Linux-Advocacy Digest #976, Volume #30 Tue, 19 Dec 00 13:13:04 EST
Contents:
Re: A Microsoft exodus! (The Ghost In The Machine)
Re: A Microsoft exodus! (The Ghost In The Machine)
Re: Windows - Is It Really Easier to Use? ("Bracy")
Re: Is Windows an operating system like Linux? (mlw)
Re: Kulkis digest, volume 2451897 (The Ghost In The Machine)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.unix.advocacy
Subject: Re: A Microsoft exodus!
Date: Tue, 19 Dec 2000 18:00:43 GMT
In comp.os.linux.advocacy, [EMAIL PROTECTED]
<[EMAIL PROTECTED]>
wrote
on Mon, 18 Dec 2000 19:41:58 GMT
<aMt%5.15552$[EMAIL PROTECTED]>:
>The Ghost In The Machine writes:
>
>>>>>>>>>>>> So, to sum up:
>>>>>>>>>>>>
>>>>>>>>>>>> [1] Nothing is intuitive.
>
>>>>>>>>>>> You need to consult a manual for everything???
>
>>>>>>>>>> "Intuitive" doesn't mean "not needing to consult a manual".
>
>>>>>>>>> You have a better definition that is simple to understand?
>
>>>>>>>> How about the one in the manual? Erm, I mean, dictionary? :-)
>
>>>>>>> Apparently the definition wasn't intuitive to you.
>
>>>>>> It wasn't supposed to be. The definition of a word -- in this
>>>>>> case, 'intuitive' -- breaks down the word, in a sense, into
>>>>>> a series of more easily digested concepts, or perhaps different
>>>>>> concepts. Occasionally, a dictionary gets into loops (I remember
>>>>>> one dictionary that defined 'nipple' as a 'teat', and a 'teat' as a
>>>>>> 'nipple'; obviously, that gets nowhere fast -- of course, that
>>>>>> particular dictionary was designed to fit into one's pocket; one
>>>>>> can only do so much :-) ). Dictionaries also have slight problems
>>>>>> with very general concepts: "go", "be", "have", "put".
>
>>>>> But the fact that you needed to consult a dictionary meant that the
>>>>> definition of "intuitive" wasn't intuitive.
>
>>>> Correct.
>
>>> Glad you agree.
>
>>>>>> I understand the definition of "intuitive", based on my earlier learning
>>>>>> and experience. It needn't be intuitive -- just understandable; that's
>>>>>> all an interface needs to be, in order to be useful.
>
>>>>> Irrelevant to the present discussion.
>
>>>>>> (The original question, IIRC, was whether a power switch was intuitive.
>
>>>>> Actually, the discussion goes back farther than that. The power switch
>>>>> example came after Aaron declared nothing about a computer is intuitive.
>>>>> That declaration came after my remark that the use of hjkl for cursor
>>>>> movement is not intuitive.
>
>>>> Both are correct. Nothing about a computer is intuitive,
>
>>> Incorrect; consider the power switch. Or did you need to consult a
>>> manual to determine how to operate it?
>
>> Have it your way;
>
>This isn't Burger King.
Feh.
>
>> a power switch is indeed intuitive in that:
>>
>> - it's a natural action to push it
>> - the effects are immediately noticeable
>> - it is prominently placed in most instances.
>
>Is that all you can think of?
- It's been in use for decades.
- They are manufacturered in the millions.
- They're cheap.
I'm not sure what you're looking for, admittedly...did you have something
specific in mind?
>
>>>> and hjkl for cursor movement is not intuitive. (Just extremely useful.)
>
>>> Not necessarily useful.
>
>> To me, it's useful. To you (and most others), the arrow keys may in fact
>> be far more intuitive.
>>
>> This is not a problem.
>
>Irrelevant, given that it wasn't described as a problem.
Obviously, it is a problem. The problem is: how to most efficiently
and/or intuitively move the cursor around in a document under
construction.
This is a problem. It's been solved multiple times, of course.
>
>>>>>> One might remark "everyone knows what a power switch is" (and most
>>>>>> do, by prior learning); hence, a power switch, being well-known,
>>>>>> is something an interface designer can leverage for new designs
>>>>>> of power switches -- and possibly other things. This makes the
>>>>>> design useful.)
