Subject: Linux-Misc Digest #455
From: Digestifier <Linux-Misc-Request@senator-bedfellow.MIT.EDU>
To: Linux-Misc@senator-bedfellow.MIT.EDU
Reply-To: Linux-Misc@senator-bedfellow.MIT.EDU
Date:     Mon, 20 Dec 93 18:13:16 EST

Linux-Misc Digest #455, Volume #1                Mon, 20 Dec 93 18:13:16 EST

Contents:
  Running off LGX cd-rom with small disk (John Bonn)
  Re: Windows emulation was Re: Microsoft Invented Inferior Personal C (Mike Dahmus)
  Re: Xwindows Terminal (Mark A. Davis)
  Re: Linux in a hospital? (Magnus Redin)
  Re: Linux in a hospital? (Kai Petzke)
  Re: Windows emulation was Re: Microsoft Invented Inferior Personal C (Jonathan Hacker 21420)
  Re: Linux in a hospital? (Kai Petzke)
  Re: Linux in a hospital? (Kai Petzke)
  3C509 Ethernet Card/ (Farmer's Copper and Industrial Supply)
  Re: Question about fractal compression (Christopher Shaulis)
  Re: Linux installation from CD (Christopher Shaulis)
  Re: Yet another benchmark results.. (John Francis Lynn)
  Re: capturing boot messag (John Will)
  Re: Windows emulation was Re: Microsoft Invented Inferior Personal C (The Great Winged One)
  Re: Linux in a hospital? (Chris Higgins - System Administrator)

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

From: bonn@unislc.slc.unisys.com (John Bonn)
Subject: Running off LGX cd-rom with small disk
Date: Mon, 20 Dec 1993 17:24:20 GMT

I have done a run-time installation and I would like the
uninstalled software (the stufff on the cd) to be 
usable.  The key is I would like it to be _transparent_
to the user that some things are installed and other
things are on the cd.  I was thinking of making links
in my hard drive tree back to where the cd-rom is mounted
thereby making it appear as though I've done a complete
installation.  Is there a better way?  Has anyone written
a script to do what I have in mind (I'm assuming there
will be a couple gotcha's that someone has already fixed).

Also, is there a cd-rom filesystem for Linux that caches 
frequently used files on the hard disk?  Is it in the LGX
distribution or where can I get information on it?

Thanks,
JP

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

Crossposted-To: alt.folklore.computers,alt.religion.kibology,comp.os.ms-windows.advocacy,alt.fan.mike-dahmus
From: mike@schleppo.bocaraton.ibm.com (Mike Dahmus)
Subject: Re: Windows emulation was Re: Microsoft Invented Inferior Personal C
Date: Mon, 20 Dec 1993 18:27:59 GMT
Reply-To: miked@vnet.ibm.com (Mike Dahmus)

In <2etbnq$hkm@wcap.centerline.com>, jimf@centerline.com (Jim Frost) writes:
>
>In all likelihood you locked up your X server by having an application
>do a server grab and not relinquish it.  This was very common with
>early Motif applications, and is still easy if you're debugging Motif
>applications today.  Killing the application frees up the server.

Regardless of whether it was the server or the app; if I hadn't been able to
telnet in from another machine, it would have been Big Red Switch time. OS/2
at least gives a fair chance at being able to resolve these problems from the
console (with the watchdog timer). If that fails, I can also try to telnet in
to the OS/2 box and kill the offending process. If that fails, well, at least
OS/2 doesn't take as long to boot as the average unix box :+)

======
Mike Dahmus                                       Internet: miked@vnet.ibm.com
Pen for OS/2 Development, IBM PSP         IBM: mike@schleppo.bocaraton.ibm.com
Disclaimer: Not an official IBM spokesman            IBM Vnet: MDAHMUS at BOCA


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

From: mark@taylor.wyvern.com (Mark A. Davis)
Subject: Re: Xwindows Terminal
Date: Mon, 20 Dec 1993 18:29:03 GMT

woody@altair.stmarys-ca.edu (Woody Weaver) writes:

>Keith Smith (keith@ksmith.com) wrote:

>>One of the Cheapest X-terminals you can buy is a 486/33 with a VLB
>>Graphics accellerator card 8MB of RAM, and ethernet card, KB, mouse and
>>nice monitor, running Linux <G>!

>How about running DOS and Desqview/X?

What about it?  If the host can do it, then all Xterminals attached can too...
Ala Merge, VP/ix, WABI, Wine, DOS-Emu, Soft-PC, etc...

