Linux-Advocacy Digest #662, Volume #31 Mon, 22 Jan 01 21:13:04 EST
Contents:
Re: NT is Most Vulnerable Server Software ([EMAIL PROTECTED])
Re: Games? Who cares about games? (Greg Yantz)
Re: Why "uptime" is important. (T. Max Devlin)
Re: Why "uptime" is important. (Charlie Ebert)
Re: Why "uptime" is important. (T. Max Devlin)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.destroy.microsoft
Subject: Re: NT is Most Vulnerable Server Software
Date: Tue, 23 Jan 2001 02:18:19 +0100
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>,
T. Max Devlin <[EMAIL PROTECTED]> wrote:
>
> Said [EMAIL PROTECTED] in alt.destroy.microsoft on Mon, 22 Jan
>>In article <[EMAIL PROTECTED]>,
>> T. Max Devlin <[EMAIL PROTECTED]> wrote:
>>>
>>> Said [EMAIL PROTECTED] in alt.destroy.microsoft on Mon, 22 Jan
>>>>In article <[EMAIL PROTECTED]>,
>>>> T. Max Devlin <[EMAIL PROTECTED]> wrote:
>>>>>
>>>>> Said [EMAIL PROTECTED] in alt.destroy.microsoft on Sun, 21 Jan
>>>>>>In article <[EMAIL PROTECTED]>,
>>>>>> T. Max Devlin <[EMAIL PROTECTED]> wrote:
>>>>>
>>>>> This is a convention, not a rule of routing or the IP protocol.
>>>>
>>>>It is not a convention. Look up the appropriate RFC's. The private IP
>>>>address ranges should never appear on the Internet.
>>>
>>> That's what we call a convention. Notice the "should". Where as,
>>> 0.0.0.0, and 127.0.0.1 *cannot* appear on the Internet. Get it?
>>
>>I 'should' have said must. Now tell me why '0.0.0.0, and 127.0.0.1
>>*cannot* appear on the Internet'? It just takes a misconfigured router.
>>There is nothing magical about these addresses. You really are clueless.
I concede this point as I only recall seeing these addresses on local
subnets (nmap can generate packets with these addresses).
>>>>Yes NAT is independant of how a firewall implements a security policy.
>>>>However most if not all firewalls implement NAT as it is the logical
>>>>place to do so.
>>>
>>> I think you mean "this is an available feature in most firewall
>>> products". Whether a firewall implements NAT is not as trivial a matter
>>> to guess. But you can generally tell those enterprises which attempt to
>>> use NAT to 'secure their network', as they are the ones with slow
>>> performance and intermittent connectivity.
>>
>>Again you just don't know what you are talking about. NAT is essential
>>if you are using private addresses on your Intranet.
>
> Boy, you are a fucking idiot, aren't you? NAT is entirely unnecessary
> if you are using private addresses on your Intranet. The only reason
> you would have to translate an address is if you are going to be using
> it on the Internet.
>
> (Think hard, and don't make assumptions about what you just read.)
Christ, I've been talking about firewalls. Any company that uses private
addresses must use NAT if they want Internet access. I thought this was
obvious but apparently not.
>>For sure NAT is an
>>overhead and you scale your firewall performance appropriately.
>
> You're just not getting it, are you? Whether a firewall and a NAT are
> entirely separate functions SAYS NOTHING ABOUT WHAT BOX THEY'RE RUNNING
> ON. Now, I can understand if, in your low-level technician-type
> understanding of networking, you believe that what box something is
> running on is the only thing of importance in the whole world. But that
> doesn't mean you should be wasting my time with your ignorance.
>
> Please, try to control yourself. Just because you don't understand what
> I'm saying doesn't mean that I don't. How many times do you think I'll
> have to repeat that before you read and comprehend it. I can understand
> your wanting to beat your chest a bit, and even to "put me in my place",
> as I'm well aware I'm not the least annoying personality on Usenet.
> Still, if you didn't have your head shoved so far up your ass you can
> whistle while you fart, you'd remember that I've already pointed out
> that I am not, despite your impression or your contentions,
> unknowledgable in the least degree, regardless of any minuscule factual
> discrepancies you may think you detect in my descriptions.
