: Re: GPL Like patents.
Date: Tue, 6 Mar 2001 13:37:26 -0300
mlw wrote:
> Roberto Alsina wrote:
>
>> I don't see support for yours in the GPL either. Are you willing to be
>> sued on your opinion?
>
> I don't see how. I have quoted the GPL in precisely the places where it is
> supported.
And everytime you quoted section 2, I have pointed out that your opinion is
based on a circular reasonaing, if anything.
You say that the app linked to a GPLd library is an independent work
because it is dynamically linked, and you say that dynamically linked apps
are not part of the larger work because they are reasonably independent.
That has no foundation whatsoever, you are trying to pull yourself up from
your shoelaces.
--
Roberto Alsina
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: KDE or GNOME?
Reply-To: hauck[at]codem{dot}com
Date: Tue, 06 Mar 2001 16:43:39 GMT
On 6 Mar 2001 02:38:50 GMT, Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
>On 05 Mar 2001 18:29:11 -0700, Craig Kelley wrote:
>>object->method ( really really really long arg one,
>> arg two,
>> result of object2->method( sub arg one,
>> sub arg two,
>> sub arg three
>> ),
>> a fourth arg
>> );
>>
>>In python, that is one huge line to fit in 80 columns.
>
>So don't fit it in 80 columns:
>
>object->method ( blah blah blah blah blah \
> arg two, \
> ... \
>
> )
You don't even need continuations for that. Python is smart enough to
figure it out. This works fine:
# snippage
#
elif (dt.calltype == INCOMING):
#
# Try to match the digits we have so far.
#
res = xlate.match (call_handle_2_port (handle),
dt.ts,
dt.destination_addr)
> My main problem with it for big projects (apart from speed critical
> ones, that is) is the fact that it's too sloppy as a language. It
> doesn't have real privacy, its syntax is even worse than that of C++
> (a remarkable achievment indeed!),
I also prefer Python because of the same issues.
--
-| Bob Hauck
-| Codem Systems, Inc.
-| http://www.codem.com/
------------------------------
From: Pete Goodwin <[EMAIL PROTECTED]>
Subject: Re: Sometimes, when I run Windows...
Date: Tue, 6 Mar 2001 16:38:52 -0000
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
says...
> >Because other people have complained of the same problems!
>
> Done several Linux installs, on about 25 different hardware configurations,
> including laptops, custom builds, Dell machines, and dreaded laptops.
> Never had hardware problems with supported hardware.
So what? I do many installations on Windows without any problems - yet
others do the same and have horrendous problems. What you find (and I
find) may not match other people's experience.
> >> Why did mine every time I have installed Linux?
> >
> >Why does Windows install perfectly every time I do it?
>
> Probably because you use hardware that works well with Windows.
Pretty much any (PC) hardware works well with Windows as everyone is
trying to support it. The same cannot be said for Linux.
> >Funny, in order to get my monitor to work on Linux, I had to at least
> >give it a basic specification.
>
> Only sync and resolution (and if you overspecify, modern monitors will
> shutdown automatically.)
So you agree you have to tell the system _something_ at least!
> >> It's just a fscking moniter. Moniters do not need drivers.
> >
> >The OS needs to know what kind of monitor you have attached. Same thing.
>
> Not exactly. It needs to know the resolution and the sync rate. Actually,
> if you tell the OS the size of your monitor, it should be able to guess
> some suitable modes (for example, any decent 19" monitor can do
> 1280*1024 @ 75 )
As above.
--
---
Pete Goodwin
All your no fly zone are belong to us
My opinions are my own
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: Sometimes, when I run Windows...
Reply-To: hauck[at]codem{dot}com
Date: Tue, 06 Mar 2001 16:59:55 GMT
On Mon, 05 Mar 2001 21:44:09 GMT, Pete Goodwin
<[EMAIL PROTECTED]> wrote:
>And how _old_ is man now? Does it have linking and cross referencing
>yet?
It does if you view the man pages using, say, the KDE help viewer.
--
-| Bob Hauck
-| Codem Systems, Inc.
-| http://www.codem.com/
------------------------------
From: Karel Jansens <[EMAIL PROTECTED]>
Subject: Re: Sometimes, when I run Windows...
