Linux-Advocacy Digest #317, Volume #34 Tue, 8 May 01 08:13:03 EDT
Contents:
Re: Microsoft's move away from perpetual licensing proves that the closed source
model doesn't work! ("Adam Warner")
Re: Linux a Miserable Consumer OS (Donn Miller)
Re: I regretfully conclude that Linux is a piece of CRAP. (Stuart Krivis)
Re: Windows makes good coasters (Chris Ahlstrom)
Re: Justice Department LOVES Microsoft! ("Edward Rosten")
Re: Justice Department LOVES Microsoft! (JamesW)
Re: Yet another IIS security bug (Giuliano Colla)
Re: Microsoft's move away from perpetual licensing proves that the (jtnews)
Re: Microsoft's move away from perpetual licensing proves that the (jtnews)
----------------------------------------------------------------------------
From: "Adam Warner" <[EMAIL PROTECTED]>
Subject: Re: Microsoft's move away from perpetual licensing proves that the closed
source model doesn't work!
Date: Tue, 8 May 2001 22:48:04 +1200
Hi jtnews,
> Microsoft's move away from perpetual
> licensing proves that the closed
> source model doesn't work!
>
> Read this article on CNET!
>
> http://news.cnet.com/news/0-1003-200-5851009.html
Yeah it's a fascinating news story. I don't think it shows that the closed
source model doesn't work though. Just that the market is saturated with
products that provide enough functionality for people to not need to
upgrade.
Instead of providing compelling reasons for large corporations to upgrade it
appears Microsoft wants to force upgrades by never allowing a lifetime
license in the first place.
> The inherent problem with the closed source
> model, is that any company that provides
> such software must come up with revised
> versions to generate a profit even if
> no further revisions are warranted or even
> desired by the people using the software.
That's just a problem with never being satisfied with a very profitable
business that is already fully servicing the market. Microsoft's wants to
keep growing and growing.
> With the open source software model, the public
> has the ability to fork off the code, keep
> their own versions, and hence retain the
> functionality and carry it to newer operating
> systems and other computing environments indefinitely
> as they see fit.
Certainly.
> So you see even Microsoft is admitting that
> the closed source model doesn't work.
Sorry I really don't think it follows. For example you can have closed
source subscription only software.
What it may show is that Microsoft's traditional business plan is in serious
trouble. If Microsoft isn't able to successfully transition to
subscription-based services then we may already be at the peak of
Microsoft's dominance.
Regards,
Adam
------------------------------
Date: Tue, 08 May 2001 06:55:39 -0400
From: Donn Miller <[EMAIL PROTECTED]>
Subject: Re: Linux a Miserable Consumer OS
Edward Rosten wrote:
centipedes? At one point in time, arthropods (of which the centipede is
> one) and vertibrates diverged from one point and gained their various
> skeletons. I doubt they split from centipedes since centipedes are land
> based arthropods and vertibrates started off as sea based life forms. I'm
> not an expert on this, and this is about the limit of my knowedge. I'd
> sugest you consult a paeleontologist.
Actually, centipedes all prefer very damp environments (except maybe the
house centipede, but even those prefer some dampness), so it's not
unreasonable to suspect that at one point in their evolution they were
water inhabiting.
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: [EMAIL PROTECTED] (Stuart Krivis)
Subject: Re: I regretfully conclude that Linux is a piece of CRAP.
Date: 8 May 2001 07:06:29 -0500
Reply-To: [EMAIL PROTECTED]
On Wed, 28 Mar 2001 20:28:28 GMT, Michael Marion <[EMAIL PROTECTED]> wrote:
> Brian Rourke wrote:
>
>> If Apple and Microsoft can make installation easier, why not emulate
>> that? Wouldn't you agree that easy installation is better than hard
>> installation?
>
> You should try a Solaris install onto a Sparc then. I've _never_ seen one
> fail (short of a hardware failure of course). The installation for Solaris 8
> is extremely simple, and you can even browse the net in a hotjava window while
> it's installing. :)
Sun has a known set of hardware to work with. Apple almost has it as
easy. MS probably has more people working on the install routines than
the total employees at RH, Mandrake, and SuSE put together.
