Linux-Misc Digest #389, Volume #19 Tue, 9 Mar 99 20:13:10 EST
Contents:
Re: Can Linux use 36-bit Xeon addressing? (Tomasz Korycki)
Re: Problems with ALSA sound driver (Jeff McCune)
Printing problem (Len Cuff)
mgetty fax receive problem ("Peter Caffin")
Re: so, how is gnome 1.0, guys? <troll> (Matthias Warkus)
Re: what "rc" scripts exist for linux? (Peter Samuelson)
Re: 2.2 Sound code (Michael McConnell)
Re: KDE? Gnome? ... confused (Matthias Warkus)
Help me with experiment (propsync)
Re: Moving directories accross partitions (Jess C. Gehin)
Re: Help with Network Printing. Pleeessse !!! (Kevin Hamilton)
Re: best offline newsreader? (Matthias Warkus)
Re: so, how is gnome 1.0, guys? <troll> (Matthias Warkus)
Re: how to set kernel limits? (Seth Van Oort)
----------------------------------------------------------------------------
From: Tomasz Korycki <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux,comp.os.linux.hardware
Subject: Re: Can Linux use 36-bit Xeon addressing?
Date: Tue, 09 Mar 1999 13:22:17 -0500
"Stuart R. Fuller" wrote:
>
> brian moore ([EMAIL PROTECTED]) wrote:
> : >
> : > Yes. And Your point, as related to "the last holdout from basing their
> : > systems on Unix concepts is Microsoft" bit? Mind You, if You look deep
> : > enough into NT architecture, You'll see.... VMS!
> :
> : VMS is based on Unix?
> :
> : Very interesting news indeed.
>
> It might be interesting, but it is certainly wrong news. If you read the
> paragraph above, it implies that the NT architecture is based on VMS.
>
> Stu
It is. Just look who the main architect was and who he brought with him
to create NT. Then look at all the arch details (at kernel level): NT
processes are the same as VMS processes; there are 32 priority levels in
process scheduler of both: 16 highest ones are realtime (or fixed)
priorities, the others are dynamic; the schedulers never lower
processes' priority; lightweight user threads in NT4.0 are
indistinguishable from VMS's threads; virtual memory address maps are
almost identical; both rely VERY heavily on memory-mapped files, using
copy-on-write (VAX has limited support for copy-on-demand, whereas x86,
MIPS and Alpha do not, so...); The concept of working-set is identical,
along with a lot of algorithms; both I/O managers are layered-driver
model, stackable and loadable; both represent resources as objects to be
managed by Object Manager, implementing uniform reference counting and
accounting; security based on same idea (DACL); then: look at
Performance Monitor and MONITOR, or NT backup tool and BACKUP. Devil's
in the details, as they say. Those arch ideas may be common among more
than those two OS's, but when You analyse (I recommend "VAX/VMS
Internals and Data Structures", Digital Press: reads like NT design
document...) actual design assumptions, algorithms, etc, You just can't
help it but conclude that VMS 5.0 and NT3 were born in the same
brain(s). Final argument, in the form of a diagram for both Executives:
Applications Applications
/ | |
Win32 | POSIX OS/2 DOS/Win16 |
======= | ===== ==== ========== |
| | User Mode |
===|=======|=========================================================|=========================
| | Kernel Mode |
=============================================
=========================================
| System Services | | System Services
|
|-------------------------------------------|
|----------------------------------------|
|---------- Process Management, I/O Mgmt, | |------------|
Process Management, |
| Device | Memory Management,...... | | Device |
I/O Management, |
| Drivers | | | Drivers | Memory
Management, |
--------------------------------------------|
|----------------------------------------|
| Harware Abstraction Layer | | Hardware
Abstraction Layer |
|-------------------------------------------|
|----------------------------------------|
'nough said?
------------------------------
From: Jeff McCune <[EMAIL PROTECTED]>
Crossposted-To: linux.dev.sound
Subject: Re: Problems with ALSA sound driver
Date: Tue, 09 Mar 1999 18:41:52 -0500
Hi there, I have an AWE 64 ISA myself, and desperately want to install
the ALSA drivers. However, I can't find any decent documentation. The
docs I have are rather vague to say the least. Could you tell me where
you got the information on how to install the ALSA drivers? I'd be much
appriciated.