Date: Tue, 06 Mar 2001 12:41:12 +0100
Pete Goodwin wrote:
>
> In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
> says...
> > > I have no interest in the raw workings of the machine I am using.
> > > Consider this - whilst driving your care, are you aware of what is
> > > happening in the engine? Is it relevant to you as you turn into
> > > somewhere?
> > >
> > I do like to have the occasional peek at RPM, engine temp, fuel level,
> > oil pressure and amps of the battery. Not constantly, mind (that would
> > mean my car trip would be quickly over), but it _is_ nice to know
> > everything is hunkadory under the hood.
>
> Whilst I'm driving it, I don't pay too much attention to what's going on
> under the hood.
>
> > So, I suppose you just drive around in your car until it stops, and then
> > you call a mechanic, right?
>
> Pretty much. I like to use the thing, not spend hours tinkering with it.
>
Neither do I. But a simple thing like an engine temperature gauge can
prevent an unfortunate nuisance (a broken thermostat) turning into an
extremely expensive engine replacement job. Obviously, one needs to
learn what the "cryptic" output of the thermometer means.
> > Or buy a new car?
>
> Funny, I did just that!
>
Good for you. I do hope it was not because your old one ran out of fuel.
> > > So we jump from 'cryptic' to 'learning stuff'. I see.
> >
> > Heh. _Everything_ is cryptic until you learn what it means. (you could
> > at least put a man between your goal posts)
>
> After learning what it does, it is still cryptic.
>
That's an oxymoron.
> > > And can I buy such a system ready made?
> >
> > Depends on how much you're willing to pay for one...
>
> The same price as a Windows machine!
>
OK. Let's talk business then. Assuming you already have the pc, does
that mean you are willing to pay the same amount as you would have paid
for the programs on your Windows pc to have a linux pc with equivalent
programs and a consistent interface?
Um, BTW, you do know that this consistence you seek is by no means
guaranteed under Windows, yes?
> > But as it is linux you are criticising, the proper question would have
> > been: "And can I easily build such a system?"
>
> I ask ze questions!
>
You'd better, 'cuz you're no good answering them. <G>
> > > I switched from the Archimedes RISC OS to Windows for a couple of
> > > reasons: (i) floating point hardware on the Intel chips and (ii)
> > > applicatons.
> >
> > Does that mean that there was an application you needed to run that
> > wasn't available for the Archimedes, or did you just switch "because
> > Windows has more applications"?
>
> The most popular platform always has more applications.
>
See my line above.
> > The former is what I said; the latter is naff.
>
> You may not like it, but the most popular platform always had more
> applications. It's naff in that respect, but that's life.
>
Success = popular ?
> > > > The measure of success are viri, spam and monopolies????
> > >
> > > Unfortunately, yes.
> >
> > Does it happen often to you that you let a marketing department hand you
> > definitions of words?
>
> I've never let a marketing department hand me definitions of words. You
> do realise I was being sarcastic, no?
>
I dunno.
"Success" is a relative word: it is basically meaningless unless you
know what was aimed for. One of the really good tricks of marketing and
advertising is turning relative words into absolute ones, thus creating
good-sounding, but meaningless battle-cries.
"Windows is successful!" should be immediately followed by: "At what?"
As it turns out, your reply appears to be: "At being popular".
--
Regards,
Karel Jansens
==============================================================
"You're the weakest link. Goodb-No, wait! Stop! Noaaarrghh!!!"
==============================================================
------------------------------
From: Karel Jansens <[EMAIL PROTECTED]>
Subject: Re: Sometimes, when I run Windows...
Date: Tue, 06 Mar 2001 12:49:34 +0100
Pete Goodwin wrote:
>
> In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
> says...
>
> > > "A picture paints a thousand words".
> > >
> > > Ever tried to edit a graphic with a CLI?
> >
> > Have you ever run Autocad?
>
> Nope, can't say I have.
>
> Ever tried to edit a graphic with a CLI?
>
Autocad creates graphical outputs; OK, so they're vector-based, and it
doesn't "do" your granny's holliday pics, but it _is_ a graphics program
and a very popular one too. Autocad has always had the option to draw
and edit from a command-line.
I've used two programs from Autodesk (makers of Autocad): Home, a
DOS-based subset of Autocad, aimed at - indeed - architectural drawings,
and 3D-Concepts, a Windows 3.x program for creating and editing
3D-objects. Both were not only command-line enabled, but were in many
instances actually easier to use than with the mouse.