It is no wonder I don't know what you are trying to say most of the time
as you are talking crap.
> [..]
>>Max it is quite clear you do not know what you are talking about.
>>At what level in the OSI stack would you place ethernet, IP, TCP, ftp
>>for example?
>
> Well, Ethernet is part of layer (the OSI "stack", by which I presume you
> mean the OSI 7-layer Reference Model, does not have 'levels') 1
> (10Base-T, et, al.) and part of layer 2, the Data Link layer (CSMA/CD).
> IP is of course the Network (routing) layer, layer 3, and TCP shares
> layer 4 with UDP (and putatively RDP). FTP is not so conducive to
> correlation, since software is not generally broken down between layers,
> particularly Internet-type software. The FTP protocol itself provides,
> as all "servers" do, a session, presentation, and application layer.
> (You're sure to disagree with this last part, owing to the universal
> confusion caused by misappropriation of the *telecommunications network*
> model developed by the OSI for the *datacommunications* industry.)
You'r just being pedantic again. Whether I meant layer or level it
was quite obvious what I meant. I'm not going to disagree about your
defintion of layers. There isn't a 1 to 1 relation between TCP/IP or
any other protocol built on IP. You said the P in PAT was proxy which
is just wrong. The P Max is for Port translation in TCP (and UDP on
stateful firewalls such as fw-1 and sunscreen).
>
>>I'm sure you cannot answer this. Proxies work at the
>>application layer (7) to give you a little help.
>
> That would depend on the proxy; many proxies work on the session layer,
> as a matter of fact, though I'm quite sure that you naively and
> routinely use the term "application layer" and that all the others using
> the term naively and routinely (because none of you, no matter how
> wondrous and grand your capabilities or skills, gives a rip about the
> fact that this usage contradicts the usage of other equally capable
> people in other specialties).
>
> I've been through this discussion with no less than a dozen people on
> Usenet, and literally hundreds, potentially thousands, of people
> throughout the industry. The only two results that have ever occurred
> is a) they give up, and consider me wrong or b) they recognize I am
> correct.
>
> Which are you going to go for?
Well I know from other threads frequented by you, pre-emptive multitasking
comes to mind, that I will never win in your mind. Still one must try. :-)
>>>>PAT is port address translation often also called
>>>>hide address translation. When you do many to one NAT you have to have
>>>>some method of knowing which packet belongs to which connection. Doing
>>>>port translation allows this. Nothing to do with proxies which work at
>>>>the application layer.
>>>
>>> Don't move! You've just stepped on a land mine. Carefully, without
>>> bending over, pretend you never said "the application layer".
>>
>>This is just too stupid to believe. Your credibility has just dropped to
>>zero.
>
> Perhaps with you; your credibility is zero as well, so that makes no
> never-mind, to me.
>
>>For example, squid is a proxy. Pray tell me what layer in the OSI
>>stack does squid function?
>
> It doesn't matter. Squid is a piece of software. Software just does
> what it does, it doesn't "function at any layer" other than whatever
> layer YOU are abstracting it as. Of course, since you don't really have
> any definition of the abstractions which each layer encompasses, you
> have no method whatsoever for consistently, accurately, or practically
> *applying* the knowledge of "on which layer does squid function".
>
> My model, in contrast, provides these things. But the first step to
> understanding it is to get rid of this ingenuous and disfunctional
> notion that "the OSI stack" is a "software architecture". It doesn't
> work like that, because the technology which it was originally designed
> to be an architecture *for* is now rather trivial and even outdated.
Christ (lucky I'm not religeous) squid is an application. It runs at
the application layer. Stop trying to squirm out of it. You were wrong.
Have the courtesy to admit it.
> Pop Quiz: What software architecture was the OSI reference model
> created for? (Hint: "the OSI architecture" is not a valid answer.)
OSI is a standard and was not created for any software architecture. If
you want to know which language was designed to implement the OSI stack
for realtime OS's then it would be CHILL. The CCITT tried to ensure
that all telecoms software was developed in CHILL (much like the US DoD
wanted ADA to be used for all defense software). I have developed software
in CHILL and know what an abortion of a committe designed language it is.
> [...]
