Subject: Linux-Misc Digest #418
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:     Sun, 12 Dec 93 17:13:09 EST

Linux-Misc Digest #418, Volume #1                Sun, 12 Dec 93 17:13:09 EST

Contents:
  Just a query (Jaime Mantel)
  Re: Debate: Time to Remove SLS From archive sites? (Daniel L Lakeland)
  Re: using Trakker tape backup system under linux (Mark Watson)
  Re: Security (Brandon S. Allbery)
  Fortran compiler? (Erik Werner)
  Re: Call for linux sourc (DAVIN GEORGE)
  how-to to include Slackware? (Marc J. Fraioli)
  Re: Linux Consortium (EurIng Chris Thoday)
  Re: [Q] Scanner software ? (Jonathan Magid)
  Re: Yet another benchmark results.. (Thomas Haywood)
  Re: Yet another benchmark results.. (Zhenya Sorokin)
  Re: Linux Foundation (was Re: Linux Consortium) (David Alan Black)
  Re: The Great Linux Debate(s) (David Alan Black)

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

From: mantel@calvin.fnal.gov (Jaime Mantel)
Subject: Just a query
Date: 12 Dec 1993 19:42:20 GMT

I was wondering what makes motif so "great"?  I use Xt and athena widgets at home
and am quit satisfied with them.  I use motif applications at work and really
don't see any advantage to the motif stuff.  If there is no real advantage to
using motif, why all the noise about programming motif applications and their
copy write.  Just don't use motif.
-- 
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Jaime Mantel                                        +  Sorry, no COOL 
e-mail:   internet :  mantel@calvin.fnal.gov (Unix) +     LOOKING    ,,,,, 
                      mantel@adcalc.fnal.gov (VAX)  +      ASCII      ^ ^
                                                    +      IMAGE    ( O O )
          bitnet   :  mantel@fnalad                 +   just this->    >   
                                                    +                `---'
Phone:    (708) 840-3721, 3194, 4977                ++++++++++++++++++++++++++
Place:    Fermilab, P.O. Box 500, MS 306, Batavia, IL 60510
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++






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

From: dlakelan@iastate.edu (Daniel L Lakeland)
Subject: Re: Debate: Time to Remove SLS From archive sites?
Date: Sun, 12 Dec 1993 19:49:03 GMT

In <2efa4v$gpu@vixen.cso.uiuc.edu> jeske@ux4.cso.uiuc.edu (David Jeske) writes:


>As sad as it is, and as much as I owe alot to SLS for being there and getting
>me started in the beginning, all the points made seem to be on the mark.
>I have had no problems WORSE then 0.99pl12 (which I ended up never running
>because of the problems) and Xfree 2.0 is light years ahead of 1.2. 

>At the very least the SLS 1.03 should have a note attached which reminds the
>user of the date of the materials and the Places to look elsewhere for newer
>material.


>-- 
>David Jeske(N9LCA)/CompEng Student at Univ of Ill at Cham-Urbana/NeXT Programmer
>CoCreator of the GTalk Chat Software System  - online at (708)998-0008
>Mail:  jeske@ux4.cso.uiuc.edu    NeXTMail: jeske@sumter.cso.uiuc.edu

WHAT IS THE MOST UP-TO-DATE DISTRIBUTION???

I want to get ahold of linux, and install it, what should I get? GNU?
slackware? what's the best?


-- 
-Daniel Lakeland

What is the most important word in the english language?
Finger for PGP 2.3 public key. 

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

From: mwa@netcom.com (Mark Watson)
Subject: Re: using Trakker tape backup system under linux
Date: Sun, 12 Dec 1993 18:44:41 GMT

Henry M.K. Yeung (hyeung@barracuda.micro.umn.edu) wrote:

: Is there anyone using trakker in a Linux box? I've the
: drive but I have no idea on how to use it to backup
: a linux partition...

