File: issue09.html

package info (click to toggle)
lg-issue09 2-3
links: PTS
area: main
in suites: hamm, slink
size: 304 kB
ctags: 35
sloc: makefile: 30; sh: 3
file content (2033 lines) | stat: -rw-r--r-- 75,937 bytes
parent folder | download | duplicates (3)
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
 <HTML>
<HEAD>
<TITLE> Linux Gazette Table of Contents #9</TITLE>
</HEAD>

<BODY>
<H1><IMG SRC="../gx/banner.gif" ALT="Linux Gazette"></H1>
<A NAME="#top"><H1>Table of Contents Issue #9</H1></A>

<H5>Copyright (c) 1996 Specialized Systems Consultants, Inc. <BR>
For information regarding copying and distribution of this material see the
<A HREF="../ssc.copying.html">COPYING</A> document.</H5>

<P> <HR> <P>
<UL>
<LI><A HREF="../lg_frontpage.html">The Front Page</A> 
<LI><A HREF="#mail">The MailBag</A> 
<LI><A HREF="#tips">More 2 Cent Tips</A>
<ul>
<li><a HREF="#emacs">Emacs Control M Trick</a>
<li><a HREF="#xterm">XTerm Title Trick 2</a>
<li><a HREF="#vi">VI Trick -- Commenting Code</a>
<li><a HREF="#sendmail">Masquerading with SendMail</a>
<li><a HREF="#upgrade">Linux Upgrade</a>
</ul>
<LI><A HREF="#news">News Bytes</A>  
<LI><A HREF="#binstats">Binstats: Finding Unusable Binaries, by Larry Ayers</A>  
<LI><A HREF="#newsserver">The Easy Way to Set Up a Local News Server by Christophe Blaess</A> 
<LI><A HREF="#filerunner">FileRunner: A New Tk/Tcl File Manager by Larry Ayers</A> 
<LI><A HREF="#staroffice">Getting Up and Running on StarOffice 3.1, by Dwight W. Johnson</A> 
<LI><A HREF="#yodl">YODL: A New, Easy-To-Use Text Formatting Language by Larry Ayers</A> 
<LI><A HREF="#backpage">The Back Page</A> 
</UL>
 
<HR>

<H4>Got any <I>great</I> ideas for improvements!  Send your
<A HREF="mailto:gazette@ssc.com">comments, criticisms, suggestions and ideas.</H4>
</a><P><hr><p>

<P>
This page written and maintained by the Editor of <I>Linux Gazette</I>,
<A HREF="mailto: gazette@ssc.com"> gazette@ssc.com</A>
<P> 

<hr>

<H2><A NAME="mail"><IMG ALIGN=MIDDLE SRC="../gx/mail.gif">The MailBag!</A></H2>

<P><HR><P>

<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Tue, 06 Aug 1996 08:45:27 -0500<br>
<b>From: "Andrew R. Cook" andy@anchtk.chm.anl.gov</b><br>
To: fiskjm@ctrvax.Vanderbilt.Edu<br>
<b>Subject: Linux Gazette comment</b><br>
<P> 
Hi John,
<P> 
Thanks for taking the time for putting together the Gazette.  I hope
you have some time left over for doing work for your employer !!!
I trust/hope that Phil Hughes will be able to keep the Gazette going ...
I've certainly enjoyed the Gazette over the last year.  One comment
though.  In the most recent issue (#8) the format has changed so that
the Gazette is no longer (easily) printable, but is split among many
pages.  I know this is supposed to be an e-zine, but I liked the old
format better :-(.  I'd hazard to guess that I'm not alone in this, but
probably in the minority.....  Any chance of putting an organized single
postscript file on the ftp site? That way those of us loosers who like
to read from paper at our leisure (and don't have laptops, sigh..) can
still easily print out a copy of the Gazette.  This may understandably
not fly with Phil, but I thought I'd suggest it and see what happens
anyway!
<P> 
Thanks again for all your work!
<P> 
-Andy Cook<br>
andy@anchtk.chm.anl.gov<br>
<P>
<BLOCKQUOTE>
<I>(Yes, you are in the minority, and I must say it is much easier for me if
the Gazette is in a multi-part format. As a result, it's very likely to
stay in multi-part format. However, you should note that it can be
downloaded as one file from the SSC ftp site -- <A HREF="ftp://ftp.ssc.com/lg/">
ftp.ssc.com/lg/</A>. So you
could download it, convert it to postscript, print it and read at your
leisure. We're not going to do all the work for you -- that would take all
the fun out of it. :-) --Editor)</I>
</BLOCKQUOTE>

<P><HR><P>

<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Fri, 16 Aug 1996 20:44:20 -0500 (CDT)<br>
<B>From: Larry Ayers layers@vax2.rain.gen.mo.us<br></B>
To: gazette@ssc.com<br>
<B>Subject: A Few Choice Gleanings From the FTP Sites</B>
<P> 
This month I've found some really nice utilities and programs in the
incoming directories of the Linux archive sites.  If any of you LG readers
find (or have written) a program you're excited about, and don't have the time
or inclination to write a piece for the Gazette, drop me a line telling me
what you like about it and its location.  I'll include a piece on it in next
month's issue.

<ul>
  <li><a href="#binstats">Binstats: Finding Unusable Binaries</a>
  <li><a href="#filerunner">FileRunner: A New Tk/Tcl File Manager</a>
  <li><a href="#yodl">YODL: A New, Easy-To-Use Text Formatting Language</a>
</ul>

<P>
Larry
<P>
<BLOCKQUOTE>
<I>(Larry has done a great job as usual on letting us know about new 
products and releases. Thanks, Larry, for your contributions. --Editor)</I>
</BLOCKQUOTE>
<P><HR><P>

<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Tue, 20 Aug 1996 14:27:34 -0700 (PDT)<br>
<b>From: hans@mlsoft.com (Hans D. Swildens)</b><br>
To: gazette@ssc.com<br>
<b>Subject: Microline Software Free LINUX GUI Version<br></b>
<P> 
I don't know if you are aware, but we have a free LINUX advanced GUI
toolkit (based on the Microline Widget Library for Motif) on our ftp
site www.mlsoft.com.  The MWL was used to build Netscape Navigator for
UNIX and is used for mission critial applications by AT&amp;T, 3Com, Merrill
Lynch, Canon, Netscape, Sun, SGI, etc.
<P> 
Please download our free LINUX version.  We would like to be reviewed or
mentioned in the Linux Journal to get the word out.  Since we are giving
it away for free, we can not pay for an ad, etc.  We welcome your
feedback and hope that the LINUX community enjoys the free download.
<P> 
Hans Swildens<br>
Microline Software<br>
hans@mlsoft.com<br>
www.mlsoft.com<br>
<P> 
<BLOCKQUOTE>
<I>(Okay, here's your chance to see your name in print. Let me know that you
have downloaded the software and want to do the review. I'll see that it
gets in both </i>Linux Gazette<i> and </i>Linux Journal<i>. First come,
first serve. --Editor)</I>
</BLOCKQUOTE>
<P><HR><P>

<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Tue, 27 Aug 1996 12:25:02 -0400<br>
<b>From: MikKass@aol.com</b><br>
To: gazette@ssc.com<br>
<b>Subject: MindQ Publishing Inc.'s Intro to Programming Java Applets</b><br>
<P> 
Dear Editor:
<P> 
"I'm a programmer and I want to spend a few hours getting acquainted with
Java applet programming. Is there a faster way to learn than books?"
"I'm not a programmer, but I want to learn about Java without wading through
Dweeb books. Is there a way?"
"My programmers are learning Java and I want to keep up without being
overwhelmed. How do I do that?"
<P> 
These are comments we overheard at the Web Interactive Show in New York this
month. The point? It's not just hard-core techies who want to learn about
Java--or have time to read the books. So where do the rest of us go?
<P> 
MindQ Publishing Inc.'s "Intro. to Programming Java Applets" is a multimedia
CD-ROM tutorial designed to teach all of the above--using animation, audio,
video and hypertext. Experienced programmers can use the hyperlinked table of
contents to find exactly what they need. Newbies can jump on any one of five
tours. People inbetween can use a combo. All of them can use the Java
Developers Kit, or the Launch button to go right to other Java-related
programs or websites (when the Launch capability is activated). The program
retails for $49.95, but I can provide you with an eval copy. Please email me:
MikKass@aol.com, or call me (203)323-4166. Thanks. I look forward to hearing
from you.
<P> 
Cordially,<br>
Michael Kassin 
<P><HR><P>
The following letters are from authors who will be having articles
in the <i>Linux Gazette</i> soon.
<P> <hr><P>