>>Listen I administer several checkpoint fw-1's, Sun's sunscreens and
>>a couple of tuxscreens (don't know what that is do you?).
>
> No, but I've explained how they work to those that do. Does that count?
>
>>I have
>>administered firewalls for 4 years and been a Unix system administrator
>>for over 12 years. Be so kind to tell us what experience you have in
>>this area. Do you understand what NAT is? From what you have said I
>>don't think so.
>
> Well, I've been teaching networking and network management all around
> the US (and a couple brief trips overseas) for about ten years, now.
> Before that I was involved in PC application training. I've done
> consulting, seminars, and professional courses for about two dozen of
> the largest carriers, service providers, and enterprises in the last few
> years, including MCI/WorldCom, Sprint, GlobalOne, Mirror Image, E*Trade,
> and several pharmaceutical companies.
Those who can do, those who can't teach. It is clear where you fit in. I
feel sorry for your students.
> I've been working with firewalls since before there were firewalls. And
> unlike practically everyone else in the industry, I never had to learn
> the difference between firewalls, screening routers, and NAT, because I
> was never under the impression they were the same thing.
Your ignorance just amazes me. It is clear you have no real world experience.
I never said these things were the same thing. I said that most if not all
firewalls have NAT functionality. Of course you can do NAT on your external
router if it has that functionality. It's a choice. I choose to do NAT on
interior firewalls. I can choose when I need NAT or not on the firewall. One
firewall I administer has 8 subnets hanging off it. Two are the Intranet
and DMZ (which goes to the outer firewall). The others are what we call
content zones where the servers live. Why so many subnets? This is a large
company with many remote branches. Many of the servers are connected via
leased lines.
>>I just don't believe a word of this. You are sounding more and more
>>like the wintrolls on this list.
>
> Let that be a lesson to you.
>
>>It is clear you have little or no
>>knowledge / experience with regard to IP networking and firewalls in
>>particular.
>
> Yes, I'm sure. Nobody engages in chest-beating like Unix admins talking
> about networking, that's one piece of knowledge and experience which
> seems accurate enough.
I know you will never admit to being wrong. It is probably genetic and
nothing I can say will convince you.
------------------------------
Subject: Re: Games? Who cares about games?
From: Greg Yantz <[EMAIL PROTECTED]>
Date: 22 Jan 2001 21:00:44 -0500
[EMAIL PROTECTED] () writes:
> On 22 Jan 2001 19:13:37 -0500, Greg Yantz <[EMAIL PROTECTED]> wrote:
> >[EMAIL PROTECTED] () writes:
> >> On 22 Jan 2001 16:32:28 -0500, Greg Yantz <[EMAIL PROTECTED]> wrote:
> >> >[EMAIL PROTECTED] () writes:
> >> >> On 22 Jan 2001 08:55:27 GMT, Perry Pip <[EMAIL PROTECTED]> wrote:
> >> >> >On Mon, 22 Jan 2001 00:41:30 -0500,
> >> >> >mlw <[EMAIL PROTECTED]> wrote:
> >> >> >>I have a Nintendo for games, why would I waste a computer on games?
> >Because Nintendo games are crap that my 8 year-old nephew finds
> >intellectually demeaning?
> This still isn't a technological limitation. The progenitors
> of most of today's PC game genre's had their origins on
> console systems.
So?
> Besides, even Nintendos aren't necessarily limited to super
> mario world and mario karts. There's significant overlap
> between current consoles and PC's actually.
True but irrelevent.
> What you're more
> likely to find only on PCs are the really mindless and or
> violent fare.
Speak for yourself. Find a good simulation to rival Civ II, or an
up-to-date non-idiotic RPG or (I hate to have to do this to you,
but you asked for it) anything TEXT-BASED for a console.
> >> >> >By any chance would any of your computers be sitting idle while you
> >> >> >play games on your Nintendo? If so, why waste your money on the
> >> >> >Nintendo??
> >Indeed, if you're going to have a computer anyway why not use it for games
> >as well and not waste money on a console? Besides, PC's have better
> >games. :P
> A PC can be quite suitable for non-gaming consumer PC use
> and be grossly inadequate for gaming, or simply inadequate
> enough to be annoying.