David Guertin wrote:
> Hi folks,
>
> In an effort to get my cheap old SB16 card working in full-duplex, I
> replaced the standard kernel sound drivers with the ALSA drivers. The
> install procedure went OK, and the modules are all loaded. (I have to
> load the snd-pcm1-oss module by hand, but that's another story...)
>
> I can cat .au files to /dev/audio, and I can listen to .au and .aifc
> files using DAP. I can also record sound in DAP.
>
> But: playing wav files with wavplay gives me:
>
> Invalid argument: Audio block size (1024 bytes)
>
> And trying to run multitrack for multitrack recording gives me:
>
> Cannot open /dev/sndstat. No sounddriver loaded.
>
> So I tried `cat /dev/sndstat' and I get:
>
> cat: /dev/sndstat: Operation not supported by device
>
> Also, I can no longer play audio CD's. There's no error, they're just
> silent.
>
> Have I missed something, or are these just ordinary bugs that go with
> early-in-development software?
>
> Regards,
> --
> Dave Guertin
> [EMAIL PROTECTED]
------------------------------
From: Len Cuff <[EMAIL PROTECTED]>
Subject: Printing problem
Date: Tue, 9 Mar 1999 19:51:24 +0000
Reply-To: Len Cuff <[EMAIL PROTECTED]>
I have SuSE 6 installed and use apsfilter/Ghostprint5.5 to print to my
Epson600 Color. It works OK but when I print for example lpr /etc/hosts,
it always prints in landscape when I want portrait. Seems that landscape
is the default somehow. When I use Star Office it prints fine but
WordPerfect 8 goes back to landscape again even though I set up the
printer as portrait ! How can I set it to default to portrait all the
time ??
Cheers,
Len
------------------------------
From: "Peter Caffin" <[EMAIL PROTECTED]>
Crossposted-To: de.alt.comm.mgetty
Subject: mgetty fax receive problem
Date: 9 Mar 1999 19:18:23 GMT
From: "Peter Caffin" <[EMAIL PROTECTED]>
Subject: mgetty fax receive problem
Newsgroups: comp.os.linux.misc
Organization:
Summary:
Keywords:
Well, I've been messing around with mgetty+sendfax and I've got the fax
spooling working (it should be fairly easy to make an /etc/printcap
script -- something to work on later)..
However, I haven't been able to get the fax receive part working at
all. Any suggestions would be gratefully appreciated :).
System info:
Linux (Debian) kernel 2.2.2 on an i586
mgetty: experimental test release 1.1.14-Apr02
Hayes Accura 33.6 (Class 2.0 compliant)
I'm able to dial into the modem line with another modem and connects to
a login prompt happily. But a fax connection never resolves.
The modem part of my /etc/inittab is:
# Example how to put a getty on a modem line.
#
# T3:23:respawn:/sbin/mgetty -x0 -s 57600 ttyS0
T3:23:respawn:/sbin/mgetty -x5 -C c2.0 -s 57600 ttyS0
The uncommented parts of my /etc/mgetty/mgetty.config are:
debug 4
fax-id
speed 38400
fax-owner root
fax-group uucp
fax-mode 0640
port ttyS0
speed 38400
My /etc/mgetty/dialin.config doesn't have anything uncommented (and so
should have the implicit allow all, if the documentation is right).
I've included the log file below.
--
03/09 10:57:04 yS0 mgetty: experimental test release 1.1.14-Apr02
03/09 10:57:04 yS0 mgetty.c compiled at May 23 1998, 14:16:58
03/09 10:57:04 yS0 user id: 0, parent pid: 1
03/09 10:57:04 yS0 reading configuration data for port 'ttyS0'
03/09 10:57:04 yS0 reading /etc/mgetty/mgetty.config...