>Its ~$300US, but the constraints upon
>the client are less, and some people would prefer running under DOS rather
>than Un*x.  Sacrilegeous, I know, but...

If one primarily wants to run MS-"DOS" software, then I'm sure that Unix,
Linux, and Xterminals are not what that person is looking for.
-- 
  /--------------------------------------------------------------------------\
  | Mark A. Davis    | Lake Taylor Hospital | Norfolk, VA (804)-461-5001x431 |
  | Sys.Administrator|  Computer Services   | mark@taylor.wyvern.com   .uucp |
  \--------------------------------------------------------------------------/

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

From: redin@lysator.liu.se (Magnus Redin)
Subject: Re: Linux in a hospital?
Date: Mon, 20 Dec 1993 16:54:03 GMT

rob@pe1chl.ampr.org (Rob Janssen) writes:

>What worries me more is "will Linus ever graduate"....   with such a large
>project his studies may well suffer, and in the end we may be left with
>a nice operating system but a person who has overrun the time allocated
>for his study and no graduation...  That would be a high price to pay for
>being famous.
>Or maybe that cannot happen in Finland?

Does it matter? The proper reaction from a hierer is "You are THE Linus
Torvalds!!!, What salary do you want?"
I would immediatly try to emply him _now_ if we had more monye to
spare in our company Signum Support. (We sell support on free software.)
He is probably drowning in suggestions already. He will probably be
signed up when we get big enought. *sigh*

--
--
Magnus Redin  Lysator Academic Computer Society  redin@lysator.liu.se
Mail: Magnus redin, Rydsv{gen 240C26, 582 51 LINK|PING, SWEDEN
Phone: Sweden (0)13 260046 (answering machine)  and  (0)120 13706

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

From: wpp@marie.physik.tu-berlin.de (Kai Petzke)
Subject: Re: Linux in a hospital?
Date: 20 Dec 1993 19:56:55 GMT

In <1993Dec20.082834.4409@unlv.edu> ftlofaro@unlv.edu (Frank Lofaro) writes:

>In article <2f1f2i$d9l@mailgzrz.TU-Berlin.DE> wpp@marie.physik.tu-berlin.de (Kai Petzke) writes:
>>
>>The system is for printing labels in a blood bank.  A system failure
>>wouldn't endanger anybody (they can return to writing labels from
>>hand).  The thing, that must not happen, are exchanged blood groups.
>>There are precautions in my system, like storing the blood group
>>several times in different tables, checksumming the blood group, etc.
>                                    ^^^^^^^^^^^^
>Actually, using a CRC would be _MUCH_ better.

You are right.  I will either use CRC or PGP's feature of creating
"signatures".  The later had the advantage, that even hackers could
not change the blood group, if they did not know the password of the
secret key.  Only doctors will know them, because they are the only
persons allowed to change a blood group.

The disadvantage is, that PGP signatures take up a lot of space, and
that checking them takes notably time, so I have not made up my mind
about it yet.  CRC's are very fast, and give you good reliability,
too.

>Checksums aren't as strong as one would hope.
>E.g. exchanged bytes don't phase a checksum check, but a CRC might spot it.

>Heck, I only trust critical data to CRC's (like gzip or kermit block 3 setting).

I know the complete source of gzip.  The problem is, that it is GNU
copyrighted software, so I should not be allowed to "steel" the
CRC code from it into my software.  On the other hand, it is one simple
function with one simple loop.  Perhaps, I should ask the author for
permission.


>Also be very wary of possible loss of data integrity of RAM.
>Make sure you have data integrity/consistency tests on data in memory.

Doesn't PC memory have parity bit checking?  So you should be able
to tell, if there are lots of memory errors going on.

But it is a good idea to protect the data in memory.  Not from hardware
failure, but from software failure.  Using X for the user interface
means, that I have lots of only 99% reliable code around, which might
end up incrementing the value in a memory cell, which contained "A"
before.

>Also, make sure your PC is in good shape. (no hidden/subtle problems).

Of course no overclocking, etc. in such an environment.

>I hope this is helpful.