Your oversimplified, one-size-fits-all approach to this issue is
really quite annoying. UT or Q3 are not the only games in the
world you know...
> You have to know well enough to get
> good PC gaming components and then pay for them. You then
> may or may not have to deal with hardware an software
> integration issues that stem from PC's being random
> collections of spare parts.
Possibly. If you want to play shooters you need to pick your
components carefully, true. But you have to deal with the integration
issues whether or not you play games.
> >> >> When's the last time you saw a computer capable of playing
> >> >> current games onsale for $100, or even $300?
> >> >For me the question is "when was the last time you saw a computer
> >> >capable of playing the kind of games you like to play onsale for
> >> >$100, or even $300?"
> >> Are we talking current games or just bargain bin stuff?
> >Yes.
> >> Even a relatively simple 2D RTS can bring older PC's to
> >> their knees. If it's not the rendering, then it's the AI
> >> & if it's not that then it's pushing the game itself too
> >> hard on upper levels.
> >That was kind of my point. The cost of the platform (and your
> >ongoing argument about the general merits of PC vs. game console)
> >matters less to me than "will it play the games *I* *want*?"
> OTOH, you're still left with a $2000 game console.
You might be, but I can assure you I am not.
> >> >Some people play games on their PC because the games they enjoy
> >> >tend to be PC games. Consoles don't do IP networking very well...
> >> Consoles do IP networking just fine.
> >Really? Which ones?
> Amiga & ST.
I hate to break the news to you, but those were PC's...
> Then there's the Dreamcast and PSX2.
I somehow doubt it.
> >> Infact, several older console based micros were doing
> >> multuser gaming long before it became trendy.
> >This doesn't parse, unless you're redefining "console" on the
> >fly. In this context, "console" equates to "game console",
> >i.e. Nintendo, NOT "system console".
> No, it's simply consistent with the definitions of the
> day and even now. An Amiga is little more than a Dreamcast
> with an internal floppy drive, mouse and keyboard.
Again I hate to have to break the news to you at this late date,
but an Amiga is a PC. It's a "real computer".
> They both represent a single static specification of hardware,
> one that can be used to more effectively program despite
> technological limitations.
That's lovely. How is it relevent?
> >The first real-time, multi-player, networked game I played
> >(and IMO still one of the very best) was text-based. So?
> >> Consoles merely haven't BOTHERED to support networking
> >> for the most part.
> >Therefore I don't own a game console. At a bare minimum, if I
> >can't play netrek or CS on it, it's not a viable gaming platform.
> >Even better if I can run an empire client on it, or even a
> >server. Run any Nintendo empire servers lately?
> No you are merely spouting gibberish.
Now your ignorance is showing.
> Whether or not a platform can run game servers is entirely
> irrelevant to the issue what makes a good game client
> platform.
And completely relevent to the issue of using your PC for "gaming"
in ways that a console could never duplicate.... short of growing up
and becoming a <wait for it> PC.
> In all likelihood, a good game server machine
> could get away with possessing characterstics that make it
> a VERY poor "game machine".
That really depends on how you define "game machine". From your
assumptions I have to conclude that I play rather a wider selection
of games than you have even *seen*, much less played.
> >> >> [deletia]
> >> >> The "money pit" is rather more likely the PC.
> >> >Why are you arguing over a matter of personal preference?
> >> Spending $2000 or more for a grossly inefficient games console
> >> should be considered in it's full context.
> >Which is?
> >Your basic assumptions are really quite remarkable. Why don't you
> >go argue with someone who spends $2000+ on a desktop PC and then
> >uses it only to play games? (When you find one, I'm sure you'll
> >get the best of him.)
> What else do YOU use your PC for that justifies paying the
> extra money for reasonable gaming components?
Pretty much everything. I'm a sysadmin.
> You can build a reasonable low profile non-game PC for
> about as much as a new games console would cost. Can
> you say the same for a PC game machine?
That rather depends on how you define "game machine". I can take
a $50-100 piece of PC hardware obsolescence, put Linux on it
and have a very fine time playing games a console could never
duplicate... unless it grew up and became a PC.