03/09 10:57:04 yS0 conf lib: read: 'debug 4'
03/09 10:57:04 yS0 conf lib: read: 'fax-id '
03/09 10:57:04 yS0 conf lib: read: 'speed 38400'
03/09 10:57:04 yS0 conf lib: read: 'fax-owner root'
03/09 10:57:04 yS0 conf lib: read: 'fax-group uucp'
03/09 10:57:04 yS0 conf lib: read: 'fax-mode 0640'
03/09 10:57:04 yS0 conf lib: read: 'port ttyS0'
03/09 10:57:04 yS0 section: port ttyS0, **found**
03/09 10:57:04 yS0 conf lib: read: 'speed 38400'
03/09 10:57:04 yS0 key: 'speed', type=0, flags=2, data=57600
03/09 10:57:04 yS0 key: 'switchbd', type=0, flags=1, data=0
03/09 10:57:04 yS0 key: 'direct', type=3, flags=1, data=FALSE
03/09 10:57:04 yS0 key: 'blocking', type=3, flags=1, data=FALSE
03/09 10:57:04 yS0 key: 'port-owner', type=1, flags=1, data=root
03/09 10:57:04 yS0 key: 'port-group', type=1, flags=1, data=dialout
03/09 10:57:04 yS0 key: 'port-mode', type=0, flags=1, data=432
03/09 10:57:04 yS0 key: 'toggle-dtr', type=3, flags=1, data=TRUE
03/09 10:57:04 yS0 key: 'toggle-dtr-waittime', type=0, flags=1, data=500
03/09 10:57:04 yS0 key: 'data-only', type=3, flags=1, data=FALSE
03/09 10:57:04 yS0 key: 'fax-only', type=3, flags=1, data=FALSE
03/09 10:57:04 yS0 key: 'modem-type', type=1, flags=2, data=c2.0
03/09 10:57:04 yS0 key: 'modem-quirks', type=0, flags=0, data=(empty)
03/09 10:57:04 yS0 key: 'init-chat', type=2, flags=1, data= \dATQ0V1H0
OK AT OK
03/09 10:57:04 yS0 key: 'force-init-chat', type=2, flags=1, data=
\d^P^C\d\d\d+++\d\d\d^M\dATQ0V1H0 OK
03/09 10:57:04 yS0 key: 'modem-check-time', type=0, flags=1, data=3600
03/09 10:57:04 yS0 key: 'rings', type=0, flags=1, data=1
03/09 10:57:04 yS0 key: 'get-cnd-chat', type=2, flags=0, data=(empty)
03/09 10:57:04 yS0 key: 'answer-chat', type=2, flags=1, data= ATA
CONNECT \c
03/09 10:57:04 yS0 key: 'answer-chat-timeout', type=0, flags=1, data=80
03/09 10:57:04 yS0 key: 'autobauding', type=3, flags=1, data=FALSE
03/09 10:57:04 yS0 key: 'ringback', type=3, flags=1, data=FALSE
03/09 10:57:04 yS0 key: 'ringback-time', type=0, flags=1, data=30
03/09 10:57:04 yS0 key: 'ignore-carrier', type=3, flags=1, data=FALSE
03/09 10:57:04 yS0 key: 'issue-file', type=1, flags=1, data=/etc/issue
03/09 10:57:04 yS0 key: 'prompt-waittime', type=0, flags=1, data=500
03/09 10:57:04 yS0 key: 'login-prompt', type=1, flags=1, data=@!login:
03/09 10:57:04 yS0 key: 'login-time', type=0, flags=1, data=240
03/09 10:57:04 yS0 key: 'fido-send-emsi', type=3, flags=1, data=TRUE
03/09 10:57:04 yS0 key: 'fax-id', type=1, flags=3, data=
03/09 10:57:04 yS0 key: 'fax-server-file', type=1, flags=0, data=(empty)
03/09 10:57:04 yS0 key: 'diskspace', type=0, flags=1, data=1024
