Linux-Misc Digest #170, Volume #27               Tue, 20 Feb 01 03:13:02 EST

Contents:
  Website Users + Admins ("Patrick Nagurny")
  Looking for something Lin (Mike Mcclain)
  Re: permissions (Mike Mcclain)
  "HOSTNAME=" problem with DSL DHCP ("Tom Edelbrok")
  Re: Gzip error when extracting Kernel 2.4.1 ("muzh")
  insmod (Mike Wilson)
  Re: permissions ("Peter T. Breuer")
  Re: Shell access for windows users? (David E. Fox)
  Re: What is the command to rebuild KDE's Menus? (David E. Fox)
  Re: "HOSTNAME=" problem with DSL DHCP (E J)
  Re: Looking for something Linux ("Peter T. Breuer")
  Re: insmod (John Dixon)
  Re: fastest SCSI CDROM drive (Robert Heller)
  Re: hard disk problem ("Eric")
  Re: Size of LINUX ("Peter T. Breuer")
  Re: replacing startx with startx -- +xinerama (Chris Elvidge)
  Re: fastest SCSI CDROM drive (Gerald Willmann)
  Re: MS to Enforce Registration - or Else ("Peter T. Breuer")

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

From: "Patrick Nagurny" <[EMAIL PROTECTED]>
Subject: Website Users + Admins
Date: Mon, 19 Feb 2001 23:58:27 -0500

I have mydomain.com, which has different people working on different parts
of the website.

Branching out from it are directories:

mydomain.com/usr1
owned by usr1 group client

mydomain.com/usr2
owned by usr2 group client
etc..

How can I make several "admin" accounts that have read/write access to these
directories.  Up until now, I have been relying on root for this, but I
would like to be able to give other "special users" adminstrative access to
these directories.  Is this possible?

Thanks,
Patrick



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

From: [EMAIL PROTECTED] (Mike Mcclain)
Subject: Looking for something Lin
Date: Tue, 20 Feb 2001 06:04:40 GMT

Howdy,
    Perhaps this excerpt from a Fido Linux message by 
John Coombes will help.

I was looking around on the Linuxberg site other
day and they have a whole section called "mini" and
they have at least "38" minimal Linux's there, which
vary from single floppy type to Zipslack and the like.
Perhaps one of those would be what your looking for?

Enjoy,
MiKe

-=> [EMAIL PROTECTED] wrote to ALL <=-

 UL> I want to be able to install the most basic part of Linux and explore
 UL> it for what it is. Then add to it. Man pages, command-line apps,
 UL> XFree86, X apps, RPM support, etc... I want the experience.
 
--- MultiMail/Linux v0.31

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

From: [EMAIL PROTECTED] (Mike Mcclain)
Subject: Re: permissions
Date: Tue, 20 Feb 2001 06:04:40 GMT

Howdy,
    I'm sorry I don't understand the connection between executable
and browsable. To me browsable means readable, NO? /mc can certainly
be read by root.
I have no idea what permissions "nobody" runs with. How do I tell?
And are you saying that /var/spool/cron/crontabs/nobody runs with
user "nobody" permissions rather than with root permissions?
TIA, 
Mike

-=> [EMAIL PROTECTED] wrote to ALL <=-

 >Since it's all owned by root, why the denial?
 >

 _M> Perhaps because you made the directory /mc not executable (browsable)
 _M> by _anyone_ including the owner? Also, does "nobody" have root
 _M> permission?

 _M> Mark Bratcher
 
--- MultiMail/Linux v0.31

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

From: "Tom Edelbrok" <[EMAIL PROTECTED]>
Subject: "HOSTNAME=" problem with DSL DHCP
Date: Tue, 20 Feb 2001 06:13:34 GMT

I have just gotten set up with TELUS DSL in BC Canada. My W98 and WinNT work
perfect using DHCP, although in the instructions they ask you to remove the
domain name and DNS server and WINS entries from your TCP/IP setup.

But when I boot under Linux it hangs on eth0 every time, then times out.

I discovered what is causing the problem: "/etc/sysconfig/network" contains
the line "HOSTNAME=blade.mville.com", (everything works fine if I just
delete this one line). But if I delete the line then my host name shows up
as a bunch of gibberish at the Linux command line, ie: "allng675yaud#".

So I put in "HOSTNAME=blade", but then it won't bring up eth0!

I've tried both pump and dhcp daemon. Both give the same problem. How can I
keep a host name but also bring up eth0 with the DSL DHCP?