>
>>>>> And perhaps even intuitive.
>
>>>> Not by the dictionary definition. Of course, YMMV.
>
>>> Whose dictionary, yours?
>
>> Yes, mine.
>
>And what dictionary do you have?
The one in my head. Not horribly useful for you, admittedly,
but it works for me. :-)
I can also reference other dictionaries (www.dictionary.com is
probably the simplest to remember). Which one is the most
relevant and/or useful depends on the user.
>
>>>>>>> How were those definitions better and simpler to understand?
>
>>>>>> Better and simpler to understand than what?
>
>>>>> Than the one I gave.
>
>>>>>> Pressing a button?
>
>>>>> Forget the definition I suggested already?
>
>>>> Yes. It has been "scrolled off" this news post and my newsreader (SLRN)
>>>> can't view messages by reference ID, unlike Netscape.
>
>>> No, it hasn't scrolled off, given that it has been retained in your
>>> follow-up.
>
>> Oh, OK. I guess that's my official definition of "intuitive", then.
>> I hadn't realized that.
>
>Non sequitur, given that we were talking about *my* "definition", which
>you had erroneously assumed to have "scrolled off".
Where is it in the above post, then? Or you can give me a
message-ID for scanning purposes. I think I snipped it.
[snip]
--
[EMAIL PROTECTED] -- insert random misquote here
up 85 days, 2:28, running Linux.
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.unix.advocacy
Subject: Re: A Microsoft exodus!
Date: Tue, 19 Dec 2000 18:02:18 GMT
References: cut back.
In comp.os.linux.advocacy, [EMAIL PROTECTED]
<[EMAIL PROTECTED]>
wrote
on Mon, 18 Dec 2000 19:45:24 GMT
<oPt%5.15553$[EMAIL PROTECTED]>:
>The Ghost In The Machine writes:
>
>>>>>>>>> Aaron R. Kulkis writes:
>
>>>>>>>> [snip]
>
>>>>>>>>>> Tholen...
>>>>>>>>>> when you finally realize how utterly worthless your life is...
>>>>>>>>>> remember to slit lengthwise.
>
>>>>>>>>> Kulkis, when you finally realize how utterly worthless your invective
>>>>>>>>> is, remember to come back here and apologize.
>
>>>>>>>> Is it me, or is there some sort of repeating pattern here? :-)
>
>>>>>>> I gather from your emoticon that you already know the answer. So why
>>>>>>> bother to ask the question?
>
>>>>>> So I have a sense of humor. :-P
>
>>>>> And you need to broadcast that information?
>
>>>> Better than broadcasting that one does *not* have a sense of humor,
>>>> methinks. :-)
>
>>> You make it sound like those are the only two options.
>
>> And a third option would be ... ?
>
>Leaving the question open.
True.
>
>> I'm curious.
>
>Do people who go to church on Sunday always "broadcast" whether they
>do or do not have a sense of humor in that forum?
Ah, I see. So we're supposed to do what in these forums (fora?), again?
--
[EMAIL PROTECTED] -- insert random misquote here
up 85 days, 2:33, running Linux.
------------------------------
From: "Bracy" <[EMAIL PROTECTED]>
Subject: Re: Windows - Is It Really Easier to Use?
Date: Tue, 19 Dec 2000 17:49:16 GMT
In article <koM%5.11119$[EMAIL PROTECTED]>, "Erik Funkenbusch"
<[EMAIL PROTECTED]> wrote:
> No, you conveniently forget that the card is claimed to be supported,
> but isn't.
Oh really? You called the maker of your video card and they claimed that
they provided support for their card under Linux?
It's also claimed that a Seagate Archive Python tape backup unit is
supported under Windows NT. The unit will work fine for a little while,
then will stop working altogether with a message "Tape Device Not
Responding." At the company I worked for, it was the single most
commonly reported problem. Seagate came out with a new driver for it
about every 2-to-4 weeks. By the time I left, there were at least 6 new
versions of the driver.
Was this Microsoft's fault? Or was it the fault of the tape drive
manufacturer?