03/09 10:57:04 yS0 key: 'notify', type=1, flags=1, data=root
03/09 10:57:04 yS0 key: 'fax-owner', type=1, flags=3, data=root
03/09 10:57:04 yS0 key: 'fax-group', type=1, flags=3, data=uucp
03/09 10:57:04 yS0 key: 'fax-mode', type=0, flags=3, data=416
03/09 10:57:04 yS0 key: 'debug', type=0, flags=2, data=5
03/09 10:57:04 yS0 key: 'statistics-chat', type=2, flags=0, data=(empty)
03/09 10:57:04 yS0 key: 'statistics-file', type=1, flags=0, data=(empty)
03/09 10:57:04 yS0 key: 'gettydefs', type=1, flags=1, data=n
03/09 10:57:04 yS0 key: 'term', type=1, flags=0, data=(empty)
03/09 10:57:05 yS0 check for lockfiles
03/09 10:57:05 yS0 checklock: stat failed, no file
03/09 10:57:05 yS0 locking the line
03/09 10:57:05 yS0 makelock(ttyS0) called
03/09 10:57:05 yS0 do_makelock: lock='/var/lock/LCK..ttyS0'
03/09 10:57:05 yS0 lock made
03/09 10:57:05 yS0 tio_get_rs232_lines: status: RTS CTS DSR DTR DCD
03/09 10:57:05 yS0 WARNING: DCD line still active, check modem settings
(AT&Dx)
03/09 10:57:05 yS0 lowering DTR to reset Modem
03/09 10:57:06 yS0 tss: set speed to 57600 (10001)
03/09 10:57:06 yS0 tio_set_flow_control( HARD )
03/09 10:57:06 yS0 waiting for line to clear (VTIME), read:
03/09 10:57:06 yS0 send: \dATQ0V1H0[0d]
03/09 10:57:06 yS0 waiting for ``OK''
03/09 10:57:06 yS0 got: ATQ0V1H0[0d][0d][0a]OK ** found **
03/09 10:57:06 yS0 send: AT[0d]
03/09 10:57:06 yS0 waiting for ``OK''
03/09 10:57:06 yS0 got: [0d][0a]AT[0d][0d][0a]OK ** found **
03/09 10:57:06 yS0 mdm_send: 'AT+FCLASS=2.0'
03/09 10:57:06 yS0 mdm_command: string 'AT+FCLASS=2.0'
03/09 10:57:06 yS0 mdm_command: string 'ERROR' -> ERROR
03/09 10:57:07 yS0 mdm_send: 'AT+FCLASS=2'
03/09 10:57:07 yS0 mdm_command: string 'AT+FCLASS=2'
03/09 10:57:07 yS0 mdm_command: string 'OK' -> OK
03/09 10:57:07 yS0 mdm_send: 'AT+FCLASS=0'
03/09 10:57:07 yS0 mdm_command: string 'AT+FCLASS=0'
03/09 10:57:07 yS0 mdm_command: string 'OK' -> OK
03/09 10:57:07 yS0 mdm_send: 'AT+FAA=1;+FCR=1'
03/09 10:57:07 yS0 mdm_command: string 'AT+FAA=1;+FCR=1'
03/09 10:57:07 yS0 mdm_command: string 'OK' -> OK
03/09 10:57:07 yS0 mdm_send: 'AT+FBOR=0'
03/09 10:57:07 yS0 mdm_command: string 'AT+FBOR=0'
03/09 10:57:07 yS0 mdm_command: string 'OK' -> OK
03/09 10:57:07 yS0 mdm_send: 'AT+FLID=""'
03/09 10:57:07 yS0 mdm_command: string 'AT+FLID=""'
03/09 10:57:07 yS0 mdm_command: string 'OK' -> OK
03/09 10:57:07 yS0 mdm_send: 'AT+FDCC=1,5,0,2,0,0,0,0'
03/09 10:57:07 yS0 mdm_command: string 'AT+FDCC=1,5,0,2,0,0,0,0'
03/09 10:57:07 yS0 mdm_command: string 'OK' -> OK
03/09 10:57:07 yS0 waiting for line to clear (VTIME), read:
03/09 10:57:07 yS0 removing lock file
03/09 10:57:07 yS0 waiting...