It's a miracle that installs of Linux (and Windows) are as easy as they
are. The enormous variety of hardware used for PCs makes for an
interesting time. :-)
One good example is the NE2000 NIC chipset. Manufacturers seem to all do
weird things when they implement cards with these. They _should_ all be
the same, but they're not.
--
Stuart Krivis
------------------------------
From: Chris Ahlstrom <[EMAIL PROTECTED]>
Crossposted-To: alt.linux.sux,alt.linux,comp.os.ms-windows.nt.advocacy
Subject: Re: Windows makes good coasters
Date: Tue, 08 May 2001 11:11:00 GMT
Ayende Rahien wrote:
>
> "Tom Wilson" <[EMAIL PROTECTED]> wrote in message
> news:btqJ6.10$[EMAIL PROTECTED]...
> >
> > "Chad Myers" <[EMAIL PROTECTED]> wrote in message
>
> > > I have a PII-400 with 256MB RAM and I can play Unreal Tournament while
> > burning
> > > a CD.
> >
> > THAT...is BS.
>
> Why? CD-Burning is not such a heavy task at all. I usually burn CDs while
> I'm developing, browsing, while the computer is being worked on by 2 - 5
> other people, etc.
> I never had a coasted CD because of that, and my CDR is nearly three years
> old. 8x4x Panasonic.
It's not a heavy task, but it requires a fast response from the
computer. Windows sometimes takes time out to do its own thing.
Chris
--
Free the Software!
------------------------------
From: "Edward Rosten" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Date: Tue, 08 May 2001 13:14:41 +0100
>> Using fprintf's under Linux to print PS, you can print graphics from
>> the command prompt. Under Windows this won't work.
>
> *fprintf* ? How the hell does *this* work?
> fprintf takes a character array, and print it to a file, how do you take
> a char array and make it into a graphic picture?
You can access the device independent printer layer in Linux using
(f)printfs. In windows, you have to use the GDI to get to the device
independent printer layer.
>> Also, some printers do not accept plain text.
>
> Few, as you mentioned.
I'll neglect them from now on, since its so few.
>> How can you print device independent graphics in Windows without doing
>> the whole GDI thing to a command line program?
>
> Same as you would do with Linux, output PS, activate GS to translate to
> device dependent output. In this case, it would be to the GDI, which
> would translate it to device dependent output.
The probelm being here that windows does not allow arbitraty print
filters to be put in to the printing sysetm. With UNIX, you dump a file
to lpr and forget. The printer subsystem can then activate a filter based
on the file type. Windows lacking this step requires you to print to a
file, then run a program on that file (such as GS) in order to print it.
This is a real pain for batch work.
>> If you have a PS printer hooked up to LPT1 then you can, but this is a
>> special case.
>
> Then translate it like you do in Linux.
The translator is not automatically triggered as in Linux.
>> > I might have gotten the file name wrong, it has been a long time
>> > since I did printing this way.
>>
>> LPT1 refers to the first printer port, PRN to the default one. Windows
>> allows upto LPT9
>
> Yeah, like I said, it has been a *long* time since I dealt with DOS
> printing, and printing code in general isn't a topic I like very much.
The practicalities of DOS printing are similar to Linux printing from the
apps point of view. If your app dumps data to STDOUT, redirecting to PRN
deals with the printing. Under Linux, redirecting to lpr instead of a
device file does the same job. the difference is that DOS does not
support print filters (which is why all text files under DOS use CR/LF
pairs).
>> > BTW, you have to go to *great* lengths to find a portable C/C++
>> > program that is portable once you start using graphics, so it doesn't
>> > count.
>>
>> Not true. If all the graphics are off-line (ie printed only) then you
>> ca write programs in standard C and standard PS and they will anywhere
>> that has a PS interpreter and a standards conformant C compiler.
>
> Okay, not exactly what I call *using* graphics, but true. BTW, it isn't
> just PS, you could say the same about JPEG, frex. (not as good, of
> course, for the purpose we are talking about)
This is what I was generally referring to, since I actually generate
quite a lot of postscript graphics in noninteractive programs.
>> Interactive graphics are a pain in the arese, portability wise, though,
>> but not all graphics are interactive.
>
> Agreed, full heartedly.
