Linux-Advocacy Digest #363, Volume #26            Thu, 4 May 00 14:13:09 EDT

Contents:
  Re: Questions about trace route? (Aaron Ginn)
  Re: Government to break up Microsoft (JEDIDIAH)
  Re: Dvorak calls Microsoft on 'innovation' (Wally Bass)
  Re: The Dream World of Linux Zealots (JEDIDIAH)
  Re: X Windows must DIE!!! (JEDIDIAH)
  Re: QB 4.5 in Win 2000 (Arclight)
  Re: Virus on the net? (Max Jester)
  Oracle 8i and Mandrake 7.0 ([EMAIL PROTECTED])
  Re: Do us a favor and leave. (Was dreamers) ("Cihl")
  Linux Installation from Hell ([EMAIL PROTECTED])
  Re: QB 4.5 in Win 2000 (Damien)
  Re: QB 4.5 in Win 2000 (T. Max Devlin)
  Re: QB 4.5 in Win 2000 (T. Max Devlin)
  Re: "Technical" vs. "Non-technical"... (was Re: Grasping perspective...) (Perry Pip)
  Re: Virus on the net? (Brian Langenberger)

----------------------------------------------------------------------------

From: Aaron Ginn <[EMAIL PROTECTED]>
Subject: Re: Questions about trace route?
Date: 04 May 2000 08:54:31 -0700

Donn Miller <[EMAIL PROTECTED]> writes:

> Pig wrote:
> 
> > I am now using the Win98 and SuSE Linux.
> > I remember that there is a command name "trace route" or "route trace".
> > This cammand is like the command "ping".
> > Once I type
> > [command] www.ibm.com
> > The screen shows the route including the gateway, DNS, ... ...
> > 
> > As I forget the command,
> > Could you please help me to figure it out?
> 
> It's "traceroute".  You could've just typed "trace" and then <tab> (if
> you're using bash as your shell).
> 
> - Donn

You may not have traceroute in your path.  In my distro (Mandrake 7.0),
it's in /usr/sbin which is not normally in a regular user's path.
Type 'locate traceroute' to find out where it is, and if it isn't in
your path, add it to your .bash_profile.

HTH,
Aaron

-- 
Aaron J. Ginn

Opinions mine, not Motorola's

------------------------------

From: [EMAIL PROTECTED] (JEDIDIAH)
Crossposted-To: comp.lang.java.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Government to break up Microsoft
Date: Thu, 04 May 2000 16:21:21 GMT

On Thu, 04 May 2000 11:37:01 GMT, Otto <[EMAIL PROTECTED]> wrote:
>
>"JEDIDIAH" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>
>> Microsoft can't take the credit for that. This again is something
>> where all the 3rd parties are doing the work. Individual ISPs make
>> it easy for WinDOS users to connect to their service.
>
>And just how did the ISPs do that? Did they develop the "three click" system
>on their own and then gave it to Microsoft? Doubtful to say the least. Prior

        Actually, some of them hack netscape.

>to any modem connection to the ISPs the OS needs to setup and initialize the
>modem. Microsoft did make these steps easy on their own, without the help of

        No. Those steps were relatively trivial even before Win9x.
        You just needed to have a definitive list of init strings.
        You could buy those from 3rd parties before Microsoft finally
        decide to make that part of Windows 'more convenient' for 
        developers.

>the ISPs. Windows user's don't need to read "Sane PPP documents" to use a

        Windows users can't run their own ISP off of their PPP implemention
        either. Mind you, I already soundly established this phenomenon as
        a vendor support issue.

        Try an 'interesting' network (like compuserve's) under Win9x.

>modem. On the Linux side Caldera's 2.4 version, AFAIK the only one, does the
>same as Windows. It correctly detects and sets up the US Robotics modem. To
>make the connection to the web, enter the phone number and user's info and
>off it goes. No need for editing cumbersome config files. 5 years after
>Windows95, but nonetheless working just fine.

        There were shiny happy PPP tools for Linux in 1995.

        However, even under Win95 something would occasionally fall beyond
        the bounds of where system developers had predicted. At times like
        those you need to be able to look under the hood (even on Win95).

-- 

                                                                        |||
                                                                       / | \
        
                                      Need sane PPP docs? Try penguin.lvcm.com.