03/09 10:59:47 yS0 checking lockfiles, locking the line
03/09 10:59:47 yS0 makelock(ttyS0) called
03/09 10:59:47 yS0 do_makelock: lock='/var/lock/LCK..ttyS0'
03/09 10:59:47 yS0 lock made
03/09 10:59:47 yS0 waiting for ``RING''
03/09 10:59:47 yS0 got: [0d][0a]RING ** found **
03/09 10:59:47 yS0 send: ATA[0d]
03/09 10:59:47 yS0 waiting for ``CONNECT''
03/09 10:59:47 yS0 got: [0d][0a]ATA[0d][0d][0a]NO CARRIER
03/09 11:00:42 yS0 found action string: ``NO CARRIER''
03/09 11:00:42 ##### failed A_FAIL dev=ttyS0, pid=3323, caller='none',
conn='', name=''
03/09 11:00:42 yS0 removing lock file
--: _ _ _ _
_oo__ |_|_ |__ _ | _ |_|_o _ pc at it dot net dot a u |
//`'\_ | (/_|(/_| |_(_|| | || | it.net.au/~pc |
/
------------------------------
From: [EMAIL PROTECTED] (Matthias Warkus)
Subject: Re: so, how is gnome 1.0, guys? <troll>
Date: Tue, 9 Mar 1999 18:37:56 +0000
Reply-To: [EMAIL PROTECTED]
It was the Tue, 09 Mar 1999 15:45:56 GMT...
..and steve mcadams <[EMAIL PROTECTED]> wrote:
> I have no problem with buffering but consider apps to be boneheaded if
> they insist on large fixed-size buffers. 35meg codesize is ridiculous
> for what it sounds like it offers, but it sounds like some of the
> 35meg is large fixed-size buffers. Still ridiculous for the level of
> function we're talking about as I see it. For this level of function
> I'd expect maybe 2-4meg codesize and buffering based on available
> memory. There must be a shitload of unused support code buried in it
> that's waiting for somebody to write apps that use it.
Man, that big number sure confused you. Hey, this is counting the
same chunk of shared libraries at least six times! Look at the RSS
values!
mawa
--
Ich habe noch einen langen Weg zu gehen und viel zu lernen, und ich
denke, das einzige Zeugnis, das ich ablegen kann, mit dem ich
wenigstens versuchen kann, zu zeigen, da� ich kein Troll bin, ist: ich
bin bereit zu lernen. -- mawa, Oktober 1998
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.development.system
Subject: Re: what "rc" scripts exist for linux?
Date: 9 Mar 1999 17:50:43 -0600
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[M Sweger <[EMAIL PROTECTED]>]
> I was just wondering what types of "rc" scripts exist out there for
> Linux and what tar file names they go by? I.E SysV vs. BSD style.
> Presently Redhat seems to use SysV whereas slackware uses BSD. I'll
> assume there is also a version that is a cross between SysV and BSD
> for linux.
Dunno about hybrids. But for another data point, Debian uses SysV.
> I'm trying to rebuild the Linux system from the ground up, but don't
> know the "rc" script package names or which of the ways is best.
It's the `init' package. For Red Hat and Debian, `sysvinit'. I think
the SysV init script style is easier to manage/customize. I also like
Debian's incarnation better than Red Hat's -- RH puts the runlevel
directories in a subdirectory of /etc rather than just in /etc. The
extra level seems unnecessary to me.
(But Digital Unix takes the cake for confusing/counterintuitive. Their
runlevel directories are /sbin/rc?.d/.)
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: Michael McConnell <[EMAIL PROTECTED]>
Subject: Re: 2.2 Sound code
Date: Tue, 9 Mar 1999 19:40:57 +0000
On 9 Mar 1999 [EMAIL PROTECTED] wrote:
> Ewan Dunbar scribbled manically:
> : I've recently tried 2.2.2 and found the sound code awful. I'm going back
> : to 2.0.36 for the time being. It freezes the whole system when it
> : tries to load after a warm boot. Has anyone else noticed this?
