Subject: Linux-Misc Digest #570
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, 7 Aug 94 21:13:12 EDT

Linux-Misc Digest #570, Volume #2                 Sun, 7 Aug 94 21:13:12 EDT

Contents:
  Re: Coherent & Linux (Was : A Truly Unbiased Opinion) (Pascal Haible)
  Re: Mild warning about slackware installed users. (David Fox)
  Re: Free Motif GUI and API (clone) for Linux (Jon Fo)
  Patch to run STk-2.1.1 under Linux (David Fox)
  Re: Mild warning about slackware installed users. (Patrick J. Volkerding)
  Re: WABI vs. SoftWindows? (James Vlcek)
  Re: Patch to run STk-2.1.1 under Linux (David Fox)
  Re: Ack!!!  Please HELP ME install on Dell Pentium with EIDE drive (David A. Berson)
  Call-Waiting and a modem, ideas??? (Sergei Naoumov)
  Word Processing Shareware (Keith Scott)
  Re: Linux book(s) (Scott Hanson)
  Re: PPP vs. SLIP vs. PLIP (Ziniu "Michael" Wei)
  Re: Adaptec 1522 at 66mhz? (Christopher M. May)
  Editors for Color Icons? (Christopher M. May)
  Re: console driver does not reset char table with "stty sane" (Kurt M. Hockenbury)
  Re: dip with a pulse phone line (Uri Blumenthal)

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

From: haible@izfm.uni-stuttgart.de (Pascal Haible)
Crossposted-To: comp.os.coherent
Subject: Re: Coherent & Linux (Was : A Truly Unbiased Opinion)
Date: 4 Aug 1994 17:25:21 GMT

In article <9408040121.39@rmkhome.com>, Rick Kelly <rmk@rmkhome.com> wrote:
>Robert Sanders (rsanders@mindspring.com) wrote:
>: Contrary to certain FUD-like allegations seen 'round here, Linux
>: developers aren't selfish or unresponsive to support requests.
[examples deleted]
>: Stephen Tweedie personally saw the filesystem race condition that they
>: spent many, many long hours a few months ago tracking and killing.
[example deleted]
>Drew Eckhardt wrote up a very good assessment of the differences
>between Coherent and Linux from a professional view.  Guess you didn't
>read it.
I don't see the connection between way you say here and what you are
quoting from Robert Sanders, sorry.

I have already seen too many 'X is better than Y' posts, but

>There are no race condition problems in the Coherent fs,

is not true, acording to Udo:

<Quote from Udo Munk>
From udo@umunk.GUN.de Thu Jun 16 19:10:47 MET DST 1994
Article: 15850 of comp.os.coherent
From: udo@umunk.GUN.de (Udo Munk)
Newsgroups: comp.os.coherent
References: <9406071633.AA26830@kc4sa.marine.geol.scarolina.edu>

<deleted>
When you create lots of files on a filesystem and it doesn't have enough
free space, it might happen that there are some inconsistent inodes,
because the needed blocks for the files weren't available. When fsck
checks the filesystem, it will report this inodes, usually one clears
them with clri and then fsck the filesystem again to correct the
superblock.
<deleted>
Udo Munk  udo@mwc.com or udo@umunk.GUN.de, CIS: 100021,2515
<End quote>

Sorry to warm up this old thing (I wouldn't if you were not saying
that the COHERENT fs code if free of race conditions), but I call this
***** a race condition in the fs CODE *****.

As we are talking about COHERENT vs Linux:
There were race conditions in the code for the filesystems Linux supports.
I think they are gone now, but I can't say for shure.
Nor can you about the COHERENT fs code.

> but of course
>they have to come up with a new one. (and they are :-) )
                                     Will it be Linux compatible? ;-)
                          [This is a joke. If you don't like it, ignore it]

>This particular box that i'm typing on is running Coherent.  If it is
>doing exactly the job that it is supposed to do, with the required amount
>of performance, then why should I trash it?

Nobody is trying to convince you to trash COHERENT. If you are happy
with it, and you apparently are, great.

But for me it's hard not to contradict when I read things in this group
that are misleading, not true anymore, not true, or completely wrong
wrt Linux.

>Do people still buy systems that don't have hardware floating point?
Shure. 486SX, e.g.