------------------------------

From: wallyb6@nospam (Wally Bass)
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Dvorak calls Microsoft on 'innovation'
Date: Thu, 04 May 2000 17:29:22 GMT

On Wed, 3 May 2000 18:52:26 -0500, "Erik Funkenbusch" <[EMAIL PROTECTED]>
wrote:

>I'll give you the same challenge I gave Mig Mig.  Prove that every one of
>microsofts patents are prior art.

My, you're so generous with the feasibility of you challenges. Since
you are allowing your opponent to prevail merely by PROVING
the case on EVERY patent, I would be inclined to make an equally
generous offer to you! Why don't you PROVE that EVERY
Microsoft patent is (a) innovative enough that everyone would
agree that is indeed real innovation, and (b) that it was not
prior art.

I'm not often moved to make it so easy for an opponent, but
in your case, I was so struck by the generousity of your offer
that I just had to make an exception.

WallyB

------------------------------

From: [EMAIL PROTECTED] (JEDIDIAH)
Subject: Re: The Dream World of Linux Zealots
Date: Thu, 04 May 2000 16:28:50 GMT

On Thu, 04 May 2000 13:40:58 GMT, [EMAIL PROTECTED] <[EMAIL PROTECTED]> 
wrote:
>[EMAIL PROTECTED] (JEDIDIAH) writes:
>>On Tue, 02 May 2000 09:19:09 GMT, [EMAIL PROTECTED] <[EMAIL PROTECTED]> 
>wrote:
>
>>>OK, here is another question: Why didn't Win98 include the "find computer"
>>>program when it installed on my laptop? It's a real pain --- instead of
>
>>      ???
>
>>      You can put the IP of your Linux Samba server in lmhosts if you
>>      really need to. You should be able to browse a Linux samba share
>>      under '98.
>
>I also didn't get a "network neighbourhood", or anything else related to
>sharing resources. I'd love to know why (and would love even more to know
>how to overcome the problem), but quite frankly, I am stumped.

        98 really seems to flake out at times. If it were only the Linux
        clients that dissappear I might be inclined to think that we are
        seeing samba bugs here. However, '95 and '98 can manage to stop
        seeing themselves as well...

>
>>>simply copying back and forth from the Samba exported filesystems of my
>>>linux boxes, I have to use the (sickening) ftp client or put the files
>>>up on my linux box's http server, and use IE to download them.
>
>>      ...samba.org has a really nice troubleshooting guide.
>
>Oh, the problem isn't on the Samba side. I have set up several machines
>to share my resources. But this machine doesn't have anything on it that
>would allow me to even start. No idea *why*, though.

        Mebbe some of these Windows show-offs can shed some light on these
        Win9x <-> Win9x type smb connectivity issues...

-- 

                                                                        |||
                                                                       / | \
        
                                      Need sane PPP docs? Try penguin.lvcm.com.

------------------------------

From: [EMAIL PROTECTED] (JEDIDIAH)
Crossposted-To: comp.os.linux.x
Subject: Re: X Windows must DIE!!!
Date: Thu, 04 May 2000 16:32:45 GMT

On Thu, 04 May 2000 13:17:03 GMT, bytes256 <[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>,
>  [EMAIL PROTECTED] wrote:
>> It was the Mon, 01 May 2000 16:37:32 GMT...
>> ...and Marty Dippel <[EMAIL PROTECTED]> wrote:
>> > Yeah- let's throw out X and rewrite it.  And NFS.  And the
>filesystem(s).
>> > And PGP.  And SSH.  And the printing system.  And the man pages.
>> > They all suck.  Let's re-write it all, not document any of it, and
>proclaim
>> > it all to be the new "standard".  And then we can all wink
>knowingly at each
>> > other, convinced that we've done something useful by creating a
>much cooler
>> > *nix than any of the other *nixes.  And *our* stuff won't suck at
>all, 'cause
>> > we're so much smarter than the unimaginative twits that wrote the
>originals!
>>
>> If you strike "not document any", this sounds like the stance of the
>> VMS developers -- or the Windows developers, for that matter :)
>>
>> mawa
>> --
>> But when she was with Chacko, old limits were pushed back. Horizons
>> expanded. She had never before met a man who spoke of the world [...]
>> in the way other men she knew discussed [...] their weekends at the
>> beach.                     -- Arundhati Roy, _The_God_of_Small_Things_
>>
>
>My point is quite simply put: XWindows does not best meet the needs of
>the average Linux user.  It is far more complicated than necessary.