<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Sat, 17 Aug 1996 19:07:19 -0400<br>
<b>From: Randy Appleton randy@dcs.uky.edu</b><br>
To: gazette@ssc.com
Subject: Re: WANTED: Linux Gazette Needs Writers
<P> 
I might be the sort of person you are looking for.  I've been hacking
on the kernel since around version 0.8.  I'm a new professor looking
to get my name published.  I'm teaching a class in System
Administration using Linux, and I've written before.  Does this sound
like the qualifications for writers for the Linux Journal?  How do I
get an article in the Linux Journal?
<P> 
-Thanks<br>
-Randy
<P> 
<BLOCKQUOTE>
<I>(Sounds qualified to me. See the <a href="../authors.html">Author</a>
section on the Front Page for more information. --Editor)</I>
</BLOCKQUOTE>
<P> <hr><P>
<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Sun, 18 Aug 1996 21:46:00 -0500 (CDT)<br>
<b>From: "jwhyche" jwhyche@scott.net</b><br>
To: gazette@ssc.com<br>
<b>Subject: Writers Needed</b><br>
<P> 
I read on Usenet where you are looking for writers for Linux Gazette. 
If you would send me some information on what is required.
<P> 
Thank you,<br>
<P> 
<BLOCKQUOTE>
<I>(Again, see the information found in the <a 
href="../authors.html">Author</a> section on the Front Page. --Editor)</I>
</BLOCKQUOTE>
<P><hr><P>
<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Thu, 29 Aug 1996 13:37:12 +0200 (MET DST)<br>
<b>From: Olof Svensson d95olofs@dtek.chalmers.se</b><br>
To: gazette@ssc.com<br>
<b>Subject: Re: WANTED: Linux Gazette Needs Writers</b><br>
<P> 
Hey.
<P> 
I would like to write some small articles for the Gazette. Like some novice
articles that were in the early articles in LJ. Or maybe it is interesting
to review some new products or small articles on how to set up different
network features.
<P> 
Yours sincerely, Olof S (and of course I will spellcheck my articles)
<P> 
<BLOCKQUOTE>
<I>(Spellchecking is nice -- also formatting them in HTML. --Editor)</I>
</BLOCKQUOTE>
<P> 
<hr><P>
<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Sat, 17 Aug 1996 17:15:41 -0500<br>
<b>From: Morrissey moz@hti.net</b><br>
To: gazette@ssc.com<br>
<b>Subject: Re: tips & tricks</b><br>
<P> 
I would like to contribute to the Linux Gazette.  I have just put a link to
the Linux Gazette because of the tips and tricks articles!
<pre>
----------------------------------------------------
"I wear blue on the outside, 'cause blue  |Richard
is how I feel in the inside." --morrissey.|N=FA=F1ez
 - - - - - - - - - - - - - - - - - - - - - - - - - -
http://ww.hti.net/~moz/moz.htm ;personal page
http://ww.hti.net/~moz/	       ;business page
</pre>
<P> 
<BLOCKQUOTE>
<I>(Thanks for the link. Mr. Nunoz's home page is pretty cool. --Editor)</I>
</BLOCKQUOTE>
<P> 
<hr><P>
<IMG ALIGN=BOTTOM SRC="../gx/letter.gif">
Date: Sat, 17 Aug 1996 14:06:54 +0200<br>
<b>From: Manuel Soriano dpsys10@dapsys.ch</b><br>
To: gazette@ssc.com<br>
<b>Subject: Ideas for Linux Gazette</b><br>
<P> 
Hello Marjorie
<P> 
First, excuse me, my english is not very fluent.
<P> 
I have several articles, covering general topics, of the intallation and
configuration of Linux, test of soft, etc... but all in spanish.
<P> 
My idea is to make a new section on Linux Gazette in others languages
than english.
<P> 
This will put LG in a new dimension and, why not, LJ :-)
<P> 
What do you think about ?
<P> 
Bye<br>
Manu
<PRE>
                \|/         dpsys10@dapsys.ch
                O-O         manu@ctv.es
    *****---oOo-(_)-oOo---**********************************************
    * Manuel Soriano             * El Perello/Valencia/Spain *
</PRE>
<P> 
<BLOCKQUOTE>
<I>(Why not, indeed. Let's try it. An article from Manuel will appear
in Issue 10. --Editor)</I>
</BLOCKQUOTE>

<P> <hr> 
<P>
This page written and maintained by the Editor of <I>Linux Gazette</I>,
<A HREF="mailto: gazette@ssc.com"> gazette@ssc.com</A>
<P> 

<hr>

<H2><A NAME="tips"><IMG ALIGN=MIDDLE ALT="" SRC="../gx/twocent.gif">More 2&#162; Tips!
</A></H2>

<p><hr><p>
<H3>Contents:</H3>
<ul>
<li><a HREF="./lg_tips09.html#emacs">Emacs Control M Trick</a>
<li><a HREF="./lg_tips09.html#xterm">XTerm Title Trick 2</a>
<li><a HREF="./lg_tips09.html#vi">VI Trick -- Commenting Code</a>
<li><a HREF="./lg_tips09.html#sendmail">Masquerading with SendMail</a>
<li><a HREF="./lg_tips09.html#upgrade">Linux Upgrade</a>
</ul>

<P>
<hr>
<!-- =====================================================================  -->
<H3><IMG ALIGN=BOTTOM ALT="" SRC="../gx/letter.gif"><a name="emacs">Emacs Control M Trick</a></H3>

<P> 
Date: Fri, 09 Aug 1996 20:15:45 -0500<br>
From: David Ishee <ishee@erc.msstate.edu><br>
To: fiskjm@ctrvax.Vanderbilt.Edu<br>
Subject: $0.02 tip for removing Control M in emacs<br>
<P> 
After looking at issue 8 about how to remove the pesky Control-M
character in vi (vim or whatever) I had to also tell how it could be
done in emacs. Say you loaded up a file that has the ^M all over the
place. No problem, say M-x replace-string RET C-q C-M RET RET and you're
done. Emacs rulz!!
<P> 
David
<P>

<hr>
<!-- =====================================================================  -->
<P>
<H3><IMG ALIGN=BOTTOM ALT="" SRC="../gx/letter.gif"><a name="xterm">XTerm Title Trick 2</a></H3>

<P>
Date: Tue, 13 Aug 1996 20:45:39 +0100<br>
From: Caolan McNamara <9312811@ul.iea><br>
To: fiskjm@ctrvax.Vanderbilt.Edu<br>
Subject: XTerm title tricks in Linux Gazette #6.<br>
<P> 
Just scanning through back issues and came across the tip to keep the hostname
of the machine your currently logged into in your xterm titlebar (and
wanted to add my 2 cents).
<P>
I work in a room of 30 linux boxes, so to keep track of which one I'm logged 
into despite logging in from one to another and logging out, I put 
<pre>
alias precmd 'echo -n "\033]2;"`hostname`"\007"' 
</pre>
(course anything could be put here)
<br> 
in my .cshrc. Keeps my titlebar updated at all times. As a precmd its run after
every command but its a tiny overhead and steps around the issue of trying to
detect log-outs.
<br>
C.
<br>
<PRE>
-- 
Real Life: Caolan McNamara                Local: caolan@skynet
College:   9312811@ul.ie                  Quote: Happiness is a small sig.
</PRE>
<P>

<hr>
<!-- =====================================================================  -->
<H3><IMG ALIGN=BOTTOM ALT="" SRC="../gx/letter.gif"><a name="vi">VI Trick--Commenting Code</a></H3>

<P>
Date: Fri, 23 Aug 96 10:30:06 EDT<br>
From: dane@cci.com (Daniel Engel)<br>
To: gazette@ssc.com<br>
Subject: vi trick<br>

<P>
This is how you comment in a block of code using vi:
<pre>
	:START,ENDs/^/# /
</PRE>
where START is the starting line number and END is the ending line number.
<PRE> 
	i.e. :10,30s/^/# /
</PRE> 
comments in line 10 through line 30 of the current buffer (file).
<P> 
same idea can be used for indentation and/or commenting out.
<P>
d. at Nortel (dane@cci.com)
<P>

<hr>
<!-- =====================================================================  -->
<H3><IMG ALIGN=BOTTOM ALT="" SRC="../gx/letter.gif"><a name="sendmail">Masquerading with
SendMail</a></H3>
<P> 
Date: Tue, 20 Aug 1996 12:38:12 -0600 (MDT)<br>
From: murphyc@cadvision.com(Jim Murphy)<br>
To: gazette@ssc.com<br>
Subject: Article submission: Newbie Tip on Finding<br>
<P> 
Hi,
<P> 
As a new linuxer, I had a hard time finding my way
around the filesystem, and discovered that I often had to find
a file for some reason or other.  I knew the find command was 
out there, but remembering all the options required to make it
search the right places, find the right files, and print the 
right answers was something I couldn't do, at first.  So I made
up my own command, using a shell script, and called it "fnd".
<P> 
"fnd" takes one argument, the name of the file you want
to locate, complete with any wildcards you may wish to include,
and pipes its output to "less", which then allows you to view a
large list of results.  What you get, on each line of output, is
the complete path to anything which you're looking for!  I
find it amazingly useful (as is a rough familiarity with the 
"less" command.)  Here's my script:
<PRE> 
#!/bin/bash
find / -iname $1 -mount -print |less
</PRE> 
<P> 
That's it!  The -iname option tells find to be case insensitive,
the $1 is a variable which subs in your commandline argument,
-mount tells find not to descend directories on other file systems
like your cd-ROM (because mine is wonky and locks up the machine if
it is accessed;).  The -print option is required or you don't get
any output! (Get used to it, it's *nix...)  The | symbol tells
find to direct its output to the "less" command so you can see
your results in style!  Don't forget the / right after the find
command, or it won't know where to look.  -  Enjoy!  You won't 
regret the time you spend keying in this little shortcut, and 
don't forget to put it in a "bin" or "sbin" directory after
chmod'ing it to be executable. 
<P> 
Jim Murphy<br>
murphyc@cadvision.com<br>
<P>

<hr>
<!-- =====================================================================  -->
<H3><IMG ALIGN=BOTTOM ALT="" SRC="../gx/letter.gif">Masquerading with SendMail</H3>