Kai
--
Kai
wpp@marie.physik.tu-berlin.de
Advertisement by Microsoft in a well-known German magazine:
        If you don't like our programmes, then make your own ones.
However, they expect you to use Microsoft products for this -:)

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

Crossposted-To: alt.folklore.computers,alt.religion.kibology,comp.os.ms-windows.advocacy,alt.fan.mike-dahmus
From: hacker@patagonia.bellcore.com (Jonathan Hacker 21420)
Subject: Re: Windows emulation was Re: Microsoft Invented Inferior Personal C
Reply-To: hacker@patagonia.bellcore.com
Date: Mon, 20 Dec 1993 19:38:33 GMT

In article <2etb0u$h8q@wcap.centerline.com>, jimf@centerline.com (Jim Frost) writes:
>tso@cephalo.neusc.bcm.tmc.edu (Dan Ts'o) writes:
>>The times that UNIX has crashed even once during a session,
>>over a 15 year period, I could count on one hand.
>
>You must not have used Solaris yet :-).
>

...and if you have, you know that Solaris randomly panics/reboots by
itself.  No need to worry about ctrl-esc at all? :-) :-)

-- 
Jon Hacker                         |   OS/2 2.1
Bellcore, Red Bank NJ              |   What NT was supposed to be! 
hacker@patagonia.bellcore.com      |   

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

From: wpp@marie.physik.tu-berlin.de (Kai Petzke)
Subject: Re: Linux in a hospital?
Date: 20 Dec 1993 20:03:13 GMT

In <1993Dec19.112836.30701@pe1chl.ampr.org> rob@pe1chl.ampr.org (Rob Janssen) writes:

>What worries me more is "will Linus ever graduate"....   with such a large
>project his studies may well suffer, and in the end we may be left with
>a nice operating system but a person who has overrun the time allocated
>for his study and no graduation...  That would be a high price to pay for
>being famous.
>Or maybe that cannot happen in Finland?

I think, that Linus will never, never have a problem to find a job,
whether graduated or not.  There are lots of companies around, that
use Linux for development, and I doubt any of these would reject
Linus, when he was looking for a job.

The other change is, that Linux keeps getting more and more popular.
I think, that distributors and re-distributors of Linux should pay
a small amount (say 1% of the end price) to something like a "Linux
foundation".  From that money, you could easily pay Linus a regular
income.

It won't reach that of Bill Gates, though.
--
Kai
wpp@marie.physik.tu-berlin.de
Advertisement by Microsoft in a well-known German magazine:
        If you don't like our programmes, then make your own ones.
However, they expect you to use Microsoft products for this -:)

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

From: wpp@marie.physik.tu-berlin.de (Kai Petzke)
Subject: Re: Linux in a hospital?
Date: 20 Dec 1993 20:18:37 GMT

In <CIByMA.Kzz@world.std.com> entropy@world.std.com (Lawrence Foard) writes:

>In article <2f1f2i$d9l@mailgzrz.tu-berlin.de>,
>Kai Petzke <wpp@marie.physik.tu-berlin.de> wrote:
>>
>>The system is for printing labels in a blood bank.  A system failure
>>wouldn't endanger anybody (they can return to writing labels from
>>hand).  The thing, that must not happen, are exchanged blood groups.
>>There are precautions in my system, like storing the blood group
>>several times in different tables, checksumming the blood group, etc.

>This is a good example of where trusting any OS is a bad idea. An incorrect
>label will kill someone.

In 99% of the cases, it will not.  Before transfusion, blood of the
donator and blood of the receiptor (the word may be wrong, sorry,
but I guess, you know what I mean) are checked against each other.

>You have to assume that anything in the system can
>and will fail, a perfect OS running over broken hardware will still fail,
>what if the printer decides that its only going to print the letter O that
>day :) 

>If at all possible you should make sure it is double checked by hand, for
>example the person applying the label should know what it should be.

The current system is by hand, and involves more triple checks than
double checks.  After installation of the computer, the checks won't
disappear.

The person who prints the labels, has to check the blood group once
again on the label.  When taking the donation, they have to verify
the donators name once again.  Not only to check for hardware or
software failure.  It could have been, that she typed the number of
the donator wrong.  There is a check digit with each number, but that
can be fooled, if you type two or more wrong digits.  Or the blood
group was entered wrong twice.

>The mechanical switchs in microwave doors make me feel much better than
>it would if it went through the micro controller....

I think so, too.  For rexactly the same reason you stated.

>Simple answer to your question:
> No OS Linux included is reliable enough for life critical work, computers
> are to complex to comprehend all the failure modes (sorry I don't take 
> people who claim to "prove" programs seriously), make sure there is
> something simple and failsafe as a backup.