        No it isn't. The average user typically only needs to let the
        X configuration utility probe the hardware and then chose 
        some resolutions and bit depths.

        This notion that end users typically need to manually tweak
        modelines is more urban legend and FUD than anything else.

>And then it leaves out important functionality that people want.
>(Standardized controls, High performance, easy installation, etc.)
>
>Don't shoot the messenger...revolutions have to start somewhere.

        X has high performance, it can even go toe to toe with DirectX
        in 3D rendering (dri).
        X has easy installation and has had it for quite awhile now.

        Stiffling standardization is of questionable value.


-- 

                                                                        |||
                                                                       / | \
        
                                      Need sane PPP docs? Try penguin.lvcm.com.

------------------------------

From: [EMAIL PROTECTED] (Arclight)
Crossposted-To: alt.lang.basic,alt.destroy.microsoft
Subject: Re: QB 4.5 in Win 2000
Date: Thu, 04 May 2000 16:32:03 GMT

On 04 May 2000 09:38:30 -0600, Craig Kelley <[EMAIL PROTECTED]>
wrote:

>[EMAIL PROTECTED] (Arclight) writes:
>
>> >And in 2001, or 2002?
>> 
>> well you can still buy office 95 from some places, so I think it'd be
>> pretty safe to say that you will probably be able to buy office 97 in
>> 2002,
>> 
>> >That doesn't solve the problem.
>> 
>> But why should you force microsoft to sell outdated products just to
>> support the few people who might want them?
>
>It's commonly called the "upgrade treadmill" by us "few people".
>
>We hate it.

well that's tough.
are you suggesting that microsoft should continue to support all its
old software because people might still be using it?


>> anyway what's stopping you using office 2000 on some of the computers
>> and 97 on others?
>
>File incompatibilites.

These can be avoided by using the right export/import filters.

>Don't tell me they don't exist; we've already encountered several
>PowerPoint problems (and we don't even have Office 2000 anywhere in
>the building).
Try installing all the import/export filters that are on the office
2000 CD, they should get rid of any problems.


TTFN
Arclight

Web Site:
http://www.daniel-davies.pwp.blueyonder.co.uk/

------------------------------

From: Max Jester <[EMAIL PROTECTED]>
Subject: Re: Virus on the net?
Date: 4 May 2000 16:32:52 GMT

Martijn Bruns <[EMAIL PROTECTED]> wrote:
> Has anyone heard of a new virus on the net?

Yup

> It's called 'I Love U'. (or something that sounds like that) and
> it seems to be infecting a LOT of companies around the globe
> right now!
> Also, it seems to infect Windows PC's running Outlook (Express?).

> It was on the radio surrounded by a mild case of panic :-)

> Does someone know more about it? Could something like this affect
> Linux-machines also?

Comes in email with a subject line of "ILOVEYOU" - it seems to be pretty
specific to win9x; goes in and diddles your registry, and sends itself
to your friends ... I haven't read the whole thing in detail yet.

If you're running Linux, it won't affect you.

HTH
Max
-- 
Speaking only for myself.
This message is over.  Why are you still reading?  Stop at once!

------------------------------

From: [EMAIL PROTECTED]
Subject: Oracle 8i and Mandrake 7.0
Date: Thu, 04 May 2000 16:51:38 GMT

I try to install Oracle 8i on the Linux Mandrake 7.0. I got the JRE 116
V5 because the installation process need this product. I put the all
libraries under the path /usr/local/jre in fact it is the link as ln -
s /usr/local/jre116_V5 /usr/local/jre. After that i start the process
from the CD of Oracle runInstaller and i got the following messages
Jaca script... and so on and the fatidique message "mauvaise adresse"
so how can i do to install Oracle. Can somebody help me?



Sent via Deja.com http://www.deja.com/
Before you buy.