<P> 
Date: Tue, 27 Aug 1996 21:49:27 +0100 (GMT+0100)<br>
From: "Robert S. Wolfram" <0wolfram01@Lelystad.Flnet.nl><br>
To: gazette@ssc.com<br>
Subject: Masquerading with sendmail<br>
<P> 
Hi Marjorie,
<P> 
First of all, I think it is wonderful that John devoted that much of
his spare time to start and maintain the Linux Gazette. I am happy to
see that it is being continued. Keep up this beautiful project! 
<P> 
I was very pleased with Leifs Queue-R-Mail Howto, published in issue 6
of the Gazette, but I was still left with one drawback. I have a PPP
dialup account at a local provider, and when sending remote mail, the 
"From" address and "Return Path" should be masqueraded to match my address
at the provider. If I would fill the $M macro in my 'sendmail.cf' (as
installed by Slackware 3.0), it would only change my domain, so it needed
some adjustment. I made some direct changes to  'sendmail.cf', but I did
make a backup before trying anything! Here are the changes I made:
<PRE> 
> # Before the edits
< # After the edits
</PRE>
<P> 
First of all, I filled the $M macro for masquerading my domain and added
a $N macro for masquerading my username:
<PRE> 
> DM
< DMlelystad.flnet.nl
< DN0wolfram01
</PRE>
<P>  
Disable masquerading for the local mailer in ruleset 40:
<PRE> 
> R$*			$: $1 @ $M		add local qualification
< #R$*			$: $1 @ $M		add local qualification
</PRE> 
<P> 
Include username in remote mail masquerading (rulesets 31 and 61):
<PRE> 
> R$* < @ $+ @ $+ >	$@ $1 < @ $3 >			$M is defined -- use it
< R$* < @ $+ @ $+ >	$@ $N < @ $3 >			$M is defined -- use it
> R$+			$: $1 < @ $M >			user w/o host
< R$+			$: $N < @ $M >			user w/o host
</PRE>
<P> 
If you send your mail as root, you might want to remove its special
treatment:
<PRE> 
> #CLroot
> CEroot
< #CLroot
< #CEroot
</PRE> 
<P>
After sending 'sendmail' a HUP signal, my mail got masqueraded so that
the receiver could just reply to the correct mailbox.<br>
Two remarks:<br>
First, to find ruleset xx, just search for Sxx in the beginning of a line.
Secondly, tab characters between the fields in the rewriting rules
are REQUIRED!!! Make sure you do not change those into spaces!
<P> 
Well, those were my $0.02. I hope it was still readable.
<P> 
Cheers,<br>
Rob.
<br>
<PRE>
Rob S. Wolfram   0wolfram01@lelystad.flnet.nl   rwolfram@wi.leidenuniv.nl
W3:                                       http://www.flnet.nl/~0wolfram01
=========================================================================
L I N U X :   T H E   C H O I C E   O F   A   G N U   G E N E R A T I O N
=========================================================================
</PRE>
<P>
<hr>
<!-- =====================================================================  -->
<H3><IMG ALIGN=BOTTOM ALT="" SRC="../gx/letter.gif"><a name="upgrade">Linux Upgrade</a></H3>

<P> 
<hr>
Date: Tue, 20 Aug 1996 11:02:03 -0400<br>
From: David Bourgin dbourgin@wsc.com<br>
Organization: WSC Technologies, Inc.<br>
To: linux@ssc.com<br>
Subject: Linux upgrade<br>
<P> 
Hello,
<P> 
Since this FAQ is comes up a very often in comp.os.linux.*,
and since it's already mentioned in kernels 2.0.12+ in
/linux/Documentation/Changes and in some News papers
(still one contacting me today: The editor of the german magazine
UNIXopen.) without any request from us.
So I'm going to ask for you to put something in your interesting
magazine. It's is how to avoid a lot of problem when upgrading
a Linux box. There's a easy way, and some scripts to run.
All the stuff is free, and is available by ftp, located at:
ftp://ftp.wsc.com/pub/freeware/linux/update.linux/
<P> 
The upgrade is full, it means it will upgrade any system
from 1.2.x (sorry I didn't test any earlier setup).
It will be right for anyone from new Linux users to experts
(no knowledge is required, no questions are done: all
is detected and worked fine for all people who tested,
except non-Lilo users since I force Lilo v19 to install.
This will change next.)
<P> 
All the upgrade is always up to date, and only fully tested
packages are included (!). For example, I didn't install
kernel 2.0.5 when it comes out since I've found out a bug
in the code (reported to Linus who did 2.0.6 as a patch).
Current upgrade contains:
<ul>
<li> New kernel: 2.0.x
<li> New gcc: 2.7.2
<li> New libc: 5.4.2
<li> New libg: 2.7.1.4
<li> New ld.so: 1.8.1
<li> New bin utils 86: 0.3
<li> New modules: 2.0.0
<li> New make: 3.74
<li> New lilo: 19
<li> New autoconf 2.10
<li> New binary utilities 2.6.0.15 (2.7 only works with kernel
2.0.8+)
<li> New dip 3.3.7o
<li> New fdisk 3.04
<li> New fdutils 4.2
<li> New fsck 1.04
<li> New ftape 2.08
<li> New gawk 3.0.0
<li> New getty_ps 2.0.7i
<li> New gpm 1.09
<li> New hd param 3.0
<li> New iBCS 2.0 960610
<li> New man 1.4h
<li> New man pages 1.12
<li> New mount 2.5k
<li> New mtools 3.0
<li> New ncurses 1.9.9e
<li> New net tools 1.32alpha
<li> New pcmcia modules 2.8.18
<li> New ppp 2.2.0f
<li> New ps,top,who,w 1.01
<li> New sendmail 8.7.5a
<li> New SysVinit 2.64
<li> New TermCap 2.0.8a
<li> New apache (HTML server) 1.1.1
<li> New boot sys 0.4a
<li> New DosEmu: 960807
<li> New ghostscript 3.33
<li> New IP masquerading ipfwadm 2.2
<li> New lclint 2.1b
<li> New loadlin 16
<li> New ntfs 951231a
<li> New quotas 1.51
<li> New samba
<li> New snd-util 3.5
<li> New socks 5b 0.16.4
<li> New Linux utilities: 2.5
<li> New WINE: 960717
<li> New Netscape: 3.0 (not a freeware!)
</ul><P> 
All is NOT installed when requested. For more details,
see README file at<br>
ftp://ftp.wsc.com/pub/freeware/linux/update.linux/
<P> 
I know some mirrors exist in Europe as:<br>
ftp://ftp.ibp.fr/pub/linux/update.linux/
<P> 
Note: I'm French, and I've moved two months ago from France.
So sorry if my American is not yet plain ;-)
<P> 
Bye,   David<br>
-- <br>
Web: http://www.accescyb.fr/~rezo1/homepage.html <br>
E-mail: dbourgin@wsc.com<br>
David Bourgin - Netware/Unix administration/security.<br>
I'm a netsurfer, and as such, a citizen of the worlda.<br>

<P> <hr> <P>
This page written and maintained by the Editor of <I>Linux Gazette</I>,
<A HREF="mailto: gazette@ssc.com"> gazette@ssc.com</A>
<P> 

<HR>
<center><h1>
<A NAME="news"><IMG SRC="../gx/bytes.gif"   ALT="News Bytes"></h1></center></A>

<P> 
<hr>

<!-- =====================================================================  -->
<H3><IMG ALT=" "   SRC="../gx/bolt.gif">GNU ID Utilities</H3>
<P> 
A new release of the GNU id utilities is available at
<a href="ftp://ftp.gnu.ai.mit.edu/pub/gnu/id-utils-3.2.tar.gz">
  ftp://ftp.gnu.ai.mit.edu/pub/gnu/id-utils-3.2.tar.gz</a>
<P> 
GNU id-utils is a package of simple, fast, high-capacity,
language-independent identifier database tools.  Actually, the term
``identifier'' is too limiting -- ID Utils stores tokens, be they
program identifiers of any form, literal numbers, or words of
human-readable text.  Database queries can be issued from the
command-line, or from within emacs, serving as an augmented tags
facility.
<P> 
Release 3.2 fixes a nasty bug in eid (a.k.a., "lid -R edit"), which
should only be a problem for users of non-emacs editors (e.g., vi).
This release also fixes as well as some minor portability problems.
If you use emacs and had no trouble compiling 3.1, there's no reason
for you to pick up this release.
<P> 
Additional information:  <a href="mailto:gnu@prep.ai.mit.edu"> 
gnu@prep.ai.mit.edu</a>
<P> 
<hr>

<!-- =====================================================================  -->
<H3><IMG ALT=" "   SRC="../gx/bolt.gif">InfoMagic Workgroup Server</H3>
<P> 
The InfoMagic Workgroup Server provides high-performance file and
printing services to PC and Macintosh clients using the Linux
operating system. It is the first Linux distribution designed
specifically for servers. Based on networking software created by the
Internet community and already in use at hundreds of companies,
universities and organizations worldwide, the InfoMagic Workgroup
Server provides simple graphical tools for system administration and
set-up. A Unix novice can set up a sophisticated server environment
in a couple of hours.
<P> 
Additional information:	
<a href="mailto:Orders@InfoMagic.com">
Orders@InfoMagic.com</a>,<br>
<a href="http://www.infomagic.com">
http://www.infomagic.com/</a><br>
<P> 
<hr>

<H3><IMG ALT=" "   SRC="../gx/bolt.gif">Linuxconf 1.6</H3>
<P> 
I am proud to announce the release of linuxconf 1.6. This release 
incorporates many enhancements and features. The last official release was 
1.3. Since, a major user interface rework has been done and many smaller 
features were added. Release 1.4 and 1.5 were only released on the 
linuxconf mailing list.
<P> 
It has been uploaded to sunsite in the pub/Linux/Incmoning directory and 
should move to /pub/Linux/System/admin/linuxconf-1.6.src.tar.gz. Binaries 
for both elf and a.out systems are provided at the same place.
<P> 
The major enhancement is the http mode. With this, you can now operate 
linuxconf using any web browser supporting forms. This makes 
administration of large linux networks a wonderful experience. You can 
navigate in linuxconf and even set bookmarks in your browser. Ultimatly 
one can build his corporate administration page with link to different 
part of linuxconf on different linux servers or workstation.
<P> 
Additional information: 
<a href="mailto:jack@solucorp.qc.ca">
jack@solucorp.qc.ca</a><br>