This is a crude approach, but I use a 10 megabyte DOS partition
for my Linux backups: I tar and gzip my home directory, and any
new packages that I load to the DOS partition.  Everytime I do
a DOS tape backup, my important Linux stuff goes on the tape also...

-- Mark Watson


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

From: bsa@kf8nh.wariat.org (Brandon S. Allbery)
Subject: Re: Security
Date: Sun, 12 Dec 1993 20:11:18 GMT

In article <1993Dec12.125316.1938@pe1chl.ampr.org>, rob@pe1chl.ampr.org (Rob Janssen) says:
+---------------
| In <2e53si$hbk@jacobs.jacobs.mn.org> root@jacobs.mn.org (Mike Horwath) writes:
| >this can be accomplished with current AMI bios' by setting a password to
| >get into the bios and YOU set the bios to boot off of C: and ignore A:
| >for booting.  Should be in the Advanced CMOS settings.
| 
| Unfortunately that won't work when you are using a SCSI controller and
| boot from it...
+---------------

Correct; the BIOS doesn't see any (MFM/IDE/ESDI) drives, so it re-enables the
floppy.  By the time the SCSI controller has chained itself into the disk BIOS
interrupt it's far too late, since the BIOS makes its decision based on the
CMOS drive information before running the disk BIOS initialization.

++Brandon
-- 
Brandon S. Allbery         kf8nh@kf8nh.ampr.org          bsa@kf8nh.wariat.org
"MSDOS didn't get as bad as it is overnight -- it took over ten years
of careful development."  ---dmeggins@aix1.uottawa.ca
Do not taunt Happy Fun Coder.   (seen on the Net...)

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

From: werner@acsu.buffalo.edu (Erik Werner)
Subject: Fortran compiler?
Date: Sun, 12 Dec 1993 20:45:09 GMT

A quick question. As an engineering student I have to write many programs in
fortran. I normally use unix at school, or dial in from home to compile my
programs. Now that I have linux running at home, is there a fortran compiler
out there that will let me do this from my pc with out the need to compile it
on the university's machine?

thanks

Erik

-- 
=====================================================================
Erik Werner                            
werner@destiny.eng.buffalo.edu             
=====================================================================

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

From: davin.george@welcom.gen.nz (DAVIN GEORGE)
Subject: Re: Call for linux sourc
Date: Sat, 11 Dec 1993 23:31:00 GMT


Gidday Phillip, I've been calling your bbs in auckland and left a couple of
messages to find out how I can open an account on it. I haven't heard back
anything yet so figured I'd ask you here. Theres a couple of us interested
so any info would be greatly appreciated.

Its either that or join Actrix down here as I really need FTP but would
prefer to go to a Linux Bbs instead of one of the others. Thanks.

                                                Catch Ye Later
                                                 Davin.George



Davin.George@Welcom.Gen.Nz Or FidoNet 3:771/370
Davin.George@f370.n771.z3.fidonet.org

 * WaveRdr 1.0 [NR] * UNREGISTERED EVALUATION COP

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

From: mjf@clark.net (Marc J. Fraioli)
Subject: how-to to include Slackware?
Date: 12 Dec 1993 16:15:59 -0500

Hey folks-

 I'm interested on running Linux, and have been lurking in this group for
some time.  I have read FAQs and how-to lists, and so on.  I am getting
the distinct impression that SLS is now out of favor (understatement).
However, the how-to lists are all geared towards it.  Is the Slackware
installation substantially similar to SLS?  Ie., will the how-to list
still be more or less accurate?

        Thanks for any info.

-- 
Marc Fraioli                           |   The opinions expressed above may
mjf@clark.net                          |   disintegrate if exposed to bright
                                       |   light.