Your rather narrow definitions of what comprises "gaming" and a
"game platform" and your tendency to push one-size-fits-all
"gaming" solutions to the detriment of individual choice is really
bitterly ironic from someone with your advocacy position.
-Greg
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Subject: Re: Why "uptime" is important.
Reply-To: [EMAIL PROTECTED]
Date: Tue, 23 Jan 2001 02:02:37 GMT
Said Bob Hauck in comp.os.linux.advocacy on Mon, 22 Jan 2001 17:17:33
GMT;
>On Sun, 21 Jan 2001 12:56:04 -0500, mlw <[EMAIL PROTECTED]> wrote:
>>Bob Hauck wrote:
>
>>> It is difficult to be productive while waving vodoo dolls about in the
>>> hope that your hard work won't get flushed down the toilet on the whim
>>> of your computer.
>>
>>While this is hysterical, it is also very true. I've never seen it
>>described before, but I have seen this sort of thing often.
>
>I did go a bit overboard in the interests of exposition, but I think it
>is a valid point. Windows users don't trust their computers. They
>expect them to fail at random. This is just not acceptable for a
>professional tool. No other business tool suffers from this kind of
>problem.
I disagree. Particularly since they've been so well trained by the
Microsoft monopoly in the office environment, many people are quite
wiling to mistrust photocopy machines, telephone systems, and just about
any other device.
>And it _does_ reduce the productivity gains one would expect
>from automation.
I'm quite sure we're already into negative values.
>If you have a reliable computer, your whole attitude toward the computer
>changes. You are more willing to use it for important tasks, and you
>are more willing to invest time in learning the ins and outs of it.
The one thing that most appeals to me about Linux is the knowledge that
the configuration will be entirely and completely persistent. Years,
even decades from now, I'll still be able to just copy a few text files,
and my new computer becomes my old computer.
Unlike Microsoft monopoly crapware, where I have to keep re-creating my
desktop, and fighting a losing battle to get it to remember anything
through a reboot, let alone enhancing functionality.
>>My mom, runs turbo tax, then reboots her computer "to be safe" before
>>running word, and vice versa. She says if she doesn't it crashes.
>
>That's just the sort of thing I was referring to. It is far less
>prevalent among NT users of course, which by itself is almost
>justification for spending the extra money on NT and appropriate
>hardware if you really need Windows apps.
I'd sooner lose a testicle than have to use Win98, so it was either NT
or unix, and, as you said, I really need Windows apps. (Not for any
functionality that Windows or Win32 itself can singularly provide, but
just because of the illegally built and maintained application barrier.)
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: [EMAIL PROTECTED] (Charlie Ebert)
Subject: Re: Why "uptime" is important.
Reply-To: Charlie Ebert:<[EMAIL PROTECTED]>
Date: Tue, 23 Jan 2001 02:05:43 GMT
In article <50Pa6.3603$[EMAIL PROTECTED]>,
Lloyd Llewellyn wrote:
>THANK YOU.
>
>Anyone who wants to see Linux compete on the desktop, please read this message
>- twice or three times even.
>
>Up-time is very important for servers, but if that's all you have to sell to
>desktop users, they won't buy it - even if it's free.
>
>Desktop users (end-users) want application availability and ease of use.
>
>To dig a phrase out of the old TQM handbook - "The customer decides what
>quality is." It's unpalatable, commercial, and crass, but that is a fact.
>
>I don't like this fact, but I do like Linux, an I want Linux to succeed.
>
>
>> "mlw" <[EMAIL PROTECTED]> wrote in message
>> news:[EMAIL PROTECTED]...
>>> At eastern bank, in Massachusetts, today. There was a teller trying to get
>>> information to a customer. Just as she was looking up the info,
>>> (remember their is a line of people waiting) she says to the customer,
>>> "I'm sorry, can you wait until I reboot my machine? It always does
>>> this."
>>>
>>> I asked what they used and the teller told me "Windows." I dropped my head in
>>> disgust.
>>>
>>> The world has been sold a bill of goods that is a fraud. Windows isn't
>>> usable. It is not a viable platform for any purpose for which you would use a
>>> computer.