>Rick Kelly  rmk@rmkhome.com  rmk@bedford.progress.com

Pascal
-- 
cand. mach. Pascal Haible haible@IZFM.Uni-Stuttgart.DE

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

From: fox@graphics.cs.nyu.edu (David Fox)
Subject: Re: Mild warning about slackware installed users.
Date: 07 Aug 1994 14:53:23 GMT

In article <glam0001.776200539@gold> glam0001@gold.tc.umn.edu (Robert P Glamm) writes:

] Of course, if you monitor the installation & do it package by package,
] you can skip adding the fake users satan, gonzo, & snake, and avoid
] all the trouble in the first place... ;)

While I do this myself, there is nothing in the installation procedure
to indicate that simply allowing slackware to install everything might
be a bad idea.
--
David Fox                                               xoF divaD
NYU Media Research Lab                     baL hcraeseR aideM UYN

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

From: jkf@shell.portal.com (Jon Fo)
Crossposted-To: comp.windows.x.motif
Subject: Re: Free Motif GUI and API (clone) for Linux
Date: 7 Aug 1994 22:18:56 GMT

jaakola@cc.helsinki.fi wrote:
: Slackware 1.something installation docs mention the OI package, which is
: some sort of Motif clone. However, when I ftp'd the files in June the OI
: files were missing and now the Slackware 2.0 docs don't even mention it!
: What has happened to OI? Isn't it freely available any more?

: My aim is to have a Motif look-and-feel environment. And after a few
: weeks I would like to try developing apps to the Motif API. Is it
: possible to achieve these goals free?
: --
: Juhani Jaakola, jaakola@cc.helsinki.fi
Try getting fvwm.  It has a "Motif" look-and-feel mode.  It's not mwm,
and some say it's better.  There are some important differences, but I
use it as my main window manager.  It's based on twm more or less.  I 
can't remember a good ftp site to get it from.  If I find the ftp site, 
I'll post a followup.  Or hopefully someone else closer to your site knows 
of one...  Actually I think it may already be in one of those Linux 
software collections that are popping up all over the place.

Hope this helps...

======================================================================
Jon Fo                                  "Mmm...Soylent Green..."
jkf@shell.portal.com                    - Homer Simpson

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

From: fox@graphics.cs.nyu.edu (David Fox)
Crossposted-To: comp.lang.tcl,comp.lang.scheme
Subject: Patch to run STk-2.1.1 under Linux
Date: 07 Aug 1994 21:01:10 GMT

This patch prevents STk-2.1.1 from crashing upon startup,
at least for me it does.  I will place a working version
of the installed files (with --prefix=/home/Installed)
on cs.nyu.edu in pub/local/fox/STk-2.1.1-linux.tgz.
--
David Fox                                               xoF divaD
NYU Media Research Lab                     baL hcraeseR aideM UYN

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

From: gonzo@magnet.mednet.net (Patrick J. Volkerding)
Subject: Re: Mild warning about slackware installed users.
Date: 7 Aug 1994 22:18:17 GMT

In article <FOX.94Aug7105323@first.cs.nyu.edu>,
David Fox <fox@graphics.cs.nyu.edu> wrote:
>In article <glam0001.776200539@gold> glam0001@gold.tc.umn.edu
>(Robert P Glamm) writes:
>
>] Of course, if you monitor the installation & do it package by package,
>] you can skip adding the fake users satan, gonzo, & snake, and avoid
>] all the trouble in the first place... ;)
>
>While I do this myself, there is nothing in the installation procedure
>to indicate that simply allowing slackware to install everything might
>be a bad idea.

Untrue. If you tell it to install everything, it puts up a dialog box 
with the following warning:

  WARNING: You should be advised that installing *everything* is
  often a bad plan. If you're just putting the base system in place
  and planning to install things like X and Emacs a little more
  selectively from your hard drive later, then there's no problem.
  Otherwise, if you try to install the whole thing without prompts
  you will not get as optimal an installation as you might like.
  Are you sure you want to install without prompts?

Nice try, though. ;^)

Pat

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

Crossposted-To: comp.sys.hp,comp.sys.hp.apps,comp.sys.sun.apps,comp.windows.x,comp.windows.x.apps,comp.unix.unixware
From: vlcek@epimbe.com (James Vlcek)
Subject: Re: WABI vs. SoftWindows?
Date: Sun, 7 Aug 1994 22:16:42 GMT

In article <320ris$g7n@kubds1.kub.nl> paai@kub.nl (J.J. Paijmans) writes:
>What it all
>seems to boil down to is that you (=Jim) think that Linux isn't going
>to make it because it is "...yet another Unix flavor..." But if Linux is
>yet another UF, then why is not UnixWare?