:)
>> >> For Linux, you do the whole PS thing using nothing but printfs.
>> >> Linux wind hands down here because the app is now totally protable
>> >> (hell it can be written in totally compliant C).
>> >
>> > See above, and it's fprintf, not printf.
>>
>> You can use printfs, as ling as you | lpr or > PRN.
>
> No, printf print to STDIN, what you meant is that you can take program's
> output and redirect it to the printer, that is not a developer's worry,
> usually.
OK, but i often work like this since using printfs (ie to STDOUT) I have
to option of dumping it to a viewer or printer without making temporary
files. That's laziness, I suppose...
>> >> I actually write quite a lot of apps that do printing under Linux.
>> >> none of them use the GUI and they were all very easy to write.
>> >
>> > We are not talking about text printing here, you know. For text
>> > printing, you can just forward the input to the printer, and that is
>> > it. No hassle to the developer at all.
>>
>> I was not referring to text printing. I went to the effort to learn
>> basic PostScript and not I can do graphics printing from all my command
>> line apps that I need to print from, with graphics. I have used this
>> ability quite a lot.
>
> Point taken, personally, I learned ASCII art :)
It only goes so far... :-)
>> >> There is another problem in windows. Because the process is opaque,
>> >> there is no way of getting a device independent printer file. If you
>> >> print to a file under Windows, you get one for your printer. Under
>> >> UNIX you get PS, which you can then take to any other system and
>> >> print happily using the full capabilities of the device. This is a
>> >> feature I use a lot.
>> >
>> > Install a PS printer driver, print to file, done deal. And *I*, as
>> > the developer, don't need to know anything about PS to do it, too. It
>> > will create a perfectly legal PS file, too.
>>
>> Yes, then take that device independent printer file to another windows
>> machine with the printer you want and try to print it. You can't
>> without installing 3rd party apps. There is no support for having a
>> device independent printer file under windows, without installing third
>> party apps on every machine you need to use.
>
> Since Linux does it with 3rd part tools (can you do the same without
> GS?) I would say it's not fair. Given postscript driver & GS, I can do
> the same on windows. And that would work on *any* application that print
> on windows.
That is true in a way, but a little unfair in another. If you take both
OSs with a normal installation (ie, both including the standard printer
subsystem), it is easier under Linux. Generally, if there is a linux
computer hooked up to a printer, it will either be a PS printer, or GS
will be installed (likely enough that you could count on it). With
Windows, the chance of any given machine having a PS interpreter is
fairly low, since PS is not windows's standard print model.
One of my gripes with windows is that there is no practical way to get a
device independent printer file that will print on any machine you choose
to take it to.
I think WMF would be an entirely suitable format (based on my limited
knowledge of WMF), but most programs won't output WMFs and even if they
did, you can't dump one straight to a printer via the GDI.
>> >> > On Linux, however, you need to have seperate code that does it,
>> >> > which is much harder than adding few "if"s for page-breaks.
>> >>
>> >> Only for GUI apps.
>> >
>> > Since most (all?) printers can accept fprintf commands, then *of
>> > course*
>> most. Proably 99%
>
> Large enough precentage so we can ignore the bitmap only printers.
OK.
>> > we are talking about GUI apps. Or, at least, of printing non text
>> > stuff.
>>
>> I was talking about non text stuff.
>
> Okay.
>
>> I still repeat my assertion that printing under UNIX is more flexible
>> and that it is much easier to print graphics from a command line app
>> than under Windows. I won't comment about GUI apps since I have little
>> experience with windows GUI work.
>
> I think that you are wrong, because it's just as easy to print from the
> CLI on windows as it's on Linux, if you are going to output PS.#
The difference is that under Linux, you can dump PS to the printer and
get out graphics. In windows, you have to dump it to a file, then run the
interpreter on the file, as a seperate step.
> And
> easier if you are bypassing PS and going straight to GDI.
For command line apps?
> You could
> probably write a wrapper functions that would output PS or GDI,
> according to where you are compiling/running this application, without
> too much trouble.
-Ed
--
You can't go wrong with psycho-rats.
u 9 8 e j r (at) e c s . o x . a c . u k
------------------------------
From: JamesW <[EMAIL PROTECTED]>
Subject: Re: Justice Department LOVES Microsoft!