Bracy
------------------------------
From: mlw <[EMAIL PROTECTED]>
Subject: Re: Is Windows an operating system like Linux?
Date: Tue, 19 Dec 2000 13:10:05 -0500
Erik Funkenbusch wrote:
>
> "mlw" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Erik Funkenbusch wrote:
> > >
> > > "mlw" <[EMAIL PROTECTED]> wrote in message
> > > news:[EMAIL PROTECTED]...
> > > > Erik Funkenbusch wrote:
> > > > >
> > > > > "mlw" <[EMAIL PROTECTED]> wrote in message
> > > > > news:[EMAIL PROTECTED]...
> > > > > > Erik Funkenbusch wrote:
> > > > > > > > A small embedded ROM system can be an OS. The X window system,
> > > though
> > > > > > > > vastly more complicated, and in some ways similar to an OS, is
> not
> > > an
> > > > > > > > OS. The same goes for MS Windows.
> > > > > > >
> > > > > > > X Windows does not offer scheduling, memory management, file
> > > systems,
> > > > > etc...
> > > > > >
> > > > > > Not relevant.
> > > > >
> > > > > It's completely relevant. An OS provides these things. If it
> doesn't,
> > > it's
> > > > > not an OS.
> > > >
> > > > DOS does not do scheduling, but it is an OS, X does have an amount of
> > > > memory management. And many embedded operating systems do not have
> file
> > > > systems. Thus not relevant.
> > >
> > > DOS does do scheduling, but it schedules things like interrupts.
> >
> > DOS does not "schedule" interrupts. It provides an API for setting an
> > interrupt address. Nothing stands between an interrupt handler and the
> > CPU under DOS. A DOS extender like QEMM, EMM386, and Windows 9x has to
> > intercept the interrupt because they are in protected mode and must
> > reflect the interrupt to the correct VM.
>
> Interrupts are prioritized, this is a crude form of scheduling.
The prioritization is done in hardware by the interrupt controller. DOS
has nothing to do with it. An address is put in the real mode interupt
vector table, this you'll need an x86 programming manual to understand,
and the processor jumps to a location when an interupt happens. DOS has
no code that changes this.
>
> > > File
> > > systems are an example of I/O, which is what an OS controls.
> > Yes, but not all operating systems have a file system, so it is not a
> > prerequisite.
>
> It was an example. If a system provides a file system, then the OS provides
> it.
So, it is not relevant.
>
> > > Typically an
> > > OS does things which a program cannot (or does not) itself do, such as
> > > arbitrating time-slices, knowing how to talk to I/O devices, etc...
> >
> > DOS does not arbitrate time slices, bit it is an OS. DOS does not even
> > talk to the disk drive, it uses the BIOS.
>
> DOS isn't much of an OS.
But, it is an OS. Magnitude is not relevant. A bacteria is a living
animal.
>
> > > Technically, an OS is just another program. As such, your arguments
> about
> > > what is and isn't an OS is largely meaningless, since to a Java program,
> the
> > > JVM is the OS, yet to the JVM, whatever API it uses is the OS, which
> itself
> > > may use an API for what it thinks is the OS.
> >
> > This is exactly my point. The Java VM is middleware, just like Windows.
>
> And just like DOS, and just like the BIOS.
>
> > > > > Oh, that explains why all those Linux-only systems out there run
> Lilo
> > > then.
> > > >
> > > > No, LILO is a convenience. As stated. While it is not absolutely
> > > > necessary, it saves the hassle of running fdisk to change the OS in
> > > > which you boot.
> > >
> > > Again, that explains why all those Linux-only system use Lilo. The real
> > > reason is that if they used your method, they could not construct a
> > > partition table.
> >
> > What are you talking about? A boot loader is a very convenient thing.
> > What does this have to do with constructing a partition table? One would
> > use fdisk to change bootable partitions.
>
> If you copied the vmlinuz file over the MBR to make it self-bootable, you
> would destroy the partitioning information making it impossible to create
> partitions.
Why are you harping on this? You copy vmlinux to the base of a partition
and mark that partition as bootable. You can even net-boot Linux which
would not even use a boot sector.