So yes, I have edited graphics with a CLI and yes, it was easier than
with a mouse.
--
Regards,
Karel Jansens
==============================================================
"You're the weakest link. Goodb-No, wait! Stop! Noaaarrghh!!!"
==============================================================
------------------------------
From: Henry_Barta <[EMAIL PROTECTED]>
Subject: Re: It's here! IBM's new Linux ad!
Date: 6 Mar 2001 18:00:48 GMT
Brent R <[EMAIL PROTECTED]> wrote:
> Exactly why is IE so bad? I find it much nicer to use than NS, hell even
> the Linux version of NS sucks (I love how it opens downloadable binaries
> as web pages).
I tried downloading the Solaris CDs from Sun a couple of days
ago. My first attempt was using IE on Win2000. Even though IE
blythely proclaimed "Done!", two out of the three files were
short and could not be used. I finally went to Opera on Linux
(practically a beta product) and downloaded the other two CD
images on the first try and with no problem.
So "I love how it" can't download the entire binary if it is
large.
--
Hank Barta White Oak Software Inc.
[EMAIL PROTECTED] Predictable Systems by Design.(tm)
Beautiful Sunny Winfield, Illinois
------------------------------
From: Open Country <[EMAIL PROTECTED]>
Subject: New Source for Open Source Applications - Easy Install
Reply-To: [EMAIL PROTECTED]
Date: Tue, 06 Mar 2001 18:03:21 GMT
Hey, new Linux desktop application installer is available. Check it
out and let us know what you think. http://www.opencountry.net
Our company helps the less technical to use and enjoy Linux by
repackaging open-source software in an easy install, QwikClick(tm),
format. The base application remains unchanged. It's like putting
the application on a CD for distribution, except that it will be
downloaded and installed over the internet. A user will be able to
subscribe to the store, and as a subscriber obtain and install a
variety of common applications or games with only one click. A
nominal fee is charged for the install convenience and a part of the
revenue obtained is sent back to the open-source developer or his
designated charity.
Consider this an open-source effort, check out our web site,
http://www.opencountry.net, and tell us your reactions. If you have
authored an open-source product that you'd like us to consider for
non-exclusive redistribution, send us e-mail at
[EMAIL PROTECTED]
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Subject: Re: GPL Like patents.
Date: 06 Mar 2001 11:13:55 -0700
mlw <[EMAIL PROTECTED]> writes:
> Roberto Alsina wrote:
> > > (2) Indicates that if you keep the GPL code separate from yours and your
> > > code is not merely an extension, yours need not be GPL.
> >
> > As I told you half a dozen times: RMS says that if your code links to the
> > GPLd code, it can not be reasonably considered separate. Why do you say
> > when it links dynamically it is separate?
>
> Dynamically linking to GPL code does not include the GPL code into
> your binary where statically linking does.
>
> This is vital to understanding the difference. When something is
> statically linked, it creates one binary. When something is
> dynamically linked, you have multiple "independent" binaries. That
> is the key.
What's the difference? They do exactly the same thing.
What about Tom Christansen's GNU condom:
GPL Code <--> Free Shared Library <--> Non-GPL Code
Where the fictitious "Free Shared Library" simply translates object
symbols from one to the other?
What about SOAP calls? If I make an XML request to a GPL'd SOAP
server, then how is this different than making a static library call?
> Static and dynamic linking are fundamentally different processes. It
> isn't just storage.
But they aren't. In the first case you have ld link your code, and in
the second case... ld links your code as well. It's only a matter of
when it happens. Besides, I think you're misunderstanding the GPL
here. If you link against a GPL work, whether it's statically or
dynamically then you must GPL your own work as well. Only the LGPL
affords protection otherwise.
> The GPL does say this, it lays out a framework in which GPL can be
> used by non-GPL code. The "litmus test" is that the works can be
> "reasonably considered independent." A shared library can be built
> completely independent of your non-GPL code. A static link
> incorporates the GPL code into your code, thus eliminating a
> reasonable definition of "independent."
>
> You are even allowed to ship the GPL code with your application and
> not be required to make yours GPL.