Date: Tue, 8 May 2001 12:17:59 +0100
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
says...
> Note that this will not be readable just as is by the sucking newsreader
> JS PL is using.
>
> Peter
>
> end
>
begin
^^
You need to put two spaces in front of the begin to fool OE. The
braindead newsreader expects an attachment - true if there was a matching
space-space-end somewhere after the space-space-begin. Since there is no
space-space-end working newsreaders ignore the begin and display the
text.
OE thinking there is an attachment will displays a paperclip instead -
another M$ product that substitutes useless paperclips for
functionality...
------------------------------
From: Giuliano Colla <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft
Subject: Re: Yet another IIS security bug
Date: Tue, 08 May 2001 11:39:44 GMT
"T. Max Devlin" wrote:
>
> Said Giuliano Colla in alt.destroy.microsoft on Sun, 06 May 2001
> >"T. Max Devlin" wrote:
>
> ['Intuitive' == 'Familiar'; a standard theory of mine]
>
> >I don't believe that 'intuitive' and 'familiar' are just the same thing,
> >on a human interface. We have a background, a psychology, a brain built
> >in a given way, etc. etc. So if a human interface has features which
> >recall other 'familiar' things (like pressing a button, following with
> >the eye a moving object, etc.) it's more 'intuitive' than another one
> >which doesn't recall previous 'familiar' behaviors.
>
> I think you're going to far, trying to say that because people can
> recognize "a button" and that pixels can represent 3-d buttons that
> "push in" then intuitive doesn't precisely equal familiar. To the
> contrary, the fact that these things are recognizable in any
> human/computer interface is because they are already familiar, as visual
> mimics of real-world objects. This background can be called psychology,
> and we can say it is 'built into the brain', but I think that's
> something of a mirage, actually. The fact is, it is built into the
> physical world, how objects work and how our eyes perceive them. To
> claim that because it is built into the brain it is intuitive is like
> saying because it is familiar, it is built into the brain, I think.
>
> >I have worked out my personal way to tell apart 'familiar' from
> >'intuitive'. If you are familiar with interface A, and then switch to B,
> >it will take some time to become familiar with B. But when you've become
> >familiar with both, it turns out that it takes you more time to get used
> >again to B after using A, than getting used to A after using B. It means
> >that A is more 'intuitive' than B. I've been using Windows since its
> >beginning, and only last year I started using KDE on Linux. But it takes
> >me seconds to feel at ease when I switch from Windows to Linux KDE,
> >while it takes me many minutes to get used again to Windows when coming
> >from Linux. I conclude that KDE is more 'intuitive' than Windows.
>
> Oh, christ, I LOVE IT! You've DONE it Giuliano; you've proven me wrong!
> THANK YOU!
>
> I think this means that the difference between 'intuitive' and
> 'familiar' is possibly "consistency", according to your description and
> my intuition. Does that make sense? Or are you still thinking that it
> is some 'consistency' between the interface and some platonic model of
> "the perfect archetypical human/computer interface' which exists
> inexplicably in our minds?
>
Well, I don't know if I've proven you wrong.
If you enlarge the concept of 'familiar' to encompass not
only computer experience, but all the previous experiences,
maybe your definition holds true. But it's very hard to
tell. Consistency is certainly a big issue. And that's
something related on how our brain works. But maybe there's
something deeper, very hard to define. We're used to infer
conclusions from insufficient data (that's what 'intuition'
means, on the end), but how it works is far from being
clear. An artists builds metaphors and analogies that make
us 'feel' what he means. How it's done we don't know.
Shakespeare is great, but no school can teach you to write
in such a way as to convey the same emotions to readers or
listeners.
If you take linguistic, you find a number of facts quite
interesting. Why the same word (to wait) means two
apparently unrelated things: to wait *on* someone and to
wait *for* someone? What has to do 'to take care of
something' with waiting? Well, this double meaning is common
in may other languages besides English. This is but one
example of a myriad of those cases. Apparently it has to do
with some basic concepts buried deeply in our mind.
You rightly described the Desktop as a metaphor, but
building the right metaphor, convincing for all listeners is
something not at all so obvious.