<P> 
<HR>

<!-- =====================================================================  -->
<H3><IMG ALT=" "   SRC="../gx/bolt.gif">Red Hat Rembrandt II</H3>
<P> 
Red Hat Linux - 3.0.4 (Rembrandt II) BETA now available on both the 
Intel and Sparc!
<P> 
The Rembrandt beta releases of Red Hat Linux include the following
nifty features:
<blockquote>
 Modular kernel (2.0.10)
 One kernel (one boot disk) for all hardware
 Increased hardware support over 3.0.3
 New, simpler installation
 PAM - Pluggable Authentication Modules
 More comprehensive X configuration
 New network configuration tool
 New version of RPM - 2.2.3
 Dependencies
 Libc 5.3.12
</blockquote>
<P> 
The Rembrandt II release fixes many bugs, and adds a few features.
Among the improvements over Rembrandt are:
<blockquote>
 config file handling
 DOS partitions
 IDE drives c-h
 flashing asterisk
 System.map
 interface cleanups
 /net /.automount updatedb
 dip, inn, fvwm95, ypbind, ftpuser
 module parameters
 package selection
</blockquote>
<P> 
The Red Hat Linux Rembrandt II is available from:
<P> 
<a href="ftp://ftp.redhat.com/pub/redhat/rembrandt">
ftp://ftp.redhat.com/pub/redhat/rembrandt</a>
<P> 
Additional information: <a href="mailto:rembrandt-list@redhat.com">
rembrandt-list@redhat.com</a>,<br>
<a href="http://www.redhat.com/redhat/rembrandt">
http://www.redhat.com/redhat/rembrandt</a> <br>
<P> 
<HR>

<H3><IMG ALT=" "   SRC="../gx/bolt.gif">SDK Software</H3>
<P> 
<P> 
URL CHANGE from The Boy With The Thorn In His Side
<P> 
The <i>free</i> Linux software which I release under the name "SDK Software"
(morepkgtools - supplements Slackware's pkgtool; tidylinks - searches
for and tidies dangling and messy symbolic links; LNET-Config -
configuration script for the LNET TCP/IP (KA9Q-like) program) has
moved from AOL to: <P>
<a href="http://www.kalika.demon.co.uk/sdk-software/index.html">
http://www.kalika.demon.co.uk/sdk-software/index.html</a>
<P>Revised versions of the software, with the correct URLs in the
manual pages etc., will be released shortly.
<P> 
If you have my AOL site bookmarked please change your records.
<P> 
FREE SOFTWARE ANNOUNCEMENT:
<P> 
<a href="http://www.kalika.demon.co.uk/sdk-software/index.html#man2html">
http://www.kalika.demon.co.uk/sdk-software/index.html#man2html</a>
<p><a href="http://www.kalika.demon.co.uk/sdk-software/myprogs/unix/man2html">
http://www.kalika.demon.co.uk/sdk-software/myprogs/unix/man2html</a>

<p>Version 0.15 of this <i>very</i> modest Bourne-shell script is now
available. The script is a CGI-bin interface to "man", converting
UNIX (Linux) manual pages into fully-legal HTML on the fly.
<P> 
Sample output can be viewed at:
<a href="http://www.kalika.demon.co.uk/sdk-software/index.html#manpages">http://www.kalika.demon.co.uk/sdk-software/index.html#manpages</a>
<P> 
Additional information: <a href="mailto:steve@kalika.demon.co.uk"> 
steve@kalika.demon.co.uk</a><br> 
<P>
<HR>

<!-- =====================================================================  -->
<H3><IMG ALT=" "   SRC="../gx/bolt.gif">Taper</H3>
<P> 
This is to announce the latest release of taper - version 6.7.4
<P> 
Taper is a user friendly, full featured tape backup solution for Linux.
Multiple features are supported including most recent restore, incremental
backups, archive verification and archive management. With triple
buffering, and internal compression, backup performance is quite
good. 
<P> 
Taper supports ftape, zftape, scsi, floppies, regular files and
removable media such as the IOMEGA ZIP drives.
<P> 
In addition, taper also supports IDE tape drives in ALPHA stage.
<P> 
PLEASE NOTE:<br>
I am going away for 2.5 months from this week and will be unreachable
via e-mail therefore, there will be no support for this version
until I return mid-late October.
<P>
Additional information: <a href="mailto:yusuf@nagree.u-net.com">
yusuf@nagree.u-net.com</a><br>
<P>
<HR>

<H3><IMG ALT=" "   SRC="../gx/bolt.gif">TranSend</H3>
<P> 
Advanced Network Products, Inc. announces the alpha release of TranSend.
TranSend, a "middleware" product, is an efficient, reliable and platform-
independent mechanism that provides communications between your applications.
<P> 
TranSend is a development toolkit that allows rapid construction of
real-time client/server and distributed applications. TranSend is the
perfect foundation for all of your network development needs, including:
<ul>
<li>Real-time client/server and distributed applications
<li>High performance data distribution systems
<li>Cross-platform connectivity
<li>Fault-tolerant systems
</ul>
Additional information: <a href="mailto:anpi@advanced-net.com">anpi@advanced-net.com</a>
<a href="http://www.advanced-net.com/examples.html">
http://www.advanced-net.com/examples.html</a><br>
<P> 
<HR> 

<H3><IMG ALT=" "   SRC="../gx/bolt.gif">Web-traversing Robot</H3>
<P> 
The VWbot Web-traversing robot is now available as shareware.
It was developed on Linux and hasn't been tested on anything else,
but it ought to run on most Unix. It uses Perl4 and libwww-perl-0.40
<P> 
The core program may be customised to perform different functions,
such as keeping track of document modifications, operating as
a restricted-domain search engine, or as a free-running agent
configured for tasks such as lexical analysis.
The robot adheres to the original Robot Exclusion Protocol and
includes code to implement the newer ROBOTS META tag.
<P> 
Additional information: <a href="mailto:robots@vancouver-webpages.com">
robots@vancouver-webpages.com</a>,<br>
<a href="http://vancouver-webpages.com/VWbot/">
http://vancouver-webpages.com/VWbot/</a><br>
<P> 
<HR>

<P><hr><p>
This page written and maintained by the Editor of <I>Linux Gazette</I>,
<A HREF="mailto: gazette@ssc.com"> gazette@ssc.com</A>
<P> 

<HR>
<center><A NAME="binstats"><h1>Binstats: Finding Unusable Binaries</h1></A></center>

<center><h4>by Larry Ayers <layers@vax2.rain.gen.mo.us></h4></center>

<center>Copyright (c) 1996</center><BR>


<center> <H5>Published in Issue 9 of the Linux Gazette</H5></center>

<HR>


Like many other Linux users, I've strayed from the path of my originally
installed distribution (Slackware 3.00) and have updated quite a large
percentage of the packages and libraries.  This can lead to problems; these
might come my attention when starting up a long-neglected executable only to
receive a message indicating that an essential library is missing, or that a
library has an incompatible executable format.<p>

Of course, you could spend a couple of hours every month or so and run ldd on
each and every executable on your system, writing down the results for every
one which has errors.  This could become tedious, I imagine.<p>

<a href="Mailto: Peter.Chang@nottingham.ac.uk">Peter Chang</a> is evidently a
whiz at cobbling together shell scripts which use various Unix utilities
chained and piped one to another.  <b>Binstats</b> is one of his, and it is
truly an ingenious contrivance.  You start it up (after editing it so that it
knows where all of your /bin directories are), the hard disc grinds away for a
minute or three, and this little script presents you with a list of all the
poor orphaned programs which can't run due to a lack of shared libs.  It also
lists how many of each type of executable (ELF, QMagic, statically linked,
etc.) you have, and which shared libs you have which aren't needed by any of
your executables.<p>

What really blew my mind was seeing a long list of duplicated executable
names, an unexpected result of many upgrades.  This can be a result of this  
imaginary scenario: Joe has been maintaining Package X for several years and
is weary of the constant email.  He gratefully transfers the maintenance of
Package X to an eager, energetic young programmer, Ed.  Ed is appalled to
find an installation procedure which doesn't follow the Linux Filesystem
Standard, and immediately changes the default installation directory from /bin
to /usr/local/bin.  Yours truly logs in at sunsite.unc.edu, finds a new
version of Package X, installs it, and is happy to see the new functionality.
Unfortunately the old Package X executable is living out the remainder of its
days, unknown to all, in /bin. Then Binstats does its work and the old X
binary and all its hoary cohorts are brought to light.<p>

All of the functions of Binstats can be done "by hand", of course.  The beauty
of this shell program is the combination of tasks into one, with the results
logged to a text file.  Then you can see at a glance several system
administration jobs which should be taken care of.<p>

Binstats is only four and one-half kb. archived in tgz format.  A copy of the
latest version is available at <a
href="http://www.ccc.nottingham.ac.uk/~etzpc/binstats.html">this UK site</a>,
as well as in <a href="ftp://sunsite.unc.edu/pub/Linux/utils/scripts/">this
sunsite directory</a>.  It's well worth the short download time, even if you
only run it once.