Thanks



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

From: "muzh" <[EMAIL PROTECTED]>
Subject: Re: Gzip error when extracting Kernel 2.4.1
Date: Tue, 20 Feb 2001 19:49:45 +1300

Your downloaded file may be corrupt -- this happens!  Especially with big
files -- it only needs a byte or two wrong and the decompressor can't
decompress --
Try downloading and tar-xzvf again
Also, make sure the Windows program you are using doesn't do funny things
to the .tar.gz file -- try using another program -- IE (puke!) perhaps?
Or perhaps try the .tar.bz2 version -- you must use bunzip2 or tar -xyvf
or tar -xIvf to decompress this --

Recently, the keys of [EMAIL PROTECTED]  's computer randomly danced and
produced <96rqrv$lv34f$[EMAIL PROTECTED]> :

> I'm having problems getting Kernel 2.4.1 installed on my machine: Red
> Hat Linux 
> 6.1 with Kernel 2.2.12-20.  I've been following the instructions
> provided at 
> kernel.org, so here's what I've been doing:
> 
> 1) Downloaded `linux-2.4.1.tar.gz' from ftp.kernel.org.  Since my
> winmodem is 
> not configured, I've downloaded this file under Windows 98 SE, using AOL
> and  the ftp utility that comes with DOS.
> 
> 2) Booted Linux using LILO, mounted my Windows partition, and proceeded
> to move 
> the `linux-2.4.1.tar.gz' file from that partition to `/usr/src'.
> 
> 3) Deleted the `/usr/src/linux' directory.
> 
> 4) Gave the command `tar xfv - | gzip -cd linux-2.4.1.tar.gz' at an
> xterm.
> 
> 5) After about a minute of seeing the code fly by, I got the following
> message:
> 
> `gzip: linux-2.4.1.tar.gz: invalid compressed data--format violatedexi'
> 
> Please help me with this problem.  I don't know if I'm doing something
> wrong;   I'm guessing it's not related to the actual file, because I've
> downloaded it  from several ftp/http sites and I still get the same
> error when decompressing  the file.  I'm wondering if Windows does
> something to it after the download  that makes it unusable under Linux. 
> Perhaps it's related to a problem with the  Linux gzip and/or tar
> utility.  I would very much appreciate a response.  Thank  you for your
> time.
> 
> PS - If you need any more system details, soft/hardware, please don't
> hesitate  to contact me.  I really need to pull this off.
> 


-- 
Never trust a man in a suit

cll

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

From: Mike Wilson <[EMAIL PROTECTED]>
Subject: insmod
Date: Tue, 20 Feb 2001 14:52:16 +0800

Hi All,
        I am having problems installing a module, I use
/sbin/insmod mobitex.o

but I get this error
mobitex.o: unresolved symbol waitqueue_lock

I have Red Hat 6.0 ...what do i need to do to get this working?

Mike


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

From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: permissions
Date: Tue, 20 Feb 2001 07:46:23 +0100

Mike Mcclain <[EMAIL PROTECTED]> wrote:
>     I'm sorry I don't understand the connection between executable
> and browsable. To me browsable means readable, NO? /mc can certainly

No .. no exactly. 

But I agree, people have largely forgotten the difference, and
it's not clear if the distinction is currently correctly implemented.
If a directory is browsable "x", then you can ls -l on any *named*
directory entry, i.e. "ls -l dir/foo", but can't do "ls -l dir".
For the latter the directory must be readable "r".

OTOH, if it's +r only, then you can't cd into it. It needs to be
+x for that, but you won't be able to see anything when you're in it
unless you know what you're looking for.

Think of the directory as a file with a list of filenames.

> be read by root.

> I have no idea what permissions "nobody" runs with. How do I tell?

nobody *runs with* permissions of any kind. Nobody in particular has
none! But people do *have* permissions depending on what groups they're
in.

> And are you saying that /var/spool/cron/crontabs/nobody runs with
> user "nobody" permissions rather than with root permissions?

Of course. It runs as nobody.

>  _M> Perhaps because you made the directory /mc not executable (browsable)
>  _M> by _anyone_ including the owner? Also, does "nobody" have root
>  _M> permission?

I'm not sure what he meant by the last sentence. Perhaps he was
asking if nobody is a member of wheel ;-), or if the cronjob runs
a setuid root executable. Probably the latter.