First of all it has to do with background and culture. A
mailbox or a trash bin convey a very precise idea, until you
go in a different country where mailboxes or trash bins have
a completely different shape.
It has to do with the degree of literacy of the user. If you
don't mind a digression (you may skip it if you do mind!) I
may tell you the experience of the 'communication error'
icon.
Some twenty years ago, while designing the first graphical
human interface for our control system, I decided to use
graphical symbols (the word 'Icon' was unheard of at that
time) to make things easier for operators, and also to avoid
the translation burden. We had only a text mode display, so
we used an alternate character generator to display
graphics. Most of the symbols were quite obvious, but then I
stumbled into a problem. Being a multiprocessor system, with
processors communicating via serial lines, you could
encounter some communication errors because of noise,
hardware failures and such. How do you show that? I thought
that a telephone symbol could be the right thing. But I
discovered that technicians and engineers could understand
the meaning of the icon without explanation, while normal
operators could not. It's a metaphor which works only if you
understand how the system works, with many processors
communicating. If you don't, it becomes meaningless.
Up to now I appear to be backing up your concept of
'familiar', but I'm not convinced that it's all the truth.
It would be nice if it were as simple as a friend of mine (a
biologist) was explaining the difference between hereditary
vs environmental factors: if the newborn resembles to the
father, it's hereditary, if it resembles to the neighbor
next door, it's environmental....
I don't believe that there is a platonic model of
computer/human interface, but I believe that there is a
model of human behavior, built by millions of years of
evolution, plus background, plus education, plus everyday
experience, plus something else, which makes some solutions
more 'intuitive' than others.
I also believe that the day a critic will be able to tell
what makes the difference between a masterpiece and rubbish,
we'll be able to pinpoint the factors not related with
familiarity.
Till then, we must rely on consistency, on trying to
understand what the user is familiar with, and on using at
best our 'intuition' to work out something 'intuitive'.
> >Coming to your question I'd say that alt-tab is connected to the idea of
> >making first of all disappear the current app and then popping up the
> >next one. But if current app has already disappeared, the scenario is
> >different, so it's not 'intuitive' to use the same command. One would
> >think that, as current app has disappeared, the next one should pop up
> >by itself.
>
> It honestly seems like, when looking suddenly at desktop, you should be
> able to hit 'ENTER' and have an app pop up! Somehow this seems more
> intuitive, even, (at least to me, based on what *I* am familiar with)
> then having the next app pop up automagically.
Well, we have our administration and production management
software which requires an F10 to confirm anything. I'm
using it since at least five years, but I still feel that
'ENTER' would be the right key, and that F10 is 'wrong'.
--
Giuliano Colla
Before activating the tongue, make sure that the brain is
connected (anonymous)
------------------------------
From: jtnews <[EMAIL PROTECTED]>
Subject: Re: Microsoft's move away from perpetual licensing proves that the
Date: Tue, 08 May 2001 11:40:33 GMT
Adam Warner wrote:
> That's just a problem with never being satisfied with a very profitable
> business that is already fully servicing the market. Microsoft's wants to
> keep growing and growing.
But the inherent problem with all software is
that you have to maintain it on an ongoing
basis as technology progresses.
As a market saturates, a company only has
two choices with a closed source product,
come up with a new version and get people
to buy the new features, or get them to pay
for "maintainence". Since the company is the
only one with the source, they are the only
ones with the ability to adapt the program
to new operating environments. They must bear
the full cost of all future maintenance
and charge for those services.
If the company's cost structure continues to bloat,
they will be increasingly pressured to increase
the price of maintainence, to the point where
the fees become more expensive than the cost to
do the R&D of the product itself! Essentially,
you're forcing the entire community to pay
for a bloated software engineering staff.
Well the commercial company could cut its staff,
some might say. But anyone who has ever worked
for a commercial software company will tell you,
software engineers will almost always continue
to add new "features" regardless of whether the
user community needs them or not, only to justify
their own existence. It's simply human nature.