------------------------------

From: "Cihl" <[EMAIL PROTECTED]>
Subject: Re: Do us a favor and leave. (Was dreamers)
Date: Thu, 04 May 2000 17:10:05 GMT

Well it IS true what you say. The Open Source-way seems to work very nicely
indeed.
As we know, it has already brought forward several entire Operating Systems,
from which one in particular :-) seems to be gaining more and more attention
from the big, dusty corporations like Microsoft, IBM and such.
There's really no way of stopping our development, even with the trolling we
see on this newsgroup on a regular basis. Linux, after all, is not based on
classic commerce.

"Salvador Peralta" <[EMAIL PROTECTED]> schreef in bericht
news:[EMAIL PROTECTED]...
> I spent the better part of my morning wading through rpg code and cl's
> on an AS/400 trying to debug an application that was purchased for
> aseveral thousand from a private vendor.  The vendor built a binary
> object which is a functional equivalent of a cgi script that takes user
> input, stores it in a database, passes some of the data to sendmail
> which then routes email and makes a sends a synamic http response.
>
> The old application (if we were to keep it) would have to get called
> from a different proprietary interface, and there is some question of
> whether all of the variables will get called correctly from the new
> programming environment.  Since we did not have rights to the source
> over the compiled object, we cannot fully test the application in the
> new environment without paying additional consulting fees.
>
> After spending 3-4 hours doing due diligince testing the legacy
> application, i wrote a perl script that handles the same functions on
> linux from an html form.  If we were so inclined, we could serve it on a
> cheap server that handles nothing but scripts.  Total cost: 8 hours
> salary
>
> ~ 3 hours of analysis
> ~ 2-3 hours documentation on old system process
> ~ 1 hour writing and testing perl scripts.
> ~ 1 hour developing the visual interface
>
> Cost of System is less than $1500 per annum including bandwidth and
> maintenance
>
> We could host that single script on a server which server no other
> purpose and including hardware and labor, it would cost less than 2 days
> of consulting on the original proprietary application.
>
> That's the power of open source.
>
> Cihl wrote:
> >
> > But you'll see, Linux will catch up sooner or later. Linux isn't there
to
> > destroy Microsoft, Linux these days is there primarily there to test a
> > different philosophy behind making software. And, to be honest, for the
> > sheer -fun- of it. :-)
>
> --
> Salvador Peralta
> [EMAIL PROTECTED]
> http://www.la-online.com



------------------------------

From: [EMAIL PROTECTED]
Subject: Linux Installation from Hell
Date: Thu, 04 May 2000 17:12:38 GMT

  It's been a very long story, so I'll try to keep it short. My question may
be a little hard to understand. Specifically, I've built another barely
functional Linux box from Peanut Linux 7.6 to replace the destroyed Linux
Mandrake 6.0 installation, and from some RPM packages from the hell-damned
Linux-Mandrake 7.0-2 CD-ROM. I've used the FHS and some other information to
build the custom (under construction) basic filesystem (the directory
structure, not the ext2 filesystem on that shared partition) manually, in a
directory on the same partition as the working Peanut Linux 7.6 installation.

  I wish to know how to begin to build a custom working Linux in this
directory. To this end, the very first thing I need to know to continue is
whether the Peanut Linux 7.6 kernel (it's from the 2.2.14 stable release
source code, apparently) can just be placed into the appropriate directory in
the custom Linux branch, and used to boot with no other access to what is on
the rest of the partition. Is "init" the only totally necessary program? Can
I start with that and build up from there to installing the gcc compilers
(again) in this custom installation, and from there to recompiling the kernel
and all the rest of the packages that might go into a custom Linux?

   I warned you that the question would be hard to understand. :)

  P.S. Yes, this is *exactly* what I want to do. It's been very hard to get
information about this -- the "chicken and egg" problem seems to utterly
escape most people, even so-called "Linux gurus". They seem to be unable to 
understand that some people just have extraordinarily bad luck with most
automated installation scripts, and that sometimes it just necessary to do
stuff the hardest of hard ways to end nightmarish, endless Catch-22's.


Sent via Deja.com http://www.deja.com/
Before you buy.