<hr>
<center><address><a href="http://vax2.rain.gen.mo.us/~layers/">Larry Ayers&lt;layers@vax2.rain.gen.mo.us&gt;</a></address></center><br>

<hr>

<A NAME="newsserver"><H1 ALIGN=CENTER>The Easy Way to Set Up a Local News
Server</H1></A>
<P ALIGN=CENTER>
By <STRONG> Christophe Blaess </STRONG> 
(<A HREF=mailto:ccb@club-internet.fr>ccb@club-internet.fr</A>).
<H2>Introduction</H2>
</P>

<P>A few months ago, I decided to set up a local news server on my Linux Box,
in order to read off-line the articles. Before then I read
the news directly from the Usenet server of my Internet Provider, at the
detriment of my phone bill...(Here, in France, even the local communications
are rather expensive)</P>

<P>Before trying to install and set up one of the two classical news servers
(<EM>Inn</EM> and <EM>CNews</EM>), I browsed a bit the Web, looking for a
possible other product. I then discovered a very powerful small package
"<A HREF="http://www.troll.no/freebies/leafnode.html"><EM>Leafnode</EM></A>" 
written by <A HREF="mailto:agulbra@troll.no">Arnt Gulbrandsen</A>.
It can be found in source form at <A 
HREF="ftp://ftp.troll.no/freebies/leafnode/"><EM>ftp://ftp.troll.no/freebies/leafnode/</EM></A> 
and the home page of this project is accessible at
<A HREF="http://www.troll.no/freebies/leafnode.html"><EM>http://www.troll.no/freebies/leafnode.html</EM></A>.</P>

<P>This package (leafnode-0.9.tar.gz, 29 Kb) contains three little programs, 
very easy to install, and to use:</P>
<UL>
<LI>
"<STRONG>Fetch</STRONG>" can feed a local news server from a remote 
Usenet server (university, ISP...).
It also posts the outgoing articles, using classical NNTP requests.
<P><LI>
"<STRONG>Leafnode</STRONG>" is a USENET server, run by the <EM>inetd</EM>
daemon, when a connection in required on the NNTP port, by a newsreader.
<P><LI>"<STRONG>Texpire</STRONG>" is generally run daily from the <EM>crontab</EM> 
to erase the oldest articles from the news spool.</P>

</UL>

<P>The main advantage of this system, is the transparent way it is 
inserted between the remote news server and the local newsreader. The 
remote Usenet server sees 
Leafnode exactly like a classical newsreader (like tin, trn, 
netscape,...) and the local
news reader sees Leafnode just like a USENET server.</P>

<P>I have been happily using Leafnode for several months, and I would like to
describe here the (very simple) steps to <A HREF="#INSTALL">install</A> and
<A HREF="#CONFIGURE">configure</A> it. Then I will explain some 
<A HREF="#HINTS">hints</A> to use it in a multi-users environment.</P>

<HR>

<H2><A NAME="INSTALL">INSTALLING LEAFNODE</A></H2>

<P>I've been using leafnode 0.8 for a few months, but I've recently upgraded
to version 0.9. The process for installing it is the same as the previous 
version, but I've had a little problem, maybe due to my version of 
<EM>make</EM>, and I'll describe how I have fixed it.</P>

<P>You first need to download the source code from 
<A HREF="ftp://ftp.troll.no/freebies/leafnode/">ftp://ftp.troll.no/freebies/leafnode/</A>. 
You'll get <STRONG>leafnode-0.9.tar.gz</STRONG>.</P>

<h3>Compiling the source code</h3>

<P>Become <EM>root</EM> and do :</P>
<PRE>
  # cd /usr/local/src
  # tar -xzf ~/leafnode-0.9.tar.gz
  # cd leafnode-0.9/
  # make	

</PRE>
<P>Everything must compile without any problem...</P>

<h3>Installing leafnode</h3>

<P>Make sure there is a "<EM>news</EM>" user and a "<EM>news</EM>" group 
on your system.</P>

<P>Then you can type:</P>
<PRE>
  # make install
</PRE>
<P>If "<EM>make install</EM>" complains when making directories in 
/var/spool/news/message.id/,
you may have the same problem as I've had. To fix it, I've modified the 
Makefile, to insert a part of the leafnode-0.8 Makefile:</P>
<P>replace the lines (in the "install:" section)</P>
<PRE>
  cd $(SPOOLDIR)/message.id
  for a in 0 1 2 3 4 5 6 7 8 9 ; do for b in 0 1 2 3 4 5 6 7 8 9 ; do \
     mkdir ${a}${b}0 ${a}${b}1 ${a}${b}2 ${a}${b}3 ${a}${b}4 ; \
     mkdir ${a}${b}5 ${a}${b}6 ${a}${b}7 ${a}${b}8 ${a}${b}9 ; done
</PRE>
<P>by the line</P>
<PRE>
  -mkdir -p $(SPOOLDIR)/message.id/{0,1,2,3,4,5,6,7,8,9}{0,1,2,3,4,5,6,7,8,9}{0,1,2,3,4,5,6,7,8,9}
</PRE>
<P>Be sure that the first character on the line is a tabulation. Then try again:</P>
<PRE>
  # make install
</PRE>

<h3>End of installation</h3>

<P>At this point, Leafnode will have installed the following files on your 
system:</P>
<PRE>
    /usr/local/sbin/fetch
    /usr/local/sbin/leafnode
    /usr/local/sbin/texpire
    /usr/local/man/fetch.8
    /usr/local/man/leafnode.8
    /usr/local/man/texpire.8
    /usr/lib/leafnode/config.example
    /var/spool/news/ ... and a lot of subdirectories ...
</PRE>
<P>First you have to copy the file /usr/lib/leafnode/config.example to
/usr/lib/leafnode/config and edit him, to put the name of your
remote NNTP server in place of:</P>
<PRE>
  server = news.hiof.no
</PRE>
<P>Edit the file /etc/inetd.conf, and look for a line like:</P>
<PRE>
  nntp   stream   tcp   nowait   root   /usr/sbin/tcpd   in.nntpd
</PRE>
<P>then modify it like this:</P>
<PRE>
  nntp   stream   tcp   nowait   news   /usr/sbin/tcpd   /usr/local/sbin/leafnode
</PRE>

<P>Depending on your configuration, there may be no TCP wrapper installed, 
so you could have to remove the "/usr/sbin/tcpd" part of this line.</P>
<P>and do</P>
<PRE>
   # killall -HUP inetd
</PRE>

<H3>First run</H3>

<P>Connect to your Usenet provider, and, while being
"<EM>root</EM>" or "<EM>news</EM>", run:</P>
<PRE>
  $ fetch
</PRE>

<P>You will have to wait for a moment, because Leafnode is asking the
NNTP server the list of all the active groups. Once fetch ends,
run a newsreader, as normal user, and ask him to contact the localhost.
for example, with <EM>tin</EM> do:</P>
	<PRE>
  $ export NNTPSERVER=localhost
  $ tin -r
	</PRE>
<P> (With Netscape you have to put <EM>localhost</EM> in "Options/Mail 
and News/Servers/NNTP server")</P>

<P>You will get the list of all available newsgroups, then choose interesting
ones, and read them. They will appear empty at this time. It's normal.</P>

<P>As <EM>root</EM> run again fetch. It will download all the content of the
previously read newsgroups. The first downloading will take a while, but
the next will obviously be very much quicker.</P>

<P>Your local USENET server is installed!</P>

<hr>

<H2><A NAME="CONFIGURE">CONFIGURING LEAFNODE</A></H2>

<P>Now that leafnode run on your Linux box, you can configure some details:<P>

<P>In the file /usr/lib/leafnode/config, there are two fields you can edit:</P>
<UL> 

<LI>the <EM>expire</EM> value tells to Texpire the number of days an 
unread thread must be kept.

<LI>the <EM>maxcount</EM> value limits the number of articles to download 
from a newsgroup, during a single execution of Fetch.
</UL>

<P>Refer to the comments in this file to choose the values (on my system I use
expire=10 and maxcount=3000)</P>

<P>You must run Texpire from time to time, and a daily entry in the 
crontab seems to be the best choice:</P>

<PRE>
  00  03  *  *  *  news  /usr/local/sbin/texpire
</PRE>

<P>This line is for a system-wide crontab (generally /etc/crontab 
maintained by root),
but you will have to remove the username "<EM>news</EM>" on user crontab 
(/var/spool/cron/news).</P>

<P>You will have to run periodically <EM>fetch</EM>. If you have a 
permanent link with your news server, there's an obvious solution: the 
crontab again, to run it once per hour for example.</P>
<PRE>
  00   *  *  *  *  news  /usr/local/sbin/fetch
</PRE>
<P>If you're using a non-permanent PPP connection, you can insert 
<EM>fetch</EM> at the end of the ip-up shell script (see Linux Gazette 7 
"<CITE>Setting up PPP's ip-up and ip-down scripts!</CITE>").</P>

<P>Put the name "<CODE>localhost</CODE>" in /etc/nntpserver or set the 
environment variable (for example in /etc/profile):</P>
<PRE>
	export NNTPSERVER=localhost =

</PRE>

<HR>

<H2><A NAME="HINTS">USING LEAFNODE ON MULTI-USERS SITE.</A></H2>

<P>There are problems to avoid, especially if there are other users 
on your system, or if the leafnode server is on a local network. First 
you must make sure of the validity of the headers
in the outgoing posts, but you also need to limit the list of the accessible 
newsgroups. Don't
forget that Fetch will download the whole content of a newsgroup if someone 
tries to read it. So, be very careful with newsgroups like 
<EM>alt.binaries.pictures...</EM></P>

<P>Fortunately, we can use some little awk and shell scripts to 
<A HREF=#VERIF>verify and correct</A> the outgoing posts, and to 
<A HREF=#LIMIT>limit the local access</A> to selected newsgroups.</P>

<H3><A NAME="VERIF">Checking the outgoing posts</A></H3>

<P>With some newsreaders, the "From:" field of the outgoing articles will 
be set to <EM>myname@my.station.on.my.lan </EM> and not 
<EM>myname@my.internet.provider.com</EM>.
With some of them you can configure the "From:" and "Reply-to:" fields, 
while the others need you to recompile them.</P>

<P>This problem can be worse if you have a Linux box with several users. 
Some of them can have misconfigured newsreader (sometimes on purpose...) 
and it may be safer to check the headers of the outgoing articles before 
posting them.</P>

<P>Here's a small awk filter which allows a kind of masquerading of the
"From:" line of an article. It will change the line
<CODE>"From: <EM>username@my.station.on.my.lan (user real name)</EM>"</CODE>  
to a line
<CODE>"From: <EM>username@my.internet.provider.com (user real name)</EM>"</CODE>.</P>

<P>You may also ensure that <EM>username</EM> is correct (i.e. in a list 
of allowed users).  The same Perl script will help us to determine the 
correct articles. Otherwise it will add a line "*** Wrong From field - 
This article must be deleted ***" to the message.</P>

<PRE>
  #! /usr/bin/gawk -f
  #
  # /usr/local/sbin/change_article_from_domain
  #
  # awk script to change the domain name on the "From:"
  # line of outgoing articles.
  # If the username is not valid a message will be added
  # at the bottom o the file, allowing a 'grep' to delete
  # him.