>>
>> And yet, millions of us use Windows all day, every day, with no problems. I
>> have three machines, one running NT and two running Win2k. In the past year,
>> they've had two unexpected crashes. The Unix box on my desk, from a highly
>> respected Unix workstation vendor, has crashed once during the same time. My
>> personal experience isn't unique. We now run somewhere around 300 NT machines
>> at my site, and have no outstanding OS issues.
>>
>>> Whilst the Winvocates defend the horrible MTTF numbers on all windows
>>> platforms, I think the real core issue needs to be addressed. If you want to
>>> play games on your computer, it doesn't really matter much what you use. If
>>> you use your computer for work. You should hold it and the operating system
>>> which it runs under the same scrutiny as you would any office equipment, such
>>> as a fax machine or a copier.
>>
>> As I mentioned, we run 300 NT machines, and have virtually no problems with
>> NT4. Our biggest single problem, by a large margin, is with Netscape, which
>> crashes regularly. Since more and more of our critical applications now
>> involve accessing databases over browser based interfaces, this has been a
>> real problem. On the Unix side of things, it's even worse. Unix Netscape has
>> lagged way behind the PC versions, and for a long time we've either been
>> unable to run the applications or had terrible reliability. Until a few months
>> ago, Unix users have been unable to use the travel reservation system or the
>> expense reporting system at all.
>>
>>> This information must be made public, not just to the techies, but everyone.
>>> People that don't want to know about cars, still know about anti-lock brakes
>>> and fuel injection, because it is important for their purchasing decisions.
>>
>> Ugh. I hate automobile analogies.
>>
http://24.27.157.74/Linux/intro.html
>> I remember Peter Lynch making a case for buying Chrysler stock in Barrons a
>> few years ago. Someone pointed out that they had the worst reliability in the
>> industry. He said something to the effect of, "So what? That's not what
>> matters to people. They want a minivan with the right features, and cars with
>> good styling." He was right. Chrysler stock did very well, as Chrysler
>> delivered the features people wanted at a good price point. You could lecture
>> customers for hours about the benefits of whatever features your car has under
>> the hood, but it wouldn't really matter.
>>
>> That leaves you in the same position you were in several weeks ago, when you
>> asked how Linux was going to get onto the desktop. Your list of important
>> features in a car is different than the woman buying the minivan, and your
>> approach seems to be to try to convince her that her important feature list is
>> wrong. It's not going to work.
>>
>> Similarly, my important feature list for a computer is different than yours.
>> You can try to lecture me all day about what's more important, but at the end
>> of the day, I'm buying the product that most closely matches my feature list.
>> If you want to sell me your product, you have to start by finding out what my
>> feature list is, and trying to match it.
>>
>> As you've pointed out in the past, there's nothing Windows can do that Linux
>> can't. But there are reasons that, even though I'm a fully competent Unix
>> user, the machines I buy for myself run Windows. On the other hand, there are
>> no reasons that it has to stay that way.
>>
>> -- Mike --
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.linux.sux
Subject: Re: Why "uptime" is important.
Reply-To: [EMAIL PROTECTED]
Date: Tue, 23 Jan 2001 02:09:40 GMT
Said [EMAIL PROTECTED] in comp.os.linux.advocacy on Mon, 22 Jan 2001
>On 22 Jan 2001 14:45:23 +0100, [EMAIL PROTECTED] (Bruce Scott TOK) wrote:
>
>>You should not force your layout on your readers... this is the WWW, not
>>a magazine.
>
>One should not have to change the way one is doing something to
>accomodate the paltry number of whiners running Linux.
Indeed, you are correct. That's why we call Windows "monopoly
crapware", for forcing you to have to change the way you are doing
something in order to accommodate people doing things the correct way.
>>I only use <pre></pre> tage for stuff like author contact lists, or
>>formulae and very simple tables, for example, these ones on fusion
>>reaction energies:
>>
>> http://www.rzg.mpg.de/~bds/phys/fusion-energies.html
>
>Is that why your page looks like crap?
No, its probably because its a technical page, and therefore he doesn't
give a shit if it looks like crap. Why, do you think the nuclear
physicists are bothered by it?
>>I just checked that in Lynx so it's OK.
>
>Oh now I see why ^^^^^^^^^^^^
That's somewhat ironic, isn't it?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
** 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
******************************