Note that under the closed source model, it's possible
that the total maintenance or subscription costs to
be dramatically higher than perpetual license model
over the lifetime of the product. What makes it
even worse is that it makes it impossible to predict
how much a piece of software will cost. Companies
can arbitrarily raise subscription fees at any point
in the future and any users will be forced to pay
if their business model is dependent on that piece
of software for doing business.
For example, suppose your entire company uses Microsoft
Outlook and is dependent on it for daily operations,
and Microsoft decides to charge $40/user/year
to use Outlook. Suppose you have 100 employees,
right, there you're out $4000/year! And you MUST
pay it immediately or else what are you going to do?
Stop using email?
What if the subscription cost
suddenly rises to $100/user/year? Even if the cost
is ridiculously high, your options with closed source
software in the near term are severely limited.
You will essentially be forced to pay up until you can
change to another competitor. And as every MIS director
knows, getting users to change their habits on a dime
is nearly impossible. In the meantime,
you'll continue to pay ridiculous prices for
subscriptions.
If you want to see a software market where
the subscription model is carried to the extreme,
take a look at the electronic design automation (EDA)
market. EDA tools are software programs used in
the design of semiconductors. The subscription
costs in that market for a single piece of software
can run into the hundreds of thousands of dollars!
This is largely because the R&D requirements to support
the technology in that business are very high mostly
because the closed source model keeps the intellectual
property and ability to simply change one line in
the source code in the hands of the EDA companies.
Thus, once they've got the semiconductor companies
hooked and locked in and can charge whatever price
they want.
------------------------------
From: jtnews <[EMAIL PROTECTED]>
Subject: Re: Microsoft's move away from perpetual licensing proves that the
Date: Tue, 08 May 2001 11:41:43 GMT
Adam Warner wrote:
> That's just a problem with never being satisfied with a very profitable
> business that is already fully servicing the market. Microsoft's wants to
> keep growing and growing.
But the inherent problem with all software is
that you have to maintain it on an ongoing
basis as technology progresses.
As a market saturates, a company only has
two choices with a closed source product,
come up with a new version and get people
to buy the new features, or get them to pay
for "maintenance". Since the company is the
only one with the source, they are the only
ones with the ability to adapt the program
to new operating environments. They must bear
the full cost of all future maintenance
and charge for those services.
If the company's cost structure continues to bloat,
they will be increasingly pressured to increase
the price of maintenance, to the point where
the fees become more expensive than the cost to
do the R&D of the product itself! Essentially,
you're forcing the entire community to pay
for a bloated software engineering staff.
Well the commercial company could cut its staff,
some might say. But anyone who has ever worked
for a commercial software company will tell you,
software engineers will almost always continue
to add new "features" regardless of whether the
user community needs them or not, only to justify
their own existence. It's simply human nature.
Note that under the closed source model, it's possible
that the total maintenance or subscription costs to
be dramatically higher than perpetual license model
over the lifetime of the product. What makes it
even worse is that it makes it impossible to predict
how much a piece of software will cost. Companies
can arbitrarily raise subscription fees at any point
in the future and any users will be forced to pay
if their business model is dependent on that piece
of software for doing business.
For example, suppose your entire company uses Microsoft
Outlook and is dependent on it for daily operations,
and Microsoft decides to charge $40/user/year
to use Outlook. Suppose you have 100 employees,
right, there you're out $4000/year! And you MUST
pay it immediately or else what are you going to do?
Stop using email?
What if the subscription cost
suddenly rises to $100/user/year? Even if the cost
is ridiculously high, your options with closed source
software in the near term are severely limited.
You will essentially be forced to pay up until you can
change to another competitor. And as every MIS director
knows, getting users to change their habits on a dime
is nearly impossible. In the meantime,
you'll continue to pay ridiculous prices for
subscriptions.
If you want to see a software market where
the subscription model is carried to the extreme,
take a look at the electronic design automation (EDA)
market. EDA tools are software programs used in
the design of semiconductors. The subscription
costs in that market for a single piece of software
can run into the hundreds of thousands of dollars!
This is largely because the R&D requirements to support
the technology in that business are very high mostly
because the closed source model keeps the intellectual
property and ability to simply change one line in
the source code in the hands of the EDA companies.
Thus, once they've got the semiconductor companies
hooked and locked in and can charge whatever price
they want.
------------------------------
** 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
******************************