UnixWare, being based upon SVR4.2, is at least the direct descendant
of the original UNIX.

But you are incorrect in assuming that I think that Linux will not
"make it" because it is yet another UNIX flavor.  While we need
another UNIX flavor about as much as the Sahara needs drought and
sand, Linux is not going to hold the line against the Microsoft
onslaught for other reasons - which I have listed in other postings in
this thread.

>I admit I have no experience with Unixware - send me a copy and I
>try and review it in the magazine I am writing for.

Buy it yourself!  It's only $166 from Information Foundation.

Cripes, all these Linuxers think everything should be given to them.
I've gotten several emails already, demanding that I buy copies of
UnixWare for Linuxers.  "If you think it's so great, why don't you buy
me a copy?"  That response didn't do much to change my basic
perception that the Linuxoids don't have a very sound understanding of
the free market system.

>So if I have to bet my shirt, I will bet on enthousiasm & creativity and
>not on commercialism like you seem to do.

There are people of equal enthusiasm and creativity providing
competitive products on Microsoft OS platforms.  Enthusiasm and
creativity are not enough; one needs a market advantage as well.  I
contend that Linux will never provide this advantage.

What Linux _does_ provide, on the other hand, is an easy route for
interested users to become introduced to UNIX.  I received one email
from a guy in high school who has installed Linux on a 386SX/16 with
four meg or memory, and who is now experiencing the sheer fun on
learning what UNIX can do for him.  This is where Linux can't be beat,
and I think/hope that UnixWare can benefit if Novell comes to view
Linux as a doorway to UnixWare.

But when one starts considering platforms upon which to base a
commercial product, Linux simply falls far short.  The lack of Windows
support alone is a killer - Windows programs can be _extremely_ useful
as front-ends to UNIX apps on a platform which can support them.  I've
been admonished by many Linuxoids "they've only just started working
on WINE - give them time!"  What the hell - I'm going to tell my
customers that?  Get real!

Jim Vlcek
vlcek@epimbe.com

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

From: fox@graphics.cs.nyu.edu (David Fox)
Crossposted-To: comp.lang.tcl,comp.lang.scheme
Subject: Re: Patch to run STk-2.1.1 under Linux
Date: 07 Aug 1994 21:10:14 GMT

In article <FOX.94Aug7170110@first.cs.nyu.edu> fox@graphics.cs.nyu.edu (David Fox) writes:

] This patch prevents STk-2.1.1 from crashing upon startup,
] at least for me it does.  I will place a working version
] of the installed files (with --prefix=/home/Installed)
] on cs.nyu.edu in pub/local/fox/STk-2.1.1-linux.tgz.

Oops.  Forgot to include the patch.

diff -u STk-2.1.1/Src/unix.c~ STk-2.1.1/Src/unix.c
--- STk-2.1.1/Src/unix.c~       Tue Jul 26 10:53:27 1994
+++ STk-2.1.1/Src/unix.c        Sun Aug  7 16:31:49 1994
@@ -85,8 +85,12 @@
 SCM internal_expand_file_name(char *s)
 {
   SCM z;
-  char abs[2 * MAX_PATH_LENGTH];  
+#ifdef linux
+  char abs[1024];  
+#else
+  char abs[2 * PATH_MAX];
     /* Warning: absolute makes no control about path overflow. Hence the "2 *" */
+#endif
 
 #ifdef USE_TK
   Tcl_DString buffer;
--
David Fox                                               xoF divaD
NYU Media Research Lab                     baL hcraeseR aideM UYN

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

From: berson@cs.pitt.edu (David A. Berson)
Crossposted-To: comp.os.linux.help
Subject: Re: Ack!!!  Please HELP ME install on Dell Pentium with EIDE drive
Date: 7 Aug 1994 23:08:00 GMT

me (silly@wrath.ugcs.caltech.edu) wrote:
:> I have a Dell Dimensions (XPS) P-90 with a 1 gigabyte hard drive.
:> The HD itself is a WD Caviar 31k,

I have the same system and hard drive.  I had no problems using Mark
Lord's EIDE patch, now available in >= 1.1.37.  You just have to have
a way to compile the kernel.  I borrowed the hard drive from my wife's
systems, a little old Connor 120MB for this.