There will be.  The double checks will be done by the stuff of that
clinic the same way, it used to be.  I agree, that a computer may
only be introduced, if it increases safety.  I currently assume, it
does.  For example, the labels printed by the machine will be much
better readable than those generated with a typewriter.
--
Kai
wpp@marie.physik.tu-berlin.de
Advertisement by Microsoft in a well-known German magazine:
        If you don't like our programmes, then make your own ones.
However, they expect you to use Microsoft products for this -:)

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

From: fcis@NeoSoft.com (Farmer's Copper and Industrial Supply)
Subject: 3C509 Ethernet Card/
Date: Mon, 20 Dec 1993 14:57:21 GMT

        I have SLS 0.99 Linux system and a 3COM 3C509 ethernet card. 
I tested this card with their disk.(work on DOS). But it doesn't work
for Linux.  The message I got from Linux is "No ethernet device found"
I had a 3C503 work without any problem before.  May anyone in the net tell me
how to change the setting and let it work?     
        Thank you in advice. 
        Julia
        Email: fics@sugar.NeoSoft.com


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

From: cjs@netcom.com (Christopher Shaulis)
Subject: Re: Question about fractal compression
Date: Mon, 20 Dec 1993 20:24:13 GMT

el@lisse.NA (Dr Eberhard W Lisse) writes:

>Hi,

>does anyone know of a program with which to compress images by way of
>fractal compression to run on linux (X)?

I believe there is a world-wide patent pending to fractal compression 
technology which eliminates most all commerical sources and I haven't 
seen an gnuish implimentation either.
 
If you *really* need it, the folks who invented it sell source for about 
$500 though I think they also want royalities. If your just interested, 
BYTE had an article on the subject about three months back. You can 
probably still find it at the library.

Christopher

  ___     _  ___   ____  _  _ ___ _____  ___  ___  __  __     ___  ___  __  __ 
 / __|_  | |/ __| / __ \| \| | __|_   _|/ __|/ _ \|  \/  |   / __|/ _ \|  \/  |
| (__| |_| |\__ \/ / _` | .` | _|  | | | (__| (_) | |\/| | _| (__| (_) | |\/| |
 \___|\___/ |___/\ \__,_|_|\_|___| |_|  \___|\___/|_|  |_|(_)\___|\___/|_|  |_|
==================\____/=======================================================


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

From: cjs@netcom.com (Christopher Shaulis)
Subject: Re: Linux installation from CD
Date: Mon, 20 Dec 1993 20:34:04 GMT

vic@france.hp.com (Vichharaks ROS ) writes:

>I should like to bye Linux on CDROM, but i have some questions before:
>a) Can i use it directly from the CDROM ?
>b) Do i need to make a new partition on my disk ?
Within reason. Certain config files and key executables should be on disk 
because either they need to be changed or you'll pull your hair out 
waiting for them to run of CD. Minimum Yggdrasil instalation in 2 MB, 
though I wouldn't suggest trying to run X off it. If your serious about 
linux, I'd set yourself up a nice little 30-40 meg partition for it to 
live in. You better ahev about 8 megs of memory too (X is big).

>c) I should like to use X11R5, what is the partition size that i need ?
>d) Is it possible to have Motif 1.2 ?
For a price. Motif is copyright and thus must be purchased.

Christopher

  ___     _  ___   ____  _  _ ___ _____  ___  ___  __  __     ___  ___  __  __ 
 / __|_  | |/ __| / __ \| \| | __|_   _|/ __|/ _ \|  \/  |   / __|/ _ \|  \/  |
| (__| |_| |\__ \/ / _` | .` | _|  | | | (__| (_) | |\/| | _| (__| (_) | |\/| |
 \___|\___/ |___/\ \__,_|_|\_|___| |_|  \___|\___/|_|  |_|(_)\___|\___/|_|  |_|
==================\____/=======================================================


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

From: jlynn@x1.engin.umich.edu (John Francis Lynn)
Crossposted-To: comp.sys.ibm.pc.hardware,comp.os.vms
Subject: Re: Yet another benchmark results..
Date: 20 Dec 1993 20:51:59 GMT


In article <wBTTec3w165w@octogard.ocunix.on.ca>, graham@octogard.ocunix.on.ca (Graham Mainwaring) writes:
> carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) writes:
> 
> > Why is it that folks who put their faith in benchmarks are generally
> > clueless?
> 
> Because benchmarks are generally meaningless.(Not just this one,all of
> them.)
> 

"There are three kinds of lies: Lies, Damned Lies & Statistics"
                                           ..... Disraeli

Substituting "benchmarks" for "statistics" in the above works 
just as well :-)

But seriously, haven't we compared apples and oranges long enough now
to return to something more mundane ?