>
> A minor "me too"...I had sound working very nicely with 2.0.36,
> yet when I compile 2.2.1 with what I would swear on a stack of Bibles _this_
> high are precisely the same settings, and boot the resulting kernel, sound
> doesn't work. Maybe it's having a cheap generic sound card...I can post
> the relevant boot messages from a 2.2.1 boot if it would help anyone, but I
> can't see why it should be this way.
I had this too. Anyway, I gave up on it, compiled my kernel without sound and
installed ALSA. Works a treat. :)
ALSA is at http://alsa.jcu.cz
-- Michael
Eridani Star System -- The Most Up-to-Date Red Hat Linux CDROMs Available
Email: [EMAIL PROTECTED] http://www.amush.cx/linux/ Fax: +44-8701-600807
------------------------------
From: [EMAIL PROTECTED] (Matthias Warkus)
Crossposted-To: comp.os.linux.setup,linux.redhat.rpm,linux.redhat.misc
Subject: Re: KDE? Gnome? ... confused
Date: Tue, 9 Mar 1999 18:34:52 +0000
Reply-To: [EMAIL PROTECTED]
It was the Tue, 09 Mar 1999 04:57:22 +0000...
..and a <[EMAIL PROTECTED]> wrote:
> "The development of the X window system started in 1984 at the Massachusetts
> Institute of Technology under the auspices of the MIT Laboratory for Computer
> Science and MIT/Project Athena."
Gee, I'd love to have a 800-page book that opens with this paragraph!
_How_The_X_Was_Made_ - sounds good :)
mawa
--
(Speaking of Beethoven's Ode to Joy:) This is certainly great music to
blow things up by!
-- Commentator Dave Nuttycombe, NPR
------------------------------
From: propsync <"NOSPAM propsync"@stratos.net>
Crossposted-To: alt.linux,alt.os.linux
Subject: Help me with experiment
Date: Tue, 09 Mar 1999 23:53:42 +0000
Hello
I have 2 partitions on my harddrive. The first one has the redhat linux
distribution on it. On the second partition, I want to create a bare
bones linux system by copying the necessary files from partition 1 to
partition 2. My goal is to see just how small I can get the operating
system. The first thing I did was to create the filesystem (ext2) on
partition 2. My next move was to add this partition to lilo to enable
it to boot. The third thing I did was to copy vmlinuz to the /boot
directory that I created. When I attempt to boot from partition 2, the
system freezes by saying "cannot find the console" or something like
that. Can anyone help me by specifying what files I need to copy to the
second partition to get it to boot?
thanks
------------------------------
From: Jess C. Gehin <[EMAIL PROTECTED]>
Subject: Re: Moving directories accross partitions
Date: 9 Mar 1999 20:31:43 GMT
Use tar:
tar -cf - from-dir | (cd to-dir; tar -xf -)
gbh wrote:
> I'm trying to move a directory to another partition.
> If I use the mv command I get an error message that
> it can't move accross file systems. What would be
> the best way to move it along with all its sub-
> directories and still keep their permissions and
> symlinks after the move?
>
> Greg
>
>
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: Kevin Hamilton <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.network,comp.os.linux.setup
Subject: Re: Help with Network Printing. Pleeessse !!!
Date: Tue, 09 Mar 1999 18:19:19 -0500
Reply-To: [EMAIL PROTECTED]
Thanks to everyone who posted in response to my question. Really
appreciated. I took bits and pieces of everyone's post and finally got
a valid printcap file to work. I don't exactly know
what I did differently but it finally worked. Just for those who are
interested in what the final
printcap file looked like here goes:
hp8000n|Hewlett_Packard_8000n:\
:[EMAIL PROTECTED]/9100:\
:mx#0:\
:rw:\
:rm=hp8000n.bradman-lake.com:\
:rp=raw:\
:lf=/usr/spool/lpd/hp8000n.log:\
:sd=/usr/spool/lpd/hp8000n:\
:sh:
hp_text|Hewlett_Packard_Text:\
:[EMAIL PROTECTED]/9100:\
:mx#0:\
:rw:\
:rm=hp8000n.bradman-lake.com:\
:rp=text:\
:lf=/usr/spool/lpd/hp_text.log:\
:sd=/usr/spool/lpd/hp_text:\
:sh:
------------------------------
From: [EMAIL PROTECTED] (Matthias Warkus)
Subject: Re: best offline newsreader?