Peter

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

From: [EMAIL PROTECTED] (David E. Fox)
Subject: Re: Shell access for windows users?
Date: Mon, 19 Feb 2001 22:42:44 -0800


On Mon, 19 Feb 2001 15:53:41 -0800, Matt O'Toole <[EMAIL PROTECTED]> wrote:
>I want to give shell access to some Windows users.  I dont want to use
>telnet.  Is there a free ssh client available for Windows?

Give PuTTY a try - http://chiark.greenend.org.uk/~sgtatham/putty/ .
It's small - fits on a floppy disk and does a fine job - it's what
I use if I'm at a Windows to ssh into my linux box here at home which
is on a dsl line.

>Matt O.

-- 
========================================================================
David E. Fox                              Thanks for letting me
[EMAIL PROTECTED]                            change magnetic patterns
[EMAIL PROTECTED]               on your hard disk.
=======================================================================

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

From: [EMAIL PROTECTED] (David E. Fox)
Crossposted-To: alt.os.linux.mandrake
Subject: Re: What is the command to rebuild KDE's Menus?
Date: Mon, 19 Feb 2001 22:47:43 -0800

On 12 Feb 2001 21:42:27 GMT, Jayfar <[EMAIL PROTECTED]> wrote:
>I had done it once before, after updating KDE to 2.0, but I neglected to
>make a note of the command.  I'm looking for the console command to
>rebuild KDE's menus. 

try as root: 'update-menus -v'.

>Jayfar


-- 
========================================================================
David E. Fox                              Thanks for letting me
[EMAIL PROTECTED]                            change magnetic patterns
[EMAIL PROTECTED]               on your hard disk.
=======================================================================

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

From: E J <[EMAIL PROTECTED]>
Subject: Re: "HOSTNAME=" problem with DSL DHCP
Date: Tue, 20 Feb 2001 07:00:35 GMT

The only solution I found that work is that you don't bring up the eth0 during
bootup and it will revert to
the hostname that you give it.
After everything is up except for the network then bring up eth0
$ su -
password: <secret>
# dhcpcd


In a new window
 # tail -f /var/log/message # check to see dhcpcd is ok and not timed out.


I have tried to edit /etc/sysconfig/network-scripts/ifup to comment out dhcpcd
from picking up the
hostname from Telus' dhcp server but no luck.

Tom Edelbrok wrote:

> I have just gotten set up with TELUS DSL in BC Canada. My W98 and WinNT work
> perfect using DHCP, although in the instructions they ask you to remove the
> domain name and DNS server and WINS entries from your TCP/IP setup.
>
> But when I boot under Linux it hangs on eth0 every time, then times out.
>
> I discovered what is causing the problem: "/etc/sysconfig/network" contains
> the line "HOSTNAME=blade.mville.com", (everything works fine if I just
> delete this one line). But if I delete the line then my host name shows up
> as a bunch of gibberish at the Linux command line, ie: "allng675yaud#".
>
> So I put in "HOSTNAME=blade", but then it won't bring up eth0!
>
> I've tried both pump and dhcp daemon. Both give the same problem. How can I
> keep a host name but also bring up eth0 with the DSL DHCP?
>
> Thanks


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

From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Looking for something Linux
Date: Tue, 20 Feb 2001 08:16:12 +0100

Jean-David Beyer <[EMAIL PROTECTED]> wrote:
> "Peter T. Breuer" wrote:
>> I've simply never been able to work out how to resize windows in gnome.
>> Choose resize from the menu, and the cursor changes shape OK. Go to a corner
>> or edge with it and nothing happens. So give up and click, and it reverts
>> to normal.

> I wonder what version of GNOME/Enlightenment you are using. I am

It's not my machine(s). It appears to be RH 6.1.

> using the following stuff:

There is no rpm dbase, but then they're clients. I wouldn't swear to
the window manager being enlightenment. I have enlightenment on my own
machine, and although it IS incomprehensible, it looks prettier and
behaves better (just a silliness with large menus overflowing the
window and also displacing the cursor into a submenu when you click,
which are, however, fatal in combination).

> valinux:jdbeyer[~]$ rpm -qa | grep gnome
> gnome-audio-1.0.0-8
> gnome-libs-1.0.55-12

THe best guess I can make is from enlightenments libs:

ldd `which enlightenment`
        libFnlib.so.0 => /usr/lib/libFnlib.so.0 (0x4001a000)
        libttf.so.2 => /usr/lib/libttf.so.2 (0x40022000)
        libesd.so.0 => /usr/lib/libesd.so.0 (0x40037000)
        libaudiofile.so.0 => /usr/lib/libaudiofile.so.0 (0x4003e000)
        ...