>
> I guess I understand why there is confusion. It is perfectly obvious
> to me, being a software developer, that dynamic linking avoids the
> GPL. I can create a program which contains NO GPL code, ships with
> NO GPL code, and yet can use GPL code installed on a users system
> when it gets there.
Then why does libstc++ have a special clause which allows non-GPL
works to be dynamically linked against it? I worked with a Linux game
developer who was extremely worried about that issue (for Shogo:
Mobile Armour Division), and we had to ask the libstdc++ folks about
it.
If it were otherwise licensed, there would have been no worries.
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Subject: Re: GPL Like patents.
Date: 06 Mar 2001 11:28:49 -0700
Roberto Alsina <[EMAIL PROTECTED]> writes:
> mlw wrote:
>
> > Roberto Alsina wrote:
> >
> >> I don't see support for yours in the GPL either. Are you willing to be
> >> sued on your opinion?
> >
> > I don't see how. I have quoted the GPL in precisely the places where it is
> > supported.
>
> And everytime you quoted section 2, I have pointed out that your opinion is
> based on a circular reasonaing, if anything.
>
> You say that the app linked to a GPLd library is an independent work
> because it is dynamically linked, and you say that dynamically linked apps
> are not part of the larger work because they are reasonably independent.
>
> That has no foundation whatsoever, you are trying to pull yourself up from
> your shoelaces.
Then why did the libstdc++ folks put an exception into their license?
http://gcc.gnu.org/ml/gcc/2000-06/msg00426.html
(for those that don't know, libstdc++ v3 is GPL and not LGPL like it's
predecessors)
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: Roberto Alsina <[EMAIL PROTECTED]>
Subject: Re: GPL Like patents.
Date: Tue, 6 Mar 2001 15:30:02 -0300
Craig Kelley wrote:
> Roberto Alsina <[EMAIL PROTECTED]> writes:
>
>> mlw wrote:
>>
>> > Roberto Alsina wrote:
>> >
>> >> I don't see support for yours in the GPL either. Are you willing to be
>> >> sued on your opinion?
>> >
>> > I don't see how. I have quoted the GPL in precisely the places where it
>> > is supported.
>>
>> And everytime you quoted section 2, I have pointed out that your opinion
>> is based on a circular reasonaing, if anything.
>>
>> You say that the app linked to a GPLd library is an independent work
>> because it is dynamically linked, and you say that dynamically linked
>> apps are not part of the larger work because they are reasonably
>> independent.
>>
>> That has no foundation whatsoever, you are trying to pull yourself up
>> from your shoelaces.
>
> Then why did the libstdc++ folks put an exception into their license?
Actually, you should ask that to mlw, not me ;-)
He's the one saying dynamical linking saves you from GPL contamination.
--
Roberto Alsina
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Subject: Re: KDE or GNOME?
Date: 06 Mar 2001 11:32:29 -0700
[EMAIL PROTECTED] (Bob Hauck) writes:
> On 6 Mar 2001 02:38:50 GMT, Donovan Rebbechi <[EMAIL PROTECTED]> wrote:
> >On 05 Mar 2001 18:29:11 -0700, Craig Kelley wrote:
>
> >>object->method ( really really really long arg one,
> >> arg two,
> >> result of object2->method( sub arg one,
> >> sub arg two,
> >> sub arg three
> >> ),
> >> a fourth arg
> >> );
> >>
> >>In python, that is one huge line to fit in 80 columns.
> >
> >So don't fit it in 80 columns:
> >
> >object->method ( blah blah blah blah blah \
> > arg two, \
> > ... \
> >
> > )
>
> You don't even need continuations for that. Python is smart enough to
> figure it out. This works fine:
>
> # snippage
> #
> elif (dt.calltype == INCOMING):
> #
> # Try to match the digits we have so far.
> #
> res = xlate.match (call_handle_2_port (handle),
> dt.ts,
> dt.destination_addr)
Thanks for the information, that's neat. Does it also handle
multi-line strings and such as well? (without using the """ operator,
IIRC)
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Subject: Re: Sometimes, when I run Windows...
Date: 6 Mar 2001 18:35:14 GMT
On Tue, 06 Mar 2001 14:07:53 GMT, chrisv wrote:
>[EMAIL PROTECTED] (Donovan Rebbechi) wrote:
>
>>Done several Linux installs, on about 25 different hardware configurations,
>>including laptops, custom builds, Dell machines, and dreaded laptops.