David A. Berson.
berson@cs.pitt.edu

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

From: naoumov@PHYSICS.UNC.EDU (Sergei Naoumov)
Crossposted-To: comp.os.linux.help
Subject: Call-Waiting and a modem, ideas???
Date: 7 Aug 1994 23:11:19 GMT


Hey, folks!
Today I was watching how my line behaves if I work over a modem
and someone calls me. The funny things start happening if a line
(in my case) have a "call-waiting" feature. If someone calls, a
telephone station does not return a busy signal but rather gives
me a ring. during this incoming call. I'm working with a line,
the former hangs until someone (who tries to get through) does not
give up and hangs up.

So, my question is the following: is there any way to catch up this
event from a modem, analyze it somehow and tell that someone is
actually calling. Then I would put my modem session "on hold", talk
to a person and go back. The idea is a little bit crazy and I feel
there should be some difficulties about a hardware but in principle...
What do you think?

        Sergei

-- 
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+   Sergei O. Naoumov  serge@envy.astro.unc.edu  tel: (919)962-3998   +
+Department of Physics & Astronomy, UNC-CH, Chapel Hill, NC 27599, USA+
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


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

From: scottk@glamslam.rtp.dg.com (Keith Scott)
Subject:  Word Processing Shareware
Date: Thu, 4 Aug 94 18:06:35 GMT

Does anyone know of any cheap Word Processing software that I can get
to run on top of linux?

Thanks
Keith Scott

-- 
/////////////////////////////////////////////////////////////////////////////
///                         |  The Quote for the Week:                    ///
/// Keith L. Scott          |  "Emancipate yourself from innerslavery.    ///
/// scottk@dg-rtp.dg.com    |   None, but ourselves can free our minds."  ///
/// (919)248-6269           |              -Bob Marley                    ///
///                         |                                             ///
/////////////////////////////////////////////////////////////////////////////

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

Date: 07 Aug 1994 17:56:00 +0100
From: scott@kirby.isys.net (Scott Hanson)
Subject: Re: Linux book(s)

In article <1994Aug4.141102.1809@cs.cornell.edu> "Re: Linux book(s)"
mdw@cs.cornell.edu (Matt Welsh) writes:
> In article <CtrtKn.4rJH@austin.ibm.com> michaelb@hobbie.bocaraton.ibm.com
> >(Michael Rogero Brown (Sys Admin)) writes: I recently got the latest
> >Springer-Verlag catalog (S-V is a large publisher of science books,
> >including computer science).  In it they list a new book that is to be
> >published soon (if not already):
> >Linux- from PC to Workstation  by Thomas Uhl and Stefan Strobel
> >(238P, softcover, $39 tentitive, ISBN 0-387-58077-8)

I think I saw the English translation of this in the bookstore yesterday.  
So it might already be out.

In article <newcombe.487.013E5BB0@aa.csc.peachnet.edu> "Re: Linux book(s)"
newcombe@aa.csc.peachnet.edu (Dan Newcombe) writes:
> I will be happy when I can go into a bookstore downtown Atlanta and get a
> book on Linux.  I'd probably get it whether I'd need it or not cause I'd
> think it's so neat.  :)
>
>       -Dan, who's happy he can no get Linux Journal in Atlanta.

In Hamburg, you can already do this... of course, the books are mostly in  
German. :) The self-published "Linux Anwenderhandbuch" (Linux User's  
Handbook) (LunetIX Softfair, 512p, DM 49.00, ISBN 3-929764-02-4), which is  
*not* simply a collection of FAQs and HOWTOs, is already in the 3rd  
edition. The JF Lehmann's bookstore sells the Slackware distribution on  
diskette as well as a number of Linux CDs (in English and German) and the  
August issue of The Linux Journal .