  BEGIN {
    # replace with the correct domains
    local_domain="my.station.on.my.lan"
    real_domain ="my.internet.provider.com"
    # insert here the name of your users allowed to post articles
    # (may be just one)
    valid_usernames["user1"]
    valid_usernames["user2"]

    must_be_deleted=0
  }

  /^From:/ {
    gsub(local_domain, real_domain)
    username=substr($2,1,index ($2, "@")  - 1)
    if (! (username in valid_usernames)) {
      must_be_deleted=1
      # you can also add a system command
      # example : mail to newsmaster with
      # the username of the wrong article
    }
  }

  END {
    if (must_be_deleted != 0) {
      print "*** Wrong From field - This article must be deleted ***"
    }
  }

  {
    print
  }
</PRE>

<P>This script can be useful if you have up to, say, ten users, otherwise
you'll need to improve it in order to read the list of allowed users in 
an otherfile for example.</P>

<P>Now we will execute the above script on all the outgoing articles, 
sitting in /var/spool/news/out.going, then delete (or move to another 
directory) those with bad usernames.</P>

<PRE>
  #! /bin/bash
  #
  # /usr/local/sbin/modify_outgoing_articles
  cd /var/spool/news/out.going
  for i in * ; do
    /usr/local/sbin/change_article_from_domain < $i >/tmp/modified_articles/$i
  done
  rm -f *
  mv /tmp/modified_articles/* .
  rm -f `grep -l "*** Wrong From field - This article must be deleted ***" *`
</PRE>

<P>(The backquote is used to catch the result of grep) Don't forget to create 
a /tmp/modified_articles/ directory.
This script cannot prevent the fake "From:" lines, when an authorized user is
hidden behind an other
correct username. This can not be easily done, and if you really don't trust 
your users, you'll have to use another Usenet package like Inn or Cnews.</P>

<P>Now all the outgoing articles will have a correct "From:" line.</P>

<H3><A NAME="LIMIT">Limiting the list of accessible news groups</A></H3>

<P>The second important point to check out is the list of fetched newsgroups. 
If you haven't got a huge disk
space, it would be better to avoid downloading <EM>alt.binaries</EM> groups 
or <EM>alt.2600.</EM> for example...
The problem is that fetch will download the content of each newsgroup 
corresponding to a file in /var/spool/news/interesting.groups, for 
example <CODE>/var/spool/news/interesting.groups/comp.os.linux.announce</CODE></P>

<P>A file in this directory is touched by leafnode every time a user tries 
to read the content of the group. Are you
sure that none of your users will try to have a look at 
alt.binaries.pictures.erotica...? 
So there are two possible solutions:</P>
<UL>
<LI>You can edit the /usr/lib/leafnode/groupinfo file, in order to suppress 
the forbidden newsgroups. But this is not a very good solution, because 
fetch will re-create it during the next connection.

<LI>You can suppress the forbidden groups from 
/var/spool/news/interesting.groups.
This can be done by shell script, in many different ways, for example:
</UL>

<PRE>
  #! /bin/bash
  #
  # modify_interesting_groups

  cd /var/spool/news
  rm -f alt.*
  rm -f *windows*
  ...
  touch comp.os.linux.announce
  touch comp.windows.x.announce
  ...
</PRE>

<P>you can also have a definite list of fetched newsgroups, for example in
/var/spool/news/official.list, manually created:</P>

<PRE>
  # mkdir /var/spool/news/official.list
  # cd /var/spool/news/official.list
  # touch comp.os.linux.announce
  # touch comp.os.linux.answer
  # touch comp.lang.c.moderated
  ...
</PRE>

<P>And this list will be copied in the interesting.groups directory by a
script before each execution of fetch.</P>

<PRE>
  #! /bin/bash
  #
  # modify_interesting_groups

  cd /var/spool/news
  rm -f *
  copy /var/spool/news/official.list .
</PRE>

<P>Now we have two scripts to insert before running fetch. For example in
 /etc/ppp/ip-up:</P>
<PRE>
  #! /bin/bash
  #
  # /etc/ppp/ip-up
  ...
  /usr/local/sbin/modify_interesting_groups
  /usr/local/sbin/modify_outgoing_articles
  fetch
</PRE>

<HR>

<H2>CONCLUSION</H2>

<P>I think that <EM>Leafnode</EM> is a very interesting package for those
 (most of us) who are running
Linux on a standalone box with intermitent connection to a Usenet server,
 or on a small local network
with few users. It's powerful and much simpler to install and to configure 
than Inn or Cnews,
designed for bigger sites. Moreover it does not require any maintenance.</P>

<HR>

<P>Christophe Blaess (<A HREF="mailto:ccb@club-internet.fr"><EM>ccb@club-internet.fr</EM></A>).</P>

<HR>
<center><A NAME="filerunner"><h1>FileRunner: A New Tk/Tcl File
Manager</h1></center></A>

<center><h4><a href="mailto: layers@vax2.rain.gen.mo.us">by Larry
Ayers</a></h4></center>

<center>Copyright (c) 1996</center><BR>

<center><H5>Published in Issue 9 of the Linux Gazette</H5></center>
<hr>

<center><h3>Introduction</h3></center>

Recently I happened upon a new Tk-based filemanager, written by <a
href="mailto: hch@cd.chalmers.se">Henrik Harmsen</a>, who evidently lives in
Sweden.  I've been using TkDesk quite a lot lately (see my review in LG #8) so
at first I was struck by the resemblance, but the program upon further
exploration comes from a different philosophy of file-management and fills a
different niche in the Linux software world.<p>

<hr>

<center><h3>Amigan Origins</h3></center>

I'll let Henrik Harmsen relate to you FileRunner's origins:<br>

<i>On my good old Amiga, there were a few great file managers called things
like DirMaster and Directory Opus. They were very simple (in concept) but
highly usable. Especially I came to love the
two-filelistings-and-command-buttons concept which is what I've done in
FileRunner. It's a fast and intuitive way of doing file handling commands as
very many file commands (mv, ln, cp etc) are happy with a source and a
destination argument. It also works great for commands that only take a list
of source files, like an image viewer and an editor. I was considering the way
the FileManager in Windows displays directory trees, but I came to the
conclusion that those tree views are more often than not a nuisance to
navigate. Instead I implemented the cool Directory menu that can take you
anywhere in the file system by mapping directories to sub-menus (even the ".."
directory :-).  So, basically I just wanted to get back the high usability I
had with the old file managers on the Amiga (plus add my own stuff like FTP
browsing, the history and hotlist etc).<p>

I had a look at a few other file managers for Unix/X11 but none of them
were even close to the convenient two-filelistings-and-command-buttons
concept. Some were bloated, some wouldn't compile, some needed Motif etc,
and none were intuitive (to me :-). So I set out to create my own. Why am
I just not a happy TkDesk user? Well I started working on FileRunner long
before TkDesk came out and TkDesk wasn't quite what I was shooting for so
I continued on FileRunner. That also gave me exactly the file manager I
wanted, of course :-)</i><p>

After I read the above explanation, I realized what FileRunner reminds me of:
I used to use a Norwegian OS/2 filemanager called Dirmaster, which was also
inspired by the Amiga filemanagers of yore.  It had a layout similar to
FileRunner's, with programmable function buttons and twin directory panes.<p>

It's interesting that software has been around long enough that
traditions have evolved.  In the text editor world there are emacs, vi, and
"windows/CUA" strains of editors, while many filemanagers have followed either
a "Norton Commander" tradition, an Amiga tradition, or a mouse-based iconic
tradition, with various hybrid strains emerging and recombining.<p>

<hr>

<center><h3>Features</h3></center>

FileRunner will inevitably be compared to TkDesk, as they share many features
and are both Tk-based.  The two programs aren't really designed for the same
purposes, though.  TkDesk is more of a desktop manager combined with a
filemanager, as it has an integral icon-bar which can serve the same purpose
as Fvwm's buttonbar, or several other similar utilities.  It's best used as an
app you would open when starting an X-window session and leave open for the
duration.<p>

FileRunner is a much less resource-hungry application which starts quickly and
lends itself to quick tasks followed by dismissal.  It uses around one-third
the memory TkDesk uses.  It is probably a more appropriate choice for a
slower, memory-constrained machine, whereas if you have a fast CPU with plenty
of RAM TkDesk or Moxfm will run well without using a disproportionate amount
of your system resources.<p>

Among the many thoughtfully designed features in FileRunner, the following I
found to be particularly useful:<br>

<ul>

  <li>A button for each directory pane which will start an xterm or rxvt in
      the current directory

  <li>Directory hotlists in a drop-down menu

  <li>A dynamically expanding directory tree which allows quick traversal of
      the entire filesystem

  <li>Quick views of text files via a single right-mouse-button-click

  <li>Function buttons in a column dividing the directory panes, with
      instructions for creating new ones

  <li>Menu showing directories visited during the session

  <li>Ability to open remote directories via FTP and browse them as if they
      were local

</ul>

Here's a screenshot of a Filerunner window:<br>

<center><a href="./gx/filerunner9.gif" >FileRunner</a></center><P>
<p>

FileRunner also has a feature which has become fashionable lately in many of
the newer apps: quick rereading of the configuration file(s), allowing
customization to be done quickly.  I first saw this in Fvwm; it's a real
time-saver.
<hr>

<center><h3>Quirks</h3></center>

FileRunner is almost entirely mouse-based in this first release.  I like to
use a mouse, but I'm fond of arrow-keys and page-up and page-down keys as
well.  Henrik Harmsen mentioned in an email message that keyboard support is
in the works.<p>

When using the FTP function, any downloading activity prevents you from doing
anything else until its done.  A separate process or thread would be nice for
this.  One way around this limitation is to open another instance of
FileRunner. The program is small and fast enough that this is feasible.<p>

Aside from these two minor complaints, I found the program to be stable and
reliable.  It's a relatively small download; why not give it a try?<p>
<hr>

<center><h3>Availability</h3></center>

As of August 15, 1996, FileRunner can be found in the /pub/Linux/Incoming
directory of ftp://sunsite.unc.edu and its mirrors.  I imagine that it will
eventually be moved to /pub/Linux/X11/xutils/managers.