...
-rwxr-xr-x   1 root     root        27769 Apr 19  1999 /usr/lib/libesd.so.0.2.10
...

and also gnomelibs:

-rwxr-xr-x   1 root     root       100860 Jul 21  1999 /usr/lib/libgnome.so.32.3.3
-rwxr-xr-x   1 root     root        56982 Jul 21  1999 /usr/lib/libgnorba.so.27.1.5

> What I run is a an Enlightenment theme called VAmetal (provided by
> VA Linux Systems), but it looks quite a bit like the stuff on my
> straight Red Hat 6.0 system. For the drawing of the stuff inside a

enlightenment looks great, but has fatal flaws that are deadly to
newbies. I know, because I had it installed as default window manager
in the labs for a few weeks until howls of protests from the people
taking labsessions forced me to switch the default back to kde, which
novices can manage without too much problem. Actually, the requests
were for fvwm95-2 to be made default, but I refused!

As I said, I have enlightenment on my own machine. At the size screen I
have and the fonts I have set, I never had basic problems with it, but
on screens 1024x764 it was a disaster.

> that I choose not to use.) At the top right of each window are three
> circles (instead of Microsoft squares) One with _, one with a
> rectangle, and one with an X. The _ iconifies the window. The
> rectangle one does different things depending on which mouse button
> you use, but the left one either expands the window to occupy the
> entire screen, or puts it back to normal. The x one will close the
> window if you left-click it, and if you right click it, it will kill
> the process with a -9 (or something like that). Grabbing a side of a
> window with the left mouse button allows changing the width, the

I believe I tried grabbing. I am honestly not sure which manager
is running in the defauolt gnome sessions on these machines. It's
not afterstep and it's not kde and it's not fvwm*. Where does gdm get
it's startup menu from? I can't see a gdmrc. xinitrc is uninformative.

Ahhh .. they run prefdm. And no manual entry. prefdm runs gdmgreeter.
No manual entry either. Grrr. 

> bottom or top allows changing the height, and a corner allows both
> at once. IIRC, these features did not work when I was running fvwm
> or lesstif (I forget which, or maybe they go together; I forget.
> That was with Red Hat Linux 5.0, where the things in the top right
> were there but did not work.)

> Also, if you right click on the top bar of a window, you get a menu
> with lots of options.

It's that menu's "resize" option that gives me the new cursor shape,
that however doesn't affect the window when I move to the side or
corner with it.

>> Whaaaaat?
>> 
>> What else can I do except move or click? Drag? It seems to revert at first
>> click.

> You can drag if you left-click the top bar of a window and slide the
> thing around, then release the left mouse button.

I meant, maybe I'm supposed to "resize" then drag a corner or side.
But no, the cursor reverts to normal on pressing the button. Any
button.

> If you do not get this kind of stuff, I would be glad to e-mail you
> the rpms you ask for, but at 56.6k or less, you would probably be
> better off getting the stuff from Red Hat or VA Linux.

It's not my system(s). I deal with them by choosing "afterstep" from
the entry menu.

> http://www.redhat.com/apps/download/
> 
>http://www.valinux.com/software/vaload/6.2/?session_hash=6fedd0c8ff766b45d963fbb71a5ec590

> The thing that annoys me most about emacs is that I have never
> figured out how to do something like this vi command:

> 1,$g/pattern1/s/pattern2/complex thingie, a function of stuff found
> by pattern2/

> So I do those things in vi (or sed sometimes, when appropriate).

Yes, that's precisely what people call me in for. One can use sed, but
sed isn't adequate when you have to sequence many such steps.

>> ALL : ALL except 127.
>> 
>> Nicer.
>> 
> Perhaps so, if you like the parsimonious approach (as I usually do),
> but I did not do that with my machines. My policy is that nobody can
> do anything if I say ALL : ALL in hosts.deny. I never put anything

I was just being careful .. there might have been a dodgy moment
for the user who edited the hosts.deny to say ALL:ALL ! He needed
to edit the hosts.allow first.