>
> > > > > > You seem to be confusing Windows 9x with NT. I refer you to Andrew
> > > > > > Schulman's book "Unauthorized Windows 95, Developers Resource Kit"
> or
> > > > > > "Inside Windows 95" by Adrian King. (MS Press). Or you could
> download
> > > > > > the Windows DDK and read the help files.
> > > > >
> > > > > No, I'm not. I've read Unauthorized Windows 95 at least a dozen
> times
> > > and
> > > > > know it inside and out.
> > > > >
> > > > > For instance, page 146 states what DPMI is:
> > > > >
> > > > > "DPMI stands for DOS Protected Mode Interface [...] Despite the
> name,
> > > DPMI
> > > > > isn't the same thing as a DOS extender. Instead, DPMI is a set of
> INT
> > > 2Fh
> > > > > and INT 31h services you can use to write a DOS extender thats
> > > compatible
> > > > > with Windows or 386 memory managers".
> > > > >
> > > > > In other words, Windows is itself a memory manager, and provides a
> DPMI
> > > > > interface to DOS applications and DOS extenders. Windows itself is
> not
> > > a
> > > > > DOS extender.
> > > >
> > > > I will not argue with Schulman's factual observations, because they
> are
> > > > largely accurate. I disagree with his conclusions, which are
> > > > questionable. DPMI is the interface by which Windows manages page
> > > > manipulation and memory management as well as other things.
> > > >
> > > > The Windows 386 executive is a DPMI provider. It always has been. The
> > > > Windows system VM runs in a DPMI virtual machine. This is verifiable.
> > >
> > > Given that the Windows system VM itself is what controls the VM's, you
> seem
> > > to have created a circular redundancy. If the VMM controls all other
> VM's,
> > > then how can it, itself live in a DPMI VM? The part you are missing out
> on
> > > is that Windows requires a DPMI server to create the VMM, but then it
> > > replaces it. This is sumarized on page 171 by a chart which shows the
> > > version numbers returned before and after Win386 is loaded.
> >
> > Lets go through this one last time. You will have to look at the Windows
> > DDK, right now you are arguing from ignorance.
>
> I'm using the exact same source you relied on to prove your point. So now,
> because I point out the flaw in your argument, your source is no longer
> valid? Further, you are now shifting the burden of proof from yourself to
> me, which indicates a failing argument.
You saw no flaw, you misunderstood what was written.
>
> > The "Windows 386 executive" is a DOS extender. It creates a system VM in
> > which the original DOS is run. The Windows environment is also run
> > within the system VM. All Windows applications are run in the system VM
> > as well. Virtual DOS boxes are run, separate from the system VM, but
> > they are not "Windows" programs.
>
> Again, Schulman (who you used to prove your argument) states specifically in
> many places that Windows runs the system copy of DOS in a V86 session under
> Windows control. You still have a circular argument.
There is no circular argument at all. My paragraph is accurate, look at
the Windows DDK.
>
> > The Windows 386 executive (VMM) does not schedule 32 bit Windows
> > threads, that is done by the Windows environment within its VM. The
> > Windows 386 executive (VMM) only schedules VMs.
>
> Your claim is that the system VM lives in a V86 mode VM (or more accurately,
> that the System DOS VM is the same VM that the system VM uses). You seem to
> indicate here that there are two scehdulers (well, three if you count the 16
> bit co-operative one). Further, your own statements here would indicate
> that the VMM is in fact an OS, since it's in charge of controlling the VM's.
You are confusing "The Windows 386 executive" with the "Windows system"
these are two different pieces. You should read the DDK documentation.
You would understand more.
>
> > The API which the "Windows system" uses to manage memory and resources
> > within its VM is largely DPMI.
>
> Oh, so now it's "largely" DPMI, rather than your previous statement that it
> was completely DPMI.
There are APIs which are provided by VxDs which can be loaded by The
Windows 386 Executive. This does not an OS make and it does not change
the basic assumptions.
>
> > All of the above is factually accurate. If you disagree, that's fine,
> > but I will ignore any argument unless you give me a page, or number of
> > pages, from the Windows DDK that refute this.
>
> Strange that you provide no such cites to back up your claims. The only
> cites you've mented are King and Schulman, and i've proven that what you
> claim (at least Schulman) says is not in fact what he says.