<hr> <center><address><a href="http://vax2.rain.gen.mo.us/~layers/">Larry
Ayers&lt;layers@vax2.rain.gen.mo.us&gt;</a></address></center><br>

<HR>
<A NAME="staroffice"><center><h1>Getting Up and Running on StarOffice
3.1</h1></center></A>

<center><H4>by Dwight William Johnson johnson@olympus.net</h4></center>

<center>Copyright (c) 1996</center><BR>


<center> <H5>Published in Issue 9 of the Linux Gazette</H5></center>

<HR>

<P> 
Happy as can be, I am composing this in Linux on StarWriter, the extremely
capable word processor that is part of the StarOffice 3.1 suite of business
applications. Below I will guide you through the labyrinth of tricks and fixes
that you will need to get this first beta of StarOffice working on your Linux
system.

<p>Star Office 3.1 is a suite of office productivity applications containing
StarWriter 3.1 (word processor), StarCalc 3.1 (spreadsheet), StarDraw 3.1
(graphics and presentation package), StarImage 3.1 (image manipulation)
StarChart 3.1 (bar-, pie- and other charts) and StarMath 3.1 (formula design).
StarOffice 3.1 makes heavy use of common code in shared libraries, therefore
using relatively few resources for the level of functionality.

<p>If you don't have the patience to fix things that aren't right, you should
wait for the next beta release. But if, like me, you don't mind fixing a few
things to get the immediate gratification of state-of-the-art business software
on Linux that you would pay $400 for on Windows 95, read on.

<p>To save you frustration and disappointment, I must also caution you that you
will need about 200 megabyes on your hard drive to install StarOffice the way
I outline below.

<p>Like many Linux users, I was delighted when the German company StarDivision
announced it would release an international Linux version of its major office
suite StarOffice, which competes head to head with Microsoft Office in Europe.
And still more pleased when I learned that for non-commercial use
StarOffice would be free.

<p>I am almost never, however, the first to jump into a new application. Let
others find the bugs; let me find a productive application is my credo.

<p>Nevertheless, the prospect of having a high-end WYSIWYG word processor to
use in Linux was irresistible. When StarDivision was pressured into an early
release of the first beta of StarOffice 3.1 for Linux on July 31, I rushed up
to the StarDivision Home Page at <A HREF="http://www.stardivision.de/index.html">
http://www.stardivision.de/index.html</a> and
linked to one of a number of possible download sites which in my case was
<A HREF="ftp://ftp.io.org/pub/mirrors/linux/sunsite/apps/staroffice">
ftp://ftp.io.org/pub/mirrors/linux/sunsite/apps/staroffice</a>.

<p>What I saw stopped me. StarOffice3.1 is a more than a 40Mb download in 53
files, mostly diskette images. The installed product takes over 120 megabytes.

<p>I decided to see what the feedback was on StarOffice before investing in that
kind of bandwidth. I aimed my Netscape newsreader for a place I knew I could
count on: comp.os.linux.development.apps.

<p>The news was bad. Many users were having problems installing StarOffice and
those who could were finding lots of bugs, mostly segmentation faults.

<p>The worst news of all was being reminded that StarOffice 3.1 requires Motif 2.0.
Even though StarOffice was free, it did not seem worth the $100-$200 investment
in Motif 2.0 to bring up an application that might be useless because of
segmentation and other faults.

<p>As the days passed, however, the news began to come in that some were getting
StarOffice installed and were quite thrilled with its look and feel.

<P>Then on August 18 Peter Klein wrote in the Redhat-List:
<br>
"to run and even to install Star Office 3.1. you don't need Motif."
<br>
I found his tips and tricks irresistible. I decided to try StarOffice.

<P>Below is step-by-step how I installed StarOffice 3.1 on my Red Hat 3.0.3
system with Metro-X server upgraded to the 2.0.10 kernel using Peter Klein's
recipe garnished with a few additional tricks and fixes which I gleaned
from the usenet and mailing lists.

<ol>
<P><LI>I went to my chosen ftp site,
<A HREF="ftp://ftp.io.org/pub/mirrors/linux/sunsite/apps/staroffice"</a>
ftp.io.org/pub/mirrors/linux/sunsite/apps/staroffice</a>,
(Alternatively, you can go to any sunsite mirror. Sunsite is at
<A HREF="ftp://sunsite.unc.edu/pub/Linux/apps/staroffice">
sunsite.unc.edu/pub/Linux/apps/staroffice</a>.)
and downloaded the file 'staroffice.README'. Here I learned about the
StarOffice for Linux license, StarDivision's plans for the product, how
to connect with the StarOffice technical support and development system 
and valuable installation notes.

<P><LI>My internet service provider has free hours from midnight to 7:30 A.M. 
So I waited till midnight, fired up xtftp, pointed it to the ftp site, 
created a ~/download/staroffice directory for the download, highlighted 
all 53 files, pressed 'copy' and 'ok' and went to bed.

<p>(The 'staroffice.README' mentions that you can install only part of 
StarOffice. So my first attempt, not recommended, was to install just 
the common files with StarWriter. Unfortunately, the install program 
comes to an error which has to be stepped past each time it finds a file 
from the complete package that is missing. After clicking 'ignore' with
my mouse for the first hundred or so times, I abandoned this attempt.)

<P><LI>If you are fortunate to have Motif 2.0 installed on your system, you
can skip down to step 6.

<P><LI>Peter Klein: "You have to install the SO libraries <I>before</I> you can start 
the install program."

<UL>
<P><LI>I created the directory ~/so-temp and executed:
<pre>
cp ~/download/staroffice/file.01-? ~/so-temp
cd ~/so-temp
unzip file.01-1
unzip file.01-2
unzip file.01-3
unzip file.01-4
unzip file.01-5
unzip file.01-6
unzip file.01-7
</pre>
In the ~/so-temp file you now have a bunch of subdirectories with the unzipped
files in them.

<P><LI>Create a directory to hold the 'so' files. I created mine in /usr/X11R6/lib
with the command:
<pre>
mkdir /usr/X11R6/lib/so-libs.
</pre>
<P><LI>Among the subdirectories created in ~/so-temp you will find the 
~/so-temp/lib* directories. I manually went through the 
~/so-temp/lib* directories and copied every file to /usr/X11R6/lib/so-libs.
I used Midnight Commander, a handy two-panel file manager for the virtual
console, for this task.

<P><LI>Next, in the /usr/X11R6/lib/so-libs directory I created the symbolic links:
<pre>
cd /usr/X11R6/lib/so-libs
ln -s libso312.so libMrm.so.2
ln -s libso312.so libXm.so.2
</pre>
<P><LI>I edited the /etc/ld.so.conf file to contain the line:
<pre>
/usr/X11R6/lib/so-libs
</pre>
<P><LI>I executed ldconfig from root.

<p>Thank you Peter Klein.

</UL>

<P><LI>Not recommended because of the large number of failures reported,
but still worth checking out for an install approach, is the 'StarInst' 
Perl script by Steffen Winterfeldt which can be downloaded at:<BR>
<A HREF="http://www.physik.uni-leipzig.de/~wfeldt/starinst/starinst.html">
http://www.physik.uni-leipzig.de/~wfeldt/starinst/starinst.html</A>.<BR>
This script is alleged to work with installed versions of Motif 1.2.

<P><LI>I next ran StarOffice Install which must be executed from root.
Install is very easy to use. Just make sure you have enough disk space
(about 125 Mb) and follow the simple prompts that Install presents in its
dialog boxes. If you have Red Hat and you are executing Install from its
own directory, you will need to remember to enter './Install'. I installed 
StarOffice into /usr/local/StarOffice3.1.

<P>At the end of Install, I was given directions for proceeding with the second
part, the user installation. During the user installation, a user-specified 
directory and configuration files are created for each user. I found it very 
easy to just follow the directions in the dialog boxes. I answered 'yes' to
the prompt to copy the templates and demo documents to my directory tree
because I had been informed that if I answered 'no', I would not be able to 
modify these documents, although I could still access them as read-only master 
copies.

<P><LI>It is now necessary to fix a couple of things that StarDivision didn't
think of in order to actually use StarOffice. I had to glean these fixes from 
posters to comp.os.linux.development.apps and the Redhat-List after finding
that my installation of StarOffice didn't work.

<P>Logged in users must get permission to use StarOffice's fonts and other files. 
Log ('su') into root, and
<pre>
chgrp -R users /usr/local/StarOffice3.1/Xp3.
</pre>
J. Maynard Gelinas proposed the alternate
<pre>
chmod -R a+r /usr/local/StarOffice3.1/Xp3
</pre>
for this fix on the Red-Hat List.

<P><LI>Printing is impossible without this next fix. While logged as root,
<pre>
mkdir /tmp/XpSp_ 
mkdir /tmp/Xp_
mkdir /tmp/XpSp_/tmp 
mkdir /tmp/Xp_/tmp
chgrp -R users /tmp/XpSp_
chgrp -R users /tmp/Xp_
</pre>
<P><LI>To get the on-line help system (unfortunately only in German in this 
release), you will need to start the two daemons, 'svdaemon' and 'svportmap' 
before starting StarOffice. This is how I did it.

<P>While logged as root, I added the lines:
<pre>
/usr/local/StarOffice3.1/linux-x86/bin/svdaemon &
/usr/local/StarOffice3.1/linux-x86/bin/svportmap &
</pre>
to my /etc/rc.d/rc.local.

<P><LI>Finally, as directed by the StarOffice user installation, I edited my 
~/.bashrc file to contain the line:
<pre>
. ~/.sd.sh
</pre>
and rebooted my computer.

</OL>

<P>And, if you followed along with me -- you are finished! All the download and 
intermediate directories can now be deleted. (Wait a few days, in case you 
find you have to redo something.)