Peter

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

From: John Dixon <[EMAIL PROTECTED]>
Subject: Re: insmod
Date: Mon, 19 Feb 2001 23:30:18 -0800

Mike Wilson wrote:
> 
> Hi All,
>         I am having problems installing a module, I use
> /sbin/insmod mobitex.o
> 
> but I get this error
> mobitex.o: unresolved symbol waitqueue_lock
> 
> I have Red Hat 6.0 ...what do i need to do to get this working?
> 
> Mike

Is the module compiled for the kernel you are using?  i.e. Have you
upgraded your kernel without recompiling support for this particular
module?  Just a thought...

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

From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: fastest SCSI CDROM drive
Date: 20 Feb 2001 01:09:27 -0600

  John Hunter <[EMAIL PROTECTED]>,
  In a message on 19 Feb 2001 14:42:58 -0600, wrote :

JH> I'm looking for a recommendation for the fastest/best for the price
JH> SCSI CDROM drive.  I noticed on pricewatch that the IDE drives go up
JH> to 72x and the SCSI ones only go to 40x.  Why is this?  Are the SCSIs
JH> competitive because of the faster bus?

These Nx numbers for CDROM drives are somewhat deceptive.  These are
*instantaneous* transfer rates and are a function of rotation speed.  There
are two major gotchas for ALL CDROM drives: 

1) It is not reasonable to spin up a CDROM and *leave it spinning*
indefinitely (like a hard drive).  Because the CDROM is a removable
device and because of the *mechanical* constraints this is just not
possible.  Also, there is this *Law*, enacted (no repeal possible), by
this dude Newton, which precludes instantaneous transition from a
stationary CDROM to one spinning at 40x or 72x times normal speed, so
there is a large (huge at 72x) spin up latency.

2) CDROM drives are not like hard disks and don't use the 'Wide/Fast'
interface (either ATA/Dma66 or SCSI-3).  Thus the continuous data
throughput is limited.  Yes, while a 72x IDE CDROM can *burst* a
*small* file at 72x (after several *seconds* spent spinning up),  it
cannot sustain this for long, so the larger the file the *slower* the
*average* transfer rate, down to about 4x for IDE.  SCSI CDROM drives
will bottom out at about 8x (the SCSI bus is better than PIO IDE). 
Faster drives can possibly be worse -- spin up (several seconds), start
bursting, flood bus, spin *down* (several seconds), wait for bus to
clear, spin up again (several *more* seconds), start bursting again,
flood bus again, spin down *again*, wait for bus to clear, etc.

Bottom line: for serious transfers, anything above about 4x for IDE and
8x for SCSI is effectively silly.  About as useful as a formula-one
race car during rush hour bumper-to-bumper traffic or even in *normal*
downtown traffic: wait at traffic light, stomp on the gas when light
goes green, race one block, slam on the brakes at the red light there,
sit there waiting for the green light, note that the guy with the GEO
Metro going 25-30mph gets to the light just as it is turning green,
every time, which makes you look like a total idiot (as you fund some
Arabian prince's daughter's wedding). Yes, a bright-red sports car is
flashy, but it won't get you there any faster than a GEO Metro.

Oh, the reason SCSI CDROMs stop at a lower 'x' factor is because the
people who *buy* SCSI are not the sort of idiot who will buy a silly
fast car and live in downtown New Your City.  The *only* reason even 40x
SCSI CDROM drives even exist is because 8x CDROM base mechanics are no
longer available, not because there is any real reason to make 40x SCSI
CDROM drives.

JH> 
JH> Seeking recommendations and links to reviews, etc...

I'd ignore the 'x' factor and look instead at price vs. MTBF ratings
(higher speed drives have a *lower* MTBF -- all of the high-speed spinning
up is rough on the bearings and motor).

JH> 
JH> I'm working under RHL 6.2.
JH> 
JH> I will be transferring very large files (1 at approx 600MB) or a small
JH> number (10 at 60MB each).
JH> 
JH> Thanks,
JH> John Hunter 
JH> 
JH>                                  






                                                    
-- 
                                     \/
Robert Heller                        ||InterNet:   [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller  ||            [EMAIL PROTECTED]
http://www.deepsoft.com              /\FidoNet:    1:321/153

Posted Via Nuthinbutnews Premium Usenet Newsgroup Services
==========================================================
          ** SPEED ** RETENTION ** COMPLETION **
==========================================================        
             http://www.nuthinbutnews.com

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

From: "Eric" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: hard disk problem
Date: Tue, 20 Feb 2001 08:39:34 +0100


> ......
> Also If I reset the machine,the BIOS does not recognize the HD.
> And I have to switch off and on the machine again in order to start the
> system again..
>
> Apart from this the HD seem to be fine and works 4 days without any
> problem....
>

Check cabling and jumper settings of this disk

And don't crosspost to so many groups!
Some don't even exist anymore!