Date: Tue, 9 Mar 1999 18:31:19 +0000
Reply-To: [EMAIL PROTECTED]
It was the Mon, 8 Mar 1999 16:56:08 -0500...
..and Rufus V. Smith <[EMAIL PROTECTED]> wrote:
[discarding elm, exmh, fetchmail, mailx, metamail, mutt, nmh, pine,
slrn, tin, trn, mailcap, procmail, sendmail, ppp, dip,
netscape-communicator, xrn, and xmailbox]
> >No, they can't, and they won't. They all serve a particular,
> >well-defined purpose.
> >
>
>
> Many particular, well-defined purposes that many users will find absolutely
> no use for. I think the "being discarded" was on an individual user basis.
> </serious commentary>
Let's have a look at them:
elm, exmh, mutt, nmh, pine - Mail clients. Don't install them if you
don't need them.
slrn, tin, trn, xrn - Newsreaders. Don't install them if you don't
need them.
fetchmail - POP3->spool mail injector. Don't use it if you can fetch
your mail in a different way.
mailx - Command-line mail client. Leave it alone, scripts need it.
mailcap, procmail, sendmail, metamail - Parts of the mail chain.
You'll never notice them. Distro should configure them.
etc. etc.
What is wrong about them? Why do you want them to be thrown away? And
what exactly do they do that users haven't got a use for?
[schnibble]
> >WRONG. Most Unix software is elegant, doing one thing in the Right
> >Way. Do you really want us to use all-in-one reinvent-the-wheel
> >packages with dumbed down interfaces and all the usual cruft you find
> >on the Windows platform?
> >
> >No man. We've got a better philosophy.
>
> ....and if the rest of the world wasn't so stupid...
Unlike you, we aren't trying to force upon others what we are
accustomed to. We've got an enormous cosmos of very elaborate,
functional software. And now there come some greenhorns from an OS
that's hardly ten years old, and they want us to throw it away.
You'd be annoyed, too.
mawa
--
Check out the Frequently Rehashed Topics on comp.os.linux.advocacy!
<URL: http://home.att.net/~nishk/frt-html/frt.html>
<URL: http://members.tripod.com/~Nishk/frt.html>
Report dead links to [EMAIL PROTECTED] or [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Matthias Warkus)
Subject: Re: so, how is gnome 1.0, guys? <troll>
Date: Tue, 9 Mar 1999 20:22:40 +0000
Reply-To: [EMAIL PROTECTED]
It was the Mon, 08 Mar 1999 23:15:13 GMT...
..and steve mcadams <[EMAIL PROTECTED]> wrote:
> [Posted & mailed, snipped, quoted is ">"]
> [EMAIL PROTECTED] (Matthias Warkus) wrote:
>
> >A nice panel, nice configurability, very good looks, some small apps
> >(gncal, gEdit etc.) that are really great, more big great apps coming
> >up.
>
> Sorry, this is going to sound stupid, but after trying KDE basically I
> quit paying any attention to the talk about gnome. I'm currently
> running fvwm2 with ktdesk. Not sure if ktdesk is its name;
You mean TkDesk. It's slick, yes.
> since I
> gave my Linux box away I've been in a dual-boot situation again until
> I get some chassis modifications made on the for-crap Packard-Bell
> system that I'm building as my new Linux-box. Anway, I haven't run
> Linux lately since I'm mostly doing things that NT is already set up
> for. So having made a short story long, this may be stupid to ask,
> but does gnome have its own window-manager, or does it run under
> whatever one is "it"?
Gnome hasn't got a particular WM, it runs under whatever WM you like
best (so does KDE; kwm is not obligatory). It works better with a
Gnome-compliant WM like Enlightenment, though.
BTW, Window Maker is both KDE- and Gnome-compliant.
> Do apps change in any particular way when
> running under gnome, for example would the gimp look and act the same
> with the same window decorations etc?