You have yet do do what you claim to have done.
>
> > > In order for a "DPMI virtual machine" to exist, there must be code
> running
> > > as the DPMI server. The DPMI server cannot itself be running in a DPMI
> > > virtual machine.
> >
> > The Windows 386 executive is separate from the system VM. The Windows
> > 386 executive is a DPMI DOS extender. AFAIK still version 0.9 as well.
>
> Ahh.. so now we get to the meat. Now you are claiming that Windows itself
> is not in fact a DOS extender, but that the Windows 386 executive is.
Yes the Windows 386 Executive is analogous to QEMM, and the Windows
system is analogous DesqView.
>
> > > No, you said "Take a look low down dude. The Windows executive is a DPMI
> > > environment and Windows run in a virtual DOS machine within it. BTW:
> Windows
> > > runs in the same virtual machine as the actual DOS OS because each
> Windows
> > > program has to have a DOS PSP."
> > >
> > > This is verifiably wrong. Yes, it's true that Windows programs have to
> have
> > > a DOS PSP, but that isn't related to running in the same VM as the DOS
> VM.
> > > Those are merely data structures which are mapped into the same virtual
> > > memory.
> >
> > DOS PSP? Perhaps not directly related, but very important.
>
> So in other words, your use in the sentance was a red herring.
No, not as such.
>
> > Also, I said, unambiguously, that each windows program ran in the same
> > VM and that memory protection was managed through page table
> > manipulation. This is a fact, you disagreed, you were wrong, so admit
> > it.
>
> I already stated I was wrong when I used the wrong term.
No, you were wrong in your understanding and want to back peddle by
saying you used the wrong term.
>
> > > NT runs 16 bit Windows in a virtual DOS machine, it's a program called
> WOW,
> > > or Windows on Windows.
> >
> > There is usually one wow session started on NT startup. This can be
> > controlled through a registry setting to save memory, if you are not
> > going to run 16 bit windows programs. When you click on a 16 bit windows
> > program, you can bring up a run dialog that has a checkbox which asks
> > "Run in separate memory space" which will run this 16 bit program in ins
> > own VM and its own copy of wow.
>
> This has nothing to do with my statements which you deleted, which is that
> the Win32 applications were all running in the same VM under NT, along with
> the Win32 subsystem.
Are you saying that all Win32 programs run in the same VM under NT?
That's wrong. All Win32 programs have their own VM under NT. All 16 bit
programs run under a single windows on windows environment, unless they
are run in their own memory space, or loaded into another 32 bit windows
programs VM.
>
> > A 32 bit program can run a 16 bit program in its memory space. Look at
> > TWAIN drivers, which are often 16 bit programs with 32 bit thunk layers.
> > This ability to run 16 bit programs within the VM of a 32 bit app is
> > very important. This bit is documented in the NT DDK or SDK, so you'll
> > have to read that too.
>
> I don't disagree with this. I never said I did.
>
> > Again, I will ignore any argument unless you can point me to a DDK page
> > that refutes this, because it is factually accurate.
>
> Prove your own statements first.
There is no need. They are documented and proven in the DDK. Should you
not care to research, that's your problem.
>
> > > > > Clearly from Schulman's picture, DOS is *NOT* in the system VM nor
> does
> > > it
> > > > > detail how Windows depends on DPMI, only that it provides extensive
> DPMI
> > > > > services.
> > > >
> > > > If you spent the time to read the text surrounding the picture, it
> makes
> > > > big mention that Microsoft has left DOS out of this picture.
> > >
> > > Yes, but it does *NOT* say that DOS runs in the same VM as the System
> VM.
> > > In fact, other parts of the book specifically say that it runs it's DOS
> > > dependant code in a *SEPERATE* VM under Windows control. Furthermore,
> DOS
> > > needs to run in V86 VM, while the System VM is *NOT* running in a V86
> VM.
> >
> > It neither supports nor denies my assertions. I leave you to read the
> > DDK documentation and do some experiments. Specifically, find out how to
> > make DOS calls from within a VxD, find out how to signal a Windows
> > program from an interrupt. When you understand how to do both of these
> > things, and what the prerequisites are, you will not be arguing.