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

From: chris@thoday.demon.co.uk (EurIng Chris Thoday)
Subject: Re: Linux Consortium
Reply-To: chris@thoday.demon.co.uk
Date: Sun, 12 Dec 1993 21:14:31 +0000

In article <2eem5e$ejb@due.uninett.no>
           magnus@ii.uib.no "Magnus Y Alvestad" writes:

>In article <1993Dec12.071148.11487@henson.cc.wwu.edu>
> markline@henson.cc.wwu.edu (Mark Line) writes:
>
>        [a lot]
>
>Mr Line, I appreciate your support, but I would like to ask you to
>SHUT UP! And that goes for the rest of you as well. I'm getting a
>headache. There will be no Linux Consortium. Let the dog die.
>
>-Magnus
>

Does this mean that the idea of a Linux Review Group is dead as well?
I hope not as, it seems to me, there is a very real job to be done.
Consider somebody who buys a Linux CD and they cannot get it to work.
If they have access to the Net they may be able to get help but only
if they are technically competent and probably with considerable expenditure
of time and money. If they bought it directly from the producer they
may be able to get help (possibly at extra cost), otherwise they may be
able to return it and get some of their money back. If these bad
experiences are repeated then Linux, and its developers, will get a
bad name.

Of course, nobody can be expected to guarantee that Linux will work
for every possible hardware combination and offering a refund is a
perfectly reasonable response to this situation. It is also true that
no significant software package is free from bugs (even if it costs a
fortune). However, most of the package should work reasonably well
and the advertising and associated documentation should be honest.
Many suppliers of CD's do not even give the date of compilation.

My bad experiences with a particular CD may not be common to other
users (perhaps with different hardware). If I shout about them on the
Net the criticisms may rub off on other suppliers. If I do nothing then
others having similar problems may not get help. It appears to me that
there are so many help requests that they are no longer being answered
by people who know what they are talking about if at all.

Because I had difficulty in installing Linux from the CD, some of my
problems may result from faulty installation. It is also possible that
I made some mistakes. I have no interest in putting the distributor out
of business but I would like to see some more effective mechanism for
solving problems. This what I hoped the Linux Review Group would do.

The main point is this: developers are only interested in those parts of the
package for which they have some responsibility but users require the parts
to work together.

I mentioned the following example in a previous posting but nobody responded.
Perhaps Matt Welsh, or somebody with similar experience, could tell me
how this problem should be dealt with. So far as I am personally concerned
I can solve the problem for myself but it seems a waste of effort if lots
of other people are solving the same problem.

   a) The binary version of `more' fails with a segementation violation.
      Since this is a very basic command I think that user's should be
      able to expect that it will work.
   b) When recompiled with the source code on the CD it still fails.
   c) The problem is related to the space allowed for loading the
      termcap information.
   d) I know that the termcap file is not satisfactory because it does
      not contain the specification for an 80x28 console and also because
      the `reset' command does not work properly (fails to reinstate
      automatic CR).
   e) A further possibility is that there is a problem with the termcap
      library or the specification of a termcap function has changed since
      `more' was written many years ago.

I had hoped that one of the functions of the Linux Review Group would be
to: collect problem reports related to a distribution, verify the reports,
and pass them on to whoever might be able to produce solutions.

In an earlier posting somebody suggested that the Linux Consortium would
not work "because testing was boring". I suspect that testing of one's
own work is boring because there is a psychological block against finding
fault with one's self. This does not apply to other people's work and,
in any case, I would suggest that the primary testing of a distribution
would be done by those with an interest in making use of it.

In case you think that my `more' example is trivial I can supply details
of a lot of other problems which have taken a considerable amount of
time and effort to deal with. And which I could not have solved without
access to the Net. It should not be assumed that everbody able to read the
news groups is also able to ftp new source packages or can do so at
reasonable cost. Several people have requested that I send them my `lc'
command by e-mail because they cannot ftp it. Obtaining a flopy disk's
worth of software by ftp costs me nearly two hours of long-distance
phone call to London or Warrington.

Finally, I hope there will be a more considered response to real
technical problems rather than the somewhat personal and argumentative
tone of recent postings. That way something good may still come from
the Consortium idea.