>>Never had hardware problems with supported hardware.
>
>That's funny, I've had problems with the majority of the machines I've
>tried it on! Examples: Voodoo3 not supported. Adaptec (!) SCSI card
>not supported.
Hint: It's easier if you use supported hardware.
--
Donovan Rebbechi * http://pegasus.rutgers.edu/~elflord/ *
elflord at panix dot com
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Subject: Re: KDE or GNOME?
Date: 6 Mar 2001 18:43:50 GMT
On 06 Mar 2001 08:47:12 -0700, Craig Kelley wrote:
>[EMAIL PROTECTED] (Donovan Rebbechi) writes:
>> My main problem with it for big projects (apart from speed critical
>> ones, that is) is the fact that it's too sloppy as a language.
>
>Yes, but it can be made better by doing this at the top of every
>program:
>
>use strict;
>use English; (or whatever language you prefer)
>use Carp;
yes, and with the -w flag ...
>> It doesn't have real privacy
>
>Who are you trying to protect code from? If people use the published
>methods only then it should be fine.
What if they inherit a class, and they choose a name that clashes with
the name of a "private" data member in the base class ?
>> its syntax is even worse than that of C++ (a remarkable achievment
>> indeed!),
>
>(see above)
I was referring to the "lines noise".
On one hand, the language doesn't know the difference between a string and
an int. On the other hand, you have to use different line noise characters
for arrays, scalars, hashes and functions.
>But in the real world, that's just fine. It's only computers that
>care about the difference between the number 123 and the string
>"123".
I don't care about this. I'm more concerned about the difference between
123 and "123hello". Perl doesn't know the difference.
> C, C++, and Java all do implicit conversions between doubles,
>floats and ints and such, depending on the situation.
which is much less dangerous, since safe conversions are used (C and C++
just promote).
> You have to
>know your language -- and that will always be a quark. If you need to
>be pedantic at a certain point in your program then you can always use
>the OO objects for types in perl or run it through the regular
>expression test: unless (/^\d+$/) {croak "bad number"}
Yes, but then the supposed advantage that you can do more in less lines
of code sort of fizzes out. Also, you're still only checking at runtime,
there's no way to do so at compile time. For large projects, it's not good
to have a lot of potential runtime errors.
The real "advantage" of perl is that it's
easy to write sloppt code that has a low amount of error checking.
>> I didn't realise this. I heard about it, but I didn't hear about any
>> work being done. This would do a lot to deal with my performance
>> concerns (because I prefer C++ to C as a "fast" language)
>
>Shouldn't everyone? :)
>
>There's no reason to use C anymore unless you need to deal with legacy
>stuff.
Tell that to the theologists and the language luddites.
--
Donovan Rebbechi * http://pegasus.rutgers.edu/~elflord/ *
elflord at panix dot com
------------------------------
From: [EMAIL PROTECTED] (Donovan Rebbechi)
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy,misc.int-property
Subject: Re: definition of "free" for N-millionth time
Date: 6 Mar 2001 18:46:28 GMT
On Tue, 6 Mar 2001 15:55:18 +0100, Stefaan A Eeckels wrote:
>In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] (Jay Maynard) writes:
>> On Tue, 6 Mar 2001 09:38:24 +0100, Stefaan A Eeckels
>> <[EMAIL PROTECTED]> wrote:
>>>They have not. The GPL even stipulates that it's not concerned
>>>with the price, only with the continued availability of the
>>>source code, and all the derivative works of the source code,
>>>if these are distributed.
>>
>> Meaningless. Since there's nothing preventing someone from buying a copy and
>> turning around and giving it away for free, the market is effectively
>> destroyed.
>
>Cheapbytes hasn't "destroyed" the expensive Linux distributions.
>And yes, you won't make money on _every_ copy. That's something
>Microsoft experiences as well.
Linux distributions are not single pieces of software, so at best you are
using an analogy, and not a counter-example.
--
Donovan Rebbechi * http://pegasus.rutgers.edu/~elflord/ *
elflord at panix dot com
------------------------------
** 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 by posting to comp.os.linux.advocacy.
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-Advocacy Digest
******************************