>
> You claimed it did support your assertion, then when proven wrong you claim
> that Schulman isn't a valid source afterall.
It proved the assertion that all Windows programs, in Windows not NT,
run in the system VM which you claimed to be false.
>
> > > > The paragraph starts with:
> > > >
> > > > "In the meantime, there's one very noticeable feature of this diagram:
> > > > MS-DOS is missing" and goes on to explain, that DOS is present only
> left
> > > > out of the diagram for marketing purposes.
> > > >
> > > > That whole chapter explains and verifies everything I written here,
> and
> > > > the only difference between Schulman's position and mine, is that I
> > > > don't believe Windows is an OS.
> > >
> > > There also seem to be factual differences between your position and
> > > Schulmans, as illustrated above.
> >
> > Also remember that Schulman has simplified some things for publishing.
> > This is sort of related to my biggest problem with what he writes. As a
> > software developer, I shudder at people using his books to figure out
> > how to write programs. He likes undocumented API sets, and (I don't know
> > if he still does) advocates using them. This seems a bit reckless to me.
> > If it weren't for both Petzold and Schulman, Windows program quality
> > would be better. Unfortunately, in Windows, only Microsoft application
> > writers get documentation to these undocumented APIs, thus a developer
> > must rely on these sorts of books to figure out how to do some of the
> > most basic things to compete with Microsoft.
>
> Well, then you shouldn't have used schulman to validate your story. You
> should have instead used cites from the DDK to back up your point.
I used Schulman to prove the point about the windows programs running in
the system VM.
>
> > I use Linux, true. But I have also done extensive work on Windows and NT
> > in both application space and kernel space. I have done real-time
> > systems in both. Many of the things you are trying to argue are not
> > opinions, but are how Windows is constructed, and arguing, or trying to
> > disprove through debate, what can easily be verified by reading the DDKs
> > is a waste of time for both of us.
>
> If it's so easy, then why haven't you provided cites?
"cites?" why, read the DDK documentation, it is all clear.
>
> Besides, I too am involved in real-time systems in NT for the last 4 years,
> and have done so in 9x in the past. Your experience doesn't mean anything.
If this is true, and I have no reason to believe it is not, let me know
what software you have worked on, I will avoid it because you don't seem
to understand the platforms on which you are developing.
--
http://www.mohawksoft.com
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.unix.advocacy
Subject: Re: Kulkis digest, volume 2451897
Date: Tue, 19 Dec 2000 18:08:34 GMT
Followups to the one newsgroup I do read. :-)
In comp.os.linux.advocacy, John Jensen
<[EMAIL PROTECTED]>
wrote
on 19 Dec 2000 03:09:07 GMT
<91mjgj$gb1$[EMAIL PROTECTED]>:
>Aaron R. Kulkis <[EMAIL PROTECTED]> writes:
>: John Jensen wrote:
>
>: > : > Unoffical Scoring:
>: > : >
>: > : > Tholen: 3 /* low on content, but not actually harmful */
>: > : > Kulkis: 1 /* low on content, and increasingly nasty */
>: >
>: > : I don't suffer fools like tholen, or you, lightly.
>: >
>: > That might be your problem, right there ;-)
>: >
>: > John
>
>: Not my problem. The only person complaining is you.
>: Why is that?
>
>John Travis said "Maybe true" and Marty gave a similar score. I think you
>count posts like you count signature lines: one, ... one, ... ah one ..
I for one might have mentioned it once -- and left it at that.
If one is from a location that requires pay-per-minute access on
a low-bandwidth Internet connection, then one might have a beef.
As it is, I've seen worse, usually in the alt.ascii-art or somesuch,
but is it really that big of a deal?
The idea is to argue the relative merits of (pick one)
- Linux
- NT
- Windows
- Macintosh
- OS/2
- Unix
(although it's not clear how one can simultaneously advocate all six :-) )
>
>John
>--
>with a sane signature
--
[EMAIL PROTECTED] -- with a non-sane signature (sanity or insanity requires
sentience, signature virii notwitstanding) :-0
up 85 days, 2:37, running Linux.
------------------------------
** 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
******************************