--
EurIng Chris Thoday,              chris@thoday.demon.co.uk, Software Engineer
8 Victoria Street, RUGBY,         a Director of the European X User Group Ltd
Warwickshire, CV21 2HN, UK        info@exug.demon.co.uk

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

From: jem@sunSITE.unc.edu (Jonathan Magid)
Subject: Re: [Q] Scanner software ?
Date: 9 Dec 1993 05:34:36 GMT

In article <2e5sgn$fnc@urmel.informatik.rwth-aachen.de>,
Guido Muesch <odiug@messua.informatik.rwth-aachen.de> wrote:
>Stupid question: Is there any scanner software available somewhere?
>And for what scanners?
>I can currently get hold of a marstek M-800 handyscanner (I think that was
>its name).

There is a driver for scanners built on the GS4500 chip-set.. It can be
found at:

<file://sunsite.unc.edu/pub/Linux/kernel/misc-patches/scanner-1.0.tar.gz>

Here is the lsm:


Begin2
Title        = Scanner driver for GS4500 and lookalikes
Version      = 1.0
Desc1        = Provides support for using GS4500 class hand
Desc2        = scanners with Linux.
Author       = Richard Lyons
AuthorEmail  = pclink@qld.npb.telecom.com.au
Maintainer   = Richard Lyons
MaintEmail   = pclink@qld.npb.telecom.com.au
Site1        =
Path1        =
File1        = scanner-1.0.tar.gz
FileSize1    = 4426
Required1    = GS4500 B&W hand scanner or workalike.
CopyPolicy1  = Freely distributable with acknowledgement of author.
Keywords     = scanner GS4500
Comment1     = Suggestions welcome.  Support for other scanners
Comment2     = particularly welcome.
Entered      = 25OCT93
EnteredBy    = Richard Lyons
CheckedEmail = pclink@qld.npb.telecom.com.au
End

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

From: tommy@zikzak.apana.org.au (Thomas Haywood)
Crossposted-To: comp.sys.ibm.pc.hardware,comp.os.vms,relcom.talk,relcom.fido.su.general
Subject: Re: Yet another benchmark results..
Date: 12 Dec 1993 22:40:48 +1100

I actually couldn't compile it.
I got error messaged like this;
unknown symbol _sin in bench.o
unknown symbol _log in bench.o
Plus for the other three maths functions.

I'm using Slackware Linux 1.1.0.
and compiled like this
gcc -O6 -m486 -c bench.c
gcc -O6 -m486 -o test bench.o

My guess is that the maths functions are missing from my standard libraries.
Any other reasons why this would happen?

Any way to fix it?

Thanks.......
-- 
Welcome to my new mail box...........
Tommy Haywood: tommy@zikzak.apana.org.au
2nd Year BCSE, Monash Uni/Clayton, Vic, Aus
Home: Wantirna Sth, Melbourne, Victoria, Australia

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

From: sor@rs6.iaee.tuwien.ac.at (Zhenya Sorokin)
Crossposted-To: comp.sys.ibm.pc.hardware,comp.os.vms,comp.benchmarks,relcom.talk,relcom.fido.su.general
Subject: Re: Yet another benchmark results..
Date: 12 Dec 1993 21:37:16 GMT

Todd Walk (walk@mrcnext.cso.uiuc.edu) wrote:
>viznyuk@mps.ohio-state.edu (Dragon Fly) writes:

>>    Notwithstanding possible critique from alleged
>>computer specialists the insightful observer might note
>>that the "benchmark" code is pretty typical for scientific
>>calculations. Whatever other merits the system might have,
>>if it's dragging its feet on this test it means the system
>>from the point of view of consumer [insightful observer] is
>>a crap. As many insightful observers probably have already
>>noticed, the crap is being limited mainly to two mainstreams:
>>SUN Sparcs and DECs running VMS.

>Well I'm not an "alleged computer specialists", I'm a PhD.
>candidate at UTK, and I'm in agreement with the others that
>say that your benchmark is "crap".

Anyway, it produced correct results.
We compared a FORTRAN program (mostly FFT) on a 486 and on a IBM R6000/350
The speed difference was about 4 times.
It is exactly the same difference as predicted by this test (27 sec/7 sec).
I hope the professor's benchmarks could be so exact.