<P>Log into your user account, 'startx' and open an 'xterm'. The StarOffice
applications start at the command line with:
<pre>
swriter3
scalc3
sdraw3
schart3
simage3
smath3
</pre>
You will find bugs in this release of StarOffice. But by experimenting, you
will also find work-arounds.

<P>For example, I found that exiting the search and replace dialog in StarWriter
would generate a fatal segmentation fault. But by activating the cursor in 
the search area and pressing <Esc> I can close the dialog box and continue 
working.

<P>Also, when you use the scroll bar, StarWriter loses its blinking cursor. I 
just go to the menu bar and activate a pull-down menu and then click my mouse 
in the document area. Wherever the mouse touches down places the blinking 
cursor.

<P>For printing on my Postscript printer I have found that I need to set the 
printer to 'NULL' and the default options to 'lpr'.

<P>There is a neat little button bar, 'soffice3', that is designed to coordinate 
all these applications. Unfortunately, bugs make it unfunctional. So look, 
but don't touch until the next release.

<P>Tip of the day for enjoying StarOffice beta one: save your work often.

<P>StarDivision operates a news server with StarOffice newsgroups at:<br>
<A HREF="news://starnews.stardivision.com/beta.staroffice.linux">
news://starnews.stardivision.com/beta.staroffice.linux</a><br>
where users share their experiences. Be sure to connect during German 
business hours -- no nights or weekends. The same goes for their Web site.

<p>StarDivision is eager to get your bug reports at:<br>
<A HREF="mailto:linux-suggest@stardivision.com">
linux-suggest@stardivision.com</A><BR>
I am still getting my feet on the ground in StarOffice. But I found 
composing this article in StarWriter quite easy and fun. I used a lot 
of cut and paste between multiple windows, formatting, changing fonts, 
printing, search and replace -- all the basic things you do in word
processing. But I didn't even scratch the surface of the capabilities 
of just StarWriter. And I haven't even looked at the other applications 
yet. I am looking forward to exploring StarOffice in the coming months.

<p>I congratulate the vision of StarDivision to recognize that Linux is an 
important platform for major application development.


<hr>

<HR>
<a name="yodl"><center><h1>YODL: A New, Easy-To-Use Text Formatting
Language</h1></center></A>
<hr>

<center><h4><a href="mailto: layers@vax2.rain.gen.mo.us">by Larry
Ayers</a></h4></center>

<center>Copyright (c) 1996</center><BR>

<center><H5>Published in Issue 9 of the Linux Gazette</H5></center>


<center><h3>Introduction</h3></center>

Linux/un*x is rich in difficult-to-learn, intricate text formatting systems.
Though the quality of printed output can be extremely high, the learning curve
can be concomitantly steep.<p>

This has led to the development of "meta-formatting" systems, which allow one
common text mark-up system to be output in several formats.  An example is
SGML, which is very capable but hardly intuitive.<p>

<a href="mailto: karel@icce.rug.nl">Karel Kubat</a>, a Dutch programmer, has
written a new text formatting system which he calls YODL, for Yet OneOther
Document Language. (I get the impression that the awkward "oneother" word
construct exists because Mr. Kubat wanted to have <i>.yo</i> as a filename
suffix, rather than <i>.ya</i>.)<p>

<hr>

<center><h3>Origins</h3></center>

Karel Kubat had found himself spending an inordinate amount of time marking up
documents for HTML, then having to repeat the process for Latex; in other
words, he wanted to have a nicely formatted printout of a document and also
make it available on a web-page.  For various reasons SGML was not satisfying
his needs so he set out to write an easy to use document language.  His
criteria included a minimum of awkward-to-type tags and the ability to include
or reference other files in one master file.  I think he succeeded remarkably
well.  

<hr>
<center><h3>Features</h3></center>

A sample document will give you a good idea of what a .yo file looks like:<br>
<pre>

COMMENT(File for testing purposes.)

htmlbodyopt(fgcolor)(#0000E0)
htmlbodyopt(bgcolor)(#E0E0C0)

article(Test article for YODL)
       (Karel Kubat)
       (1996)
       
sect(First section) label(first)

This is the first section. Now for a subsection:

subsect(Subsection of first section.)
   
This is the subsection.
    
sect(Second section.)

This is the second section, but ref(first) is the first one.
Lets try some verbatim text.

verb(
 #include <stdio.h>

 int main ()
 {
     printf ("Hello World!\n");
     return (0);
 })

Now some weird characters: !@#$%^*[]{}\|"~`'.

Accents? em(Ich m\"ochte bitte \"uberhaupt ein Bier!)
bf(Apr\`es moi la d\'eluge.) 

subsect(Some lists.)

subsubsect(An itemized list.)

itemize(

    it() Item one.
    
    it() Item two.
    
)
 
subsubsect(A descriptive list.)

description(

    dit(First:) Item one.
    
    dit(Second:) Item two.
    
)

subsubsect(An enumerated list.)

enumerate(

    eit() Item one.
    
    eit() Item two.
    
)

</pre>
This test file shows you what some of the tagging is like.  Here's a brief
excerpt from a .yo file showing how other files from the current directory can
be included:<br>
<pre>
sect(Using the yodl program)
includefile(using)

subsect(Language elements)
includefile(elements)
</pre>

As you can see the tags are in an abbreviated verbal form, which makes them
easier to remember.  The included files can be completely devoid of tagging.
This allows you to concentrate more on content rather than structure; the
structure can be mostly contained within the master document.  YODL documents
are noticeably easier to read in their source format than either HTML or Latex
source.  Format-specific tags are also allowed, as in the above HTML body
tags. They will be ignored by the Latex parser, for example. <p>

YODL is very well documented, and the installation of the docs is an effective
demonstration of the system.  The docs come in .yo source form; these files
are copied to /usr/local/yodl/doc, and then YODL is run on them during the
installation, with the end result being several HTML files.<p>

As is true with any "meta-formatter", a user is better-off knowing something
about the output format.  I think it's necessary to at least glance through
the output, just to catch any obvious errors. The advantage of a program like
YODL is that it will enable you to avoid the grunt-work of starting from
scratch.

<hr>

<center><h3>Package Contents</h3></center>

The main YODL program is written in C, and it works in conjunction with
several shell scripts and macro files.  The main supported formats are HTML,
Latex, and the troff/groff man and ms formats. There is limited support for
conversion to plain ASCII and SGML, as well.  As previously mentioned, the
documentation is unusually extensive, being about sixty pages worth of
well-written material.

<hr>
<center><h3>Availability</h3></center>

YODL can be downloaded from <a href="ftp://ftp.icce.rug.nl/pub/unix">its home
site</a> under the filename yodl-X.YY.tar.gz, with X.YY being a version
number.  It is also available from ftp://sunsite.unc.edu and its mirrors, as
of this writing in the /pub/Linux/Incoming directory.


<hr> <center><address><a href="http://vax2.rain.gen.mo.us/~layers/">Larry
Ayers&lt;layers@vax2.rain.gen.mo.us&gt;</a></address></center><br>

<hr>

<A NAME="backpage"><H1><IMG SRC="../gx/banner.gif" alt="Linux
Gazette"></H1></A>
<H1>The Back Page</H1>

<H5>Copyright (c) 1996 Specialized Systems Consultants, Inc.<br>
For information regarding copying and distribution of this material see the
<A HREF="../ssc.copying.html">COPYING</A> document.</H5>
<P>

<HR>
<H4> Plans </H4>
<P> 
Now that <i>Linux Journal</i> has taken over the <i>Linux Gazette</i>
from John Fisk,
I plan to post a new issue of the <i>Gazette</i> sometime during the
first week of each month -- preferably on the first.
Whether or not I can make this happens will 
depend on writers getting articles to me each month, AND on time
constraints due to my other projects for <i>Linux Journal</i> and SSC. 
<P> 
I wish to keep the Gazette both helpful and fun as John has over the past
year. And, of course, it will also remain free. Any ideas and suggestions,
as well as criticisms that you might have, for improvements to the Gazette
will be welcome. Most of all, I will welcome your contributions -- after
all, without you <i>Linux Gazette</i> would disappear.
<P> 
I'd also like to thank our webmaster, Michael Montoure, for his invaluable
help in checking the HTML and designing neat graphics.
<P> <hr> <P> 
<H4> Not Linux </H4> 
<P>
In the hopes that everyone had a nice Labor Day weekend, I thought I'd tell
you about mine. I went camping in Northern Washington and Canada with
my husband, Riley, our son, Keith, and Keith's dog Lucky. We had a grand
time exploring the Okanagan Lake area of British Columbia. In driving to a
campground
on Pennask Lake, we traversed what must be one of the worst roads in Canada
-- thank goodness for high clearance vehicles! Keith was ahead of us in his
4 wheel drive truck, and was really bouncing through the ditches and holes
in the road. Our camper was bouncing and swaying enough to make me worry
that it might cause us to tip over -- Riley assured me it would not. At any
rate, after
30 minutes of hard driving, we reached the beautiful lake only to find the
campground full of fishermen. So we headed back out that wonderful road,
looking for side roads, and anyplace that we might camp. We found a place
down a side road that had only two huge ditches across it, and set up camp.
No bears showed up to spoil the fun, so we had a very
good time playing in an isolated spot away from other campers. 
Canada is certainly
a very beautiful country, and the British Columbians very friendly. This
jaunt was not our first trip to Canada nor will it be the last.
<P> 
If you would like some personal information about me, clicking on my
name below will take you to my home page. It's not very jazzy at the moment,
but I'm looking for the time to fix it up.
<P> <hr> <P> 
<A HREF="http://www.ssc.com/ssc/Employees/Margie/margie.html">
Marjorie L. Richardson</a><br>
Editor, <i>Linux Gazette</i> <A HREF="mailto:gazette@ssc.com">gazette@ssc.com</a>
<p><hr><p>
This page written and maintained by the Editor of <I>Linux Gazette</I>,
<A HREF="mailto: gazette@ssc.com"> gazette@ssc.com</A>
<P> 

</BODY>
</HTML>