------------------------------

From: [EMAIL PROTECTED] (Damien)
Crossposted-To: alt.lang.basic,alt.destroy.microsoft
Subject: Re: QB 4.5 in Win 2000
Reply-To: [EMAIL PROTECTED]
Date: 04 May 2000 17:27:11 GMT

On Thu, 04 May 2000 16:32:03 GMT, in alt.destroy.microsoft,
Arclight <[EMAIL PROTECTED]> wrote:
| On 04 May 2000 09:38:30 -0600, Craig Kelley <[EMAIL PROTECTED]>
| wrote:
| 
| >[EMAIL PROTECTED] (Arclight) writes:

| >> anyway what's stopping you using office 2000 on some of the computers
| >> and 97 on others?
| >
| >File incompatibilites.
| 
| These can be avoided by using the right export/import filters.
| 
| >Don't tell me they don't exist; we've already encountered several
| >PowerPoint problems (and we don't even have Office 2000 anywhere in
| >the building).
| Try installing all the import/export filters that are on the office
| 2000 CD, they should get rid of any problems.

He doesn't have Office 2000.  

Another thing, the default on all these programs is to save into the
new incompatible file format.  Is there reason for this other than to
turn the upgrade treadmill?  Is there any way to change the system to
make it save into the older formats by default?  And why aren't these
file formats backwards compatible?  Any decent file format would have
an extensible design making new versions automatically compatible with
older versions.
        

------------------------------

From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.lang.basic,alt.destroy.microsoft
Subject: Re: QB 4.5 in Win 2000
Date: Thu, 04 May 2000 13:32:45 -0400
Reply-To: [EMAIL PROTECTED]

Quoting Arclight from alt.destroy.microsoft; Thu, 04 May 2000 16:32:03 GMT
>On 04 May 2000 09:38:30 -0600, Craig Kelley <[EMAIL PROTECTED]>
    [...]
>>> But why should you force microsoft to sell outdated products just to
>>> support the few people who might want them?
>>
>>It's commonly called the "upgrade treadmill" by us "few people".
>>
>>We hate it.
>
>well that's tough.
>are you suggesting that microsoft should continue to support all its
>old software because people might still be using it?

I think he's suggesting, as am I, that Microsoft should continue to support
its *customers*, even after its taken their money, as an on-going license does
in fact indicate *some* level of long-term commitment.  Let me ask you this:
when you bought your licenses, did you see any indication whatsoever that it
was going to be made almost entirely useless by the guy who sold it to you
after only 18 to 24 months?

Microsoft's well documented history of spuriously "upgrading" simply to "hold
up its customers for more money" so that they can continue to benefit from
their original investment (and, indeed, incur even more of an investment in
the installed system outside of monetary costs) makes yours a straw-man
argument at best, and so pointedly dismissive as to seem ludicrous.

   [...]
>>File incompatibilites.
>
>These can be avoided by using the right export/import filters.

This is a trivial response to a grievous problem; unmitigated teflon bullshit.
If Microsoft didn't know damn well that leveraging of file incompatibilities
didn't provide *massive* coercion for the installed base to upgrade, they
wouldn't do it.  In other words: these can be avoided by using someone else's
software that doesn't so whole-heartedly support the "upgrade merry-go-round"
scam.  Filters are handy in a pinch, but they are *not* the equivalent of
functional inter-operability.


--
T. Max Devlin
Manager of Research & Educational Services
Managed Services
ELTRAX Technology Services Group 
[EMAIL PROTECTED]
-[Opinions expressed are my own; everyone else, including
   my employer, has to pay for them, subject to
    applicable licensing agreement]-


====== 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: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: alt.lang.basic,alt.destroy.microsoft
Subject: Re: QB 4.5 in Win 2000
Date: Thu, 04 May 2000 13:40:19 -0400
Reply-To: [EMAIL PROTECTED]

Quoting Damien from alt.destroy.microsoft; 04 May 2000 17:27:11 GMT
    [...]
>Is there any way to change the system to
>make it save into the older formats by default?

Depends on the program.  I haven't seen Office 2000 yet, but in Office 97, I
think all three (Word, Excel, PowerPoint) finally have this feature.  Not that
it is really anywhere near as functional as you would imagine.  The
applications don't, for instance, register themselves as servers for the older
file types.  So even though they can read the file, and the extension is the
same, any program other than the Operating System that tries to launch a file
as an embedded object and the like will fail.  I would guess that this is
excused with claims that it allows alternate applications to support older
file types, but that doesn't wash, as none of Microsoft's applications are
capable of co-existing with older versions very well in any other way.

The point is that "save as" is a barely expectable work-around for trivial
cases, not a valid way of addressing the theft of control of a user's data
which Microsoft's handling of file formats constitutes.