BTW LaTeX speed difference was only 2 times. Dvips - 4 times.

All this only proves that the best benchmark is the application itself.
And for _some_ groups of sci. calculations this test _is_ correct.
Naturally, the generalization of the author is ridiculous, but ... well,
the author is known in selected newsgroups for such generalizations.

>Inaccuarate benchmarking is easy.
>Accuarate benchmarking is something that the Federal Government
>spends millions of $$$ on for grants to university professors
>who then work for YEARS refining test suites.

>(At UTK here Jack Dongarra does a lot of work on benchmark programs,
>esp. Linpack.  He's one of those million $$$ professors.
>Take a good look a Linpack and then compare it to your little
>code blurb, then if you're still interrested come back
>with a new, more reasonable program.)


--

==============
Zhenya Sorokin
Vienna, Austria


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

From: dblack@pilot.njin.net (David Alan Black)
Subject: Re: Linux Foundation (was Re: Linux Consortium)
Date: 12 Dec 93 21:38:24 GMT

mdw@cs.cornell.edu (Matt Welsh) writes:

>The best solution would be to move Linux towards a stronger, more
>centralized development process. (I'm not suggesting that we should do
>this, but it would be able to implement your stated goals for the
>so-called LC.) Something along the lines of the "Linux Foundation"
>should be formed, spearheaded by the Linux developers themselves. LF
>would be a very close analogue to the Free Software Foundation. It
>would provide Linux with a centralized legal entity to promote and
>take responsibility for the software. All Linux software could be
>copyrighted by the Linux Foundation instead of just by individual
>authors. (However, it should still be possible for individuals to
>copyright software; the FSF permits this.) Low-cost memberships could
>be sold, which would help to support Linux and the LF. Major Linux
>developers would draw from the LF funds... e.g., giving money to the
>LF would be equivalent to donating money to the developers, although
>it would be clearly defined who would get what money and how much
>(instead of in the form of monthly "awards" as previously proposed).
>The LF would be nothing more than the "official" LInux development and
>support organization. Developers, documentors, and distributors would
>work with the LF to produce Linux. 

Anticipating your later question ("What do people think?") - I basically
think that it's always too bad when something which might (and I don't
say this lightly) actually represent something new, chooses to assimilate
itself to things that are not new.  By the latter I don't specifically
mean FSF, but organizations, officialdom, and corporation-inspired
structure in general.  I add in haste that I am not in any sense opposed 
to the accrual of money to people who work on Linux.  I suppose it's just that
I find things quite astonishing at present; they will, of course, change,
as everything does.

To put it in a sort of gut-reaction way:  I installed Linux about a month
ago, and have been extremely active in e-mail correspondence about it,
fairly active on the net.  Let's just say that, during my initial discovery
of Linux, if there had been indication among the introductory materials
that there existed something called the Linux Foundation, to which the
adjective "official" applied and which was a centralizing organization
for development, documentation, and distribution (or in your words, an
organization with which d's, d's, and d's worked to produce Linux), the
whole thing would have had an extremely different "feel" to it.  I'm not
saying I wouldn't have been intrigued at all; it just would have been
different.

>With all of this organization, still, the actual development and
>distribution process for Linux would not need to change. It would
>still be an open-development over the Net, as it is now. In fact,
>technically, not much would change. All LF would provide is an
>official entity to take responsibility for Linux. It would also
>provide a channel for people to donate funds to support the Linux
>development effort. 

I can't think of a greater change than the introduction of an official
entity with responsibility for Linux.  In what sense does that not
constitute change?