It's the window manager that does the window decorations.
> >> Is it a fat pig like KDE,
> >
> >I did some memory statistics recently, it consumes far less memory
> >than KDE in a similar situation. To be more concrete: with exactly
> >equivalent applications running, KDE with kwm consumes 53,196 KB of
> >memory with an RSS of 8,340 KB, while Gnome with FLWM only uses 35,612
> >KB with an RSS of 4,884 KB.
>
> <sigh> Only 35meg. Well, moving right along...
You misunderstood me. Those 35 megabytes are all the application sizes
summed up - counting shared libraries several times. The RSS (resident
set size) is the actual memory real estate it consumes.
> >As you can see, the stuff is smaller, and the percentage of RSS is
> >about two percent less - sounds good to me.
>
> Sounds good to me too, what's RSS mean?
As I said: resident set size. Basically, the RSS is the SIZE minus the
SHRD value (do "ps m" to see what this means) minus the amount of
swapped-out pages.
Shared memory is shared between applications, that means, if you run
ten Gnome applications, you get a total SIZE of more than forty
megabytes, while in reality, they all share one copy of the Gnome and
GTK+ libraries.
> > Note that Gnome with
> >Enlightenment, while still smaller than KDE, is significantly worse.
>
> What does "significanly worse" mean in english please? Slower,
> uglier, less reliable, more likely to burn your eggs, what?
Simple. Enlightenment is window manager which uses A LOT of resources.
And it's still somewhat alpha. But it's cool, yes.
> >And note that I compiled Gnome for maximum speed and not for minimum
> >memory usage (i.e. I did -O7 -march=pentium).
>
> Frankly it seems to me that one should be able to compile something
> that provides the level of functionality you're describing for maximum
> speed and still have it come out smaller than 35meg, but what I don't
> know about Linux and X would fill a small encyclopedia.
Right. 35 megs of total SIZE (I shouldn't have mentioned that number,
I knew it was only going to confuse people) doesn't mean a memory
footprint of 35 megs. Actually, what I should have done is sum up the
SIZE and deduct the SHRD values.
Anyway, I'll have to redo the statistics for Gnome 1.0.
[schnibble]
> >A nice panel, nice configurability, very good looks, some small apps
> >(gncal, gEdit etc.) that are really great, more big great apps coming
> >up.
>
> A configurable panel-thing that's pretty. What does the panel
> control?
The Gnome panel is like kpanel, only better - i.e. a bar with menu
buttons, launcher buttons and drawers on it, and furthermore, it can
embed little applets into itself. IMHO, it is way cooler than the KDE
panel :).
> Is it like YaST and the non-working RedHat X-based
> rpm-manager?
Nope; you're confusing `panel' with `control panel' here. Gnome has
got a nice Control Center, too; there is even a Gnome frontend to Red
Hat linuxconf.
BTW, methinks I totally forgot to mention the extremely slick Gnome
Midnight Commander file manager, which is still heavily under
development, though.
mawa
--
It is a common characteristic of all democracies that intelligence is
so highly regarded as to exempt the holder from the cares of office
-- major@pyrmania
------------------------------
From: Seth Van Oort <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.networking
Subject: Re: how to set kernel limits?
Date: Wed, 10 Mar 1999 00:25:14 +0000
David Henry wrote:
>
> How do I know how many processes my system will allow? how many open
> sockets?
> If not enough, how can I change the limits?
>
> I've scanned the LDP pages and my copy of System Afmin Guide but can't
> see anything there.
>
> Please help. I'm planning an application that gets input from 250
> network devices. My idea was to have a separate thread and socket for
The default max nr of tasks is 512, but you can set it up to 4092 or
4090 with APM configured (stated in include/linux/tasks.h, this is where
you can set it also). I would think select() would give a lot better
performance. Since the normal max fds per task is 256 maybe split it
over two tasks. Just wondering if you're aware that linux threads are
always separate tasks.
Seth
> each device. Is that practical?
>
> David Henry
> [EMAIL PROTECTED]
> ICQ 23144758
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