I visited the US last month, and though I didn't look very hard, I was  
disappointed to see nothing about Linux in any of the bookstores I  
visited.
--
** Scott Hanson, 21271 Asendorf, Germany **  scott@kirby.isys.net  **
## CrossPoint v3.02 ##

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

From: ziniuwei@cs.Buffalo.EDU (Ziniu "Michael" Wei)
Subject: Re: PPP vs. SLIP vs. PLIP
Date: Thu, 4 Aug 1994 14:08:35 GMT

Jeff Smith (ind00261@pegasus.cc.ucf.edu) wrote:
> Adrian Mancini (amancini@bmerh263.bnr.ca) wrote:

> : I'm wondering what the pros and contras are of those 4 network protocols,
> : assuming that the two computers I like to connect are next to each other.
> :  
> I only see three...Anyway are you speaking about connecting to a Internet 
> provider or connecting computers in a local  network without ethernet?  
> At my University I use a PPP connection to put my computer on the net.  
> From my understanding, ppp allows correction of packets between your box 
> and your provider,  wheras SLIP is corrected at the host.  This taken 
> from the NET-2.HOWTO.

> I would suggest taking a look at the the howto....It can be found, of 
> course, at sunsite.unc.edu /pub/Linux/docs/howto or something similar.


PPP does error detection (not clear if it does error *correction*),
while SLIP doesn't care about any errors.  As we know we have error
correction on link level V.42, error detection on IP, and error
*correction* on transport layer TCP.  Whether we need error correction
on every layer is questionable.  This is kind of a religious argument
between network experts (here i mean those protocol designers).  Do we
need to ensure reliable tramsmission on low level such as network
layer (the way X.25 works) or link level, leaving a simple transport 
protocol, or do we need a unreliable network layer and having a strong 
transport protocol to do the rest (like TCP/IP)?


--
Ziniu Wei               CEDAR, SUNY at Buffalo       ziniuwei@cs.buffalo.edu
~{@O3LPrT1S@T62;K@#,K{CGV;JGLx5=PB5D5XV7~}
Old programmers never die, they just branch to a new address.

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

From: cmay@titan.ucs.umass.edu (Christopher M. May)
Subject: Re: Adaptec 1522 at 66mhz?
Date: 4 Aug 1994 18:11:26 GMT

Nancy Robison (robison@cliomatups.fr) wrote:
: I'm having a problem running Linux on my 486DX/2 66mhz at 66mhz.  There
: is an Adaptec 1522 clone on board and it keeps giving me a

: "AHA1522x: more data than expected"  error.

: I can run it out of turbo mode but I have to crawl at 16mhz.  Anyone have
: any ideas??

Well, you should check your busclock setting in the Advanced CMOS setup
The standard ISA bus speed is about 8Mhz.  Some cards will run
much higher than that.  The 1522 is the bottom of the line, though,
and probably isn't tolerant of much higher.  I could be wrong.

The bus clock is usually given in divisions of the external CPU
speed, 33Mhz in your case. Busclock/4 yields 8.25 Mhz.

--

-Chris May, Computer Science, University of MA, Amherst
-       Technical Assistant, P.C. Maintenance Lab


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

From: cmay@titan.ucs.umass.edu (Christopher M. May)
Subject: Editors for Color Icons?
Date: 4 Aug 1994 18:16:52 GMT

Hi, I've got the color icons going in fvwm by correcting the 
path to the colorpixmaps (.xpm files) in the .fvwmrc 
(copied from the /var/X11/lib/X11/fvwm/sample_configs/system.fvwmrc)

Is there a color pixmap editor available?  Does this come with
the full xpm package?  Is this already in slackware?

Any help is appreciated.
--

-Chris May, Computer Science, University of MA, Amherst
-       Technical Assistant, P.C. Maintenance Lab


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

From: kmh@linux.stevens-tech.edu (Kurt M. Hockenbury)
Subject: Re: console driver does not reset char table with "stty sane"
Date: Thu, 4 Aug 1994 18:25:32 GMT

Tim Smith (tzs@u.washington.edu) wrote:
: Is there any command that will *completely* reset a virtual console
: back to the state it comes up in?

Try the following, it doesn't do too bad:
stty sane; echo ^{c
(That's <ESC>c.  To enter it under bash, type "echo ^v<ESC>c").

        -Kurt Hockenbury

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

From: uri@valhalla.watson.ibm.com (Uri Blumenthal)
Subject: Re: dip with a pulse phone line
Date: 8 Aug 1994 00:16:23 GMT
Reply-To: uri@watson.ibm.com

In the next release of DIP-uri this will return to
the old way: DIP will know only "ATD" and it will
be up to you to prefix the actual number with 
either "T" or "P"...
--
Regards,
Uri.            uri@watson.ibm.com     N2RIU
============
<Disclaimer>

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


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