--
T. Max Devlin
Manager of Research & Educational Services
Managed Services
ELTRAX Technology Services Group 
[EMAIL PROTECTED]
-[Opinions expressed are my own; everyone else, including
   my employer, has to pay for them, subject to
    applicable licensing agreement]-


====== 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] (Perry Pip)
Subject: Re: "Technical" vs. "Non-technical"... (was Re: Grasping perspective...)
Reply-To: [EMAIL PROTECTED]
Date: Thu, 04 May 2000 17:42:07 GMT

On 4 May 2000 10:48:36 GMT, Donal K. Fellows <[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>,
>Perry Pip <[EMAIL PROTECTED]> wrote:
>> Let's say you are asynchronously waiting on 100 descriptors, i.e. you have
>> 100 async i/o calls outstanding. When one becomes ready, you want to
>> process it and then make a new async i/o call on it. 
>> 
>> If the library uses 100 threads, each one waiting on a single i/o
>> operation (read(), write(), etc.) then they are all sleeping. [...]
>> 
>> Now lets say the library uses one thread, waiting on the poll()
>> call. [...]
>
>The problem with using a hundred threads as opposed to a single thread
>is that every thread has to have its own stack, and you are doing very
>well in most practical situations if you can guarantee that each
>thread will only need a single page for its stack.  

I am very well aware of issues related to having alot of threads, in
regards to stack, scheduling, etc. However, in the particular context of
my post, i.e. implementing a POSIX async i/o API in glibc, each thread
simply makes a single i/o call, waits on it, then sends a signal after
returning from the call, and then dies. I fail to see how that could use
more than one stack page.



>In contrast, the
>single threaded scheme will have a single stack and will therefore
>have much lower memory overhead.  Plus, it will have better
>locality-of-reference properties and so will be able to run faster.

But you are overlooking the other issues in my post (which you
conveniently erased) in regards to implementing an async i/o API with a
single thread using poll(). In particular, every single time a new async
i/o is made the library must forcefully interrupt the thread's poll() call
with a pipe to to have the thread add a new descriptor to the array and
make the poll() call again. And everytime the thread makes or returns from
the poll() call the kernel must check the status of *all* the descriptors
in the array. Furthermore, your thread must scan the array everytime it
returns from the poll() call. This becomes a problem on the order of
O(n^2). In contrast, creating a thread and having it wait on single i/o
call, return and send a signal and then die for each async i/o call made
is of order O(n).


>FYI, high-end server software tends to only spawn as many threads as
>there are processors available.  

Perhaps you have understood my post out of context. I wasn't talking about
server software. I was discussing the specific implemention of POSIX async
i/o that is in glibc. And I didn't say their solution was optimal. But I
certainly understand why they chose not to use a single thread.

>More is usually counter-productive as
>the additional costs bite harder than any increase in performance.

For a good rundown on the issues invlolved in regards to threads vs.
poll() for handling i/o events take a look at:

http://www.atnf.csiro.au/~rgooch/linux/docs/io-events.html



>(Not too many people have hundred-node multiprocessor machines,

We have 128 CPU SGI/IRIX machines were I work, but they are used in very
specialized engineering applications, not as servers.

>Beowulf notwithstanding!)
>


Beowulfs don't do threads, they do processes, at least across multiple
nodes. IPC between Beowulf nodes is via message passing, not shared
memory. (Although buiding a beowulf like cluster using something like
reflective memory would be, to say the least, interesting.)

Perry

------------------------------

From: Brian Langenberger <[EMAIL PROTECTED]>
Subject: Re: Virus on the net?
Date: 4 May 2000 17:57:37 GMT

Max Jester <[EMAIL PROTECTED]> wrote:

<snip!>

: Comes in email with a subject line of "ILOVEYOU" - it seems to be pretty
: specific to win9x; goes in and diddles your registry, and sends itself
: to your friends ... I haven't read the whole thing in detail yet.

: If you're running Linux, it won't affect you.

Sure it'll affect you.  If you're running Linux at the office,
you're probably the one that has to update sendmail's filters
in order to keep everyone safe from another innovation Microsoft
needed the freedom to create.  :)


------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.advocacy) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Advocacy Digest
******************************

Reply via email to