John
  
-- 
| John F. Lynn <jlynn@engin.umich.edu>  |   (X.500 service at umich.edu)   |
| The University of Michigan, Ann Arbor |   Aerospace Engineering & LASC   |

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

Subject: Re: capturing boot messag
From: john.will@satalink.com (John Will)
Date: Mon, 20 Dec 93 11:51:00 -0640

B >| >I remember in some earlier version of UNIX there was a file which was
B >| >created on every boot which contained all of the boot time messages (for
B >| >later reference).
B >| 
B >| Yes, look for "/usr/adm/messages", or use "find".
B >|   find / -name messages -print
B >
B >Thanks, but, that file doesn't exist in /usr/adm, and the one in /usr/bin
B >notifies you if you have e-mail. <grin>

Actually, this may be as simple as an incorrect entry in your rc.local
file.  I also couldn't find this file, because syslogk was set up 
incorrectly in rc.local.  When I fixed that (and created the /usr/log
directory), the kernel file started showing up properly...  You may want
to check and see if syslogk is being started at all...

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

Crossposted-To: comp.os.ms-windows.advocacy
From: eagle@garfield.catt.ncsu.edu (The Great Winged One)
Subject: Re: Windows emulation was Re: Microsoft Invented Inferior Personal C
Date: Mon, 20 Dec 1993 18:34:22 GMT

tjrc1@mbfs.bio.cam.ac.uk (Tim Cutts (Zoology)) writes:
>This is the point you are all missing.  Windows and Unix/X are desgined for
>utterly different purposes.  Windows is designed to be easy to use, and
>primarily for single desktop computers with at most a couple of programs
>running at once for the average user.  Unix/X is designed more to be flexible
>and robust due to the multi-user nature of the beast.

So, as a single user on a single-user system, I should SETTLE for lower
quality just because it's only ONE person's work instead of the work of
MANY people at one time?  Come on!  My work is VERY portant to me!

>Windows does the best it can as far as protection goes given that it is sitting
>on top of DOS, and I don't think it does a bad job.

Well, do what I did: get rid of DOS!  I run OS/2, and I have never had
anything not run.  I keep DOS around simply to run games (all of which,
by the way, run well under OS/2; I just prefer the single-tasking system
that DOS provides, when running games).

>And for those of you saying Unix/X never crashes, just the other day Linux
>locked up completely on me.

It's unfair to say that UNIX crashes simply on the basis that a
pre-release of a single version of it (Linux, that is) does.

Daniel
--
Daniel "eagle" L'Hommedieu / Senior, NCSU CSC Dept / eagle@catt.ncsu.edu
Other stupid UNIX error statements:                             |\|/|
% man: why did you get a divorce?                               |-o-|
man:: Too many arguments.                                       |/|\|

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

From: chris@csvax1.ucc.ie (Chris Higgins - System Administrator)
Subject: Re: Linux in a hospital?
Reply-To: chris@csvax1.ucc.ie
Date: Mon, 20 Dec 1993 21:08:32 GMT

In article <2f51at$eu8@mailgzrz.TU-Berlin.DE>, wpp@marie.physik.tu-berlin.de (Kai Petzke) writes:
>There will be.  The double checks will be done by the stuff of that
>clinic the same way, it used to be.  I agree, that a computer may
>only be introduced, if it increases safety.  I currently assume, it
>does.  For example, the labels printed by the machine will be much
>better readable than those generated with a typewriter.

Only if the labels are not printed _after_ they have been put on the 
blood bags. :-)

>--
>Kai
>wpp@marie.physik.tu-berlin.de
>Advertisement by Microsoft in a well-known German magazine:
>       If you don't like our programmes, then make your own ones.
>However, they expect you to use Microsoft products for this -:)

                                                  Chris.

+ J.C. Higgins,    + Chris@csvax1.ucc.ie      + If you love something, set it +
+ VMS Sys. Admin,  + Chris@cs.ucc.ie (broken) + free. If it doesn't come back +
+ Comp.Sc.Dept.    + Chris@odyssey.ucc.ie     + to you, hunt it down and      +
+ UCC, Ireland     + C.Higgins@bureau.ucc.ie  + KILL it.                      +

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


** 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: Linux-Misc-Request@NEWS-DIGESTS.MIT.EDU

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

    Internet: Linux-Misc@NEWS-DIGESTS.MIT.EDU

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

End of Linux-Misc Digest
******************************