>It differs in the proposed "LC" in a number of major ways. First of
>all, the LF *would* be an "official" organization, composed of the
>Linux developers themselves, not some arbitrary third party trying to
>claim responsibility for Linux. All of the "duties" of the proposed LC
>would be handled by people "within" the LF. The LF would provide a
>certain degree of logistics to legal and monetary matters. As far as
>actual development of Linux goes, however, things would remain the
>same. The FSF/GNU people still essentially use an open development
>cycle (for most projects). The primary function of the FSF (as an
>organization alone) is to promote free software, financially, and to 
>provide a legal entity for dealing with such matters. Of course, the
>FSF does much, much, more than this, but you don't need an
>organization to do those other things (namely, develop and distribute
>free software), as we have proved with Linux.

>I'm not saying that we should do this. However, if people want there
>to be some kind of "official" Linux organization, this would be the
>most general and open-ended solution. It would provide a great deal of
>structure to all of these metaissues such as where donations go, who
>has the right to "speak" for the Linux development community, and so
>on. The LF would be modeled on the FSF, in certain ways, and would
>hopefully bring together all of these loose ends under one roof.

But what is exactly IS the Linux development community?  I think that the
creation of an organization which could "speak" for that community would
entail - necessitate - an excessively concrete definition of it, meaning
basically a list of names.  I'm not saying there are not people who are
obviously at the vanguard of the project.  But there are things at stake
here.  Example:  I have, through e-mail, done things like work out the
correct syntax for a patch to the source for an important, much-used
Linux utility - "through e-mail" meaning I had help, I then nailed the
solution home, and I shared it.  That felt cool.  It meant that lots of
people could now compile the program.  I do not feel like a "developer"
in the same sense that the people who wrote the program are its developers,
but I do feel woven into the life cycle of the program.  Does that mean
the LF would "speak for" me?  Of course not.  The only effect of such an
organziation (in this regard) would be to clarify and codify the very 
lines whose unclarity is so delightful.  It would virtually guarantee that
contributions of the kind I just described were understood as marginal, on
the fringes of something bigger.  And maybe in some sense they are.  But at
the moment, it's a matter of degree; giving the stamp of officialdom to one
subset of work would make it a matter of kind.

(This is completely different issue from giving credit to people who do the
actual work - a practice which I am NOT suggesting should be diminished.
But, in reality, I don't think that's a big problem.  I don't think there's
any lack of clarity about authorship and creative contribution in Linux -
at least, I've never felt any.)

>What do people think?

I love the sense of a huge tapestry.  What people have always said about
unix in general seems to me to be played out in the evolution of Linux.
Then again, in terms of its actual implementation in business environments,
I am probably among the least goal-oriented users.  So I can afford to
romanticize it  :-)


David Black
dblack@pilot.njin.net


                "And watch out for flying chandeliers!"

                        - John Steed

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

From: dblack@pilot.njin.net (David Alan Black)
Subject: Re: The Great Linux Debate(s)
Date: 12 Dec 93 21:44:18 GMT

jgifford@thor.fcs.uga.edu (Jim Gifford) writes:

>Due to the number of large flame wars that crop up on the c.o.l.*
>groups every month or two, I propose the following:

>Create a channel on the niksula mailing list called DEBATES.  Then
>when a small or large flame war erupts, people can kindly take it
>there, and anyone interested in the results could subscribe to it.
>That would keep the debate 'public' so that interested parties can
>keep track of it, (as opposed to private via e-mail which suggestion
>is ALWAYS ignored), but still not subject thousands of people not
>interested in the flame war to so much mindless
>drivell^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H err... net.wisdom.  :)

>I am surprised recently by the number of big-time linuxers makeing
>such a big scene (between less than a dozen persons) rather than
>takeing it to e-mail.  Perhaps this suggestion has some merit?

Just a quick vote in the negative.  Nothing is as good as the net for
debate (mailing lists aren't close), and nothing so easy as to avoid
reading postings.  Moreover, I for one have found the "debates" to be
as substantive as anything else.  And among the benefits to having
a .misc group is that it IS a place where such things can be conducted
without cluttering up the .help and other groups.


David Black
dblack@pilot.njin.net


                "I'm sorry, sir, I don't believe you."

                        - Glover

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


** 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
******************************