Eric



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

From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Size of LINUX
Date: Tue, 20 Feb 2001 08:35:33 +0100

Rolie Baldock <[EMAIL PROTECTED]> wrote:

> On Mon, 19 Feb 2001 07:54:00 +0100, "Peter T. Breuer"
> <[EMAIL PROTECTED]> wrote:
>>Rolie Baldock <[EMAIL PROTECTED]> wrote:
>>> other software servers on the market? I have a couple of 486 DX-33
>>> servers with 6 SCSI disks attached, the servers have 8 Mb of RAM and
>>> their own 50Mb HDDs. I have two workstations running an old DOS which
>>
>>The MHz is your problem here. Those are going to be turtles at
>>processing anything. They may work OK as file servers, but they won't
>>be fast.

> I do not recall saying I had a problem because I don't. I simply asked

You do have a problem, whether you think so or not. Those machines
will have trouble reaching even 1MB/s transfer rate. The isa bus
(8MHz*4) is not limiting, but the cpu MHz is, since they have to
look after the system as well. Since they have to handle the nfs code
.. if that's what you are using .. that will be the limiting factor.
If that's not a concern, that's your affair. But it is the problem.

> if a LINUX based server would run better in the hardware I have in my
> two servers or would it require a horrendous upgrade like the rumours
> I have heard. 

I am not a prophet. I can only  tell you that you need about 8MB of
disk, such memory as you can obtain, and you will have a problem
getting more than 1MB/s out of the system. What "upgrade" are you
talking about? You're changing the entire o/s. That's not an upgrade
.. do you mean hardware? No, you don't have to change anything.

Peter

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

From: Chris Elvidge <[EMAIL PROTECTED]>
Crossposted-To: 
alt.os.linux.redhat,comp.os.linux.hardware,comp.os.linux.admin,comp.os.linux.setup
Subject: Re: replacing startx with startx -- +xinerama
Date: Tue, 20 Feb 2001 11:51:11 +0400

Chris Nelson wrote:
> 
> Hello: This weekend I finally got a working config file for using 2
> monitors on a g400 max.
> 
> My problem however is that right now my etc/inittab file has me going to
> only run level 3. if I change this value to 5, the systems takes me to a
> graphical logon..but with only one monitor. My guess is that the system
> is still strying to run startx as opposed to startx -- +xinerama. Where
> do I need to make a change so this can be automoated? Any advice is
> appreciated...
> 
> P.S. Any idea how I can have the system default to KDE as opposed to
> gnome? Thought I would throw this in as well
> 
> Chris

Have you got a file ~/.xinitrc? Try:

/path/to/xinerama &
/path/to/startkde

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

From: Gerald Willmann <[EMAIL PROTECTED]>
Subject: Re: fastest SCSI CDROM drive
Date: Tue, 20 Feb 2001 08:54:07 +0100

On 20 Feb 2001, Robert Heller wrote:

> 2) CDROM drives are not like hard disks and don't use the 'Wide/Fast'
> interface (either ATA/Dma66 or SCSI-3).  Thus the continuous data

there is a UW SCSI cdrom drive from plextor
but even the narrow version is highly recommended
                                                       Gerald


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

From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy
Subject: Re: MS to Enforce Registration - or Else
Date: Tue, 20 Feb 2001 08:49:46 +0100

In comp.os.linux.misc Robert Surenko <[EMAIL PROTECTED]> wrote:
> In comp.os.linux.misc [EMAIL PROTECTED] wrote:
>> On 16 Feb 2001 23:36:41 GMT, Steve Mading <[EMAIL PROTECTED]> wrote:
>>>In comp.os.linux.advocacy Robert Surenko <[EMAIL PROTECTED]> wrote:
>>>
>>>: That's why Materialism is hopelessly flawed. We all know that JFK
>>>: was shot, but can't repeat the experiment. How do we go about
>>>: proving a historical event.
>>>
>>>WTF does that have to do with materialism?

>>      Whether or not JFK was shot or not is a question of history.
>>      The techniques of historians are not the subject of this discussion.

> I though we were disscussing how does a person "know" something?

> Many have claimed that a repeatable experiment is the only way.

> I've given an example of a fact that can not be repeated and can't
> be proven with the Scientific Method.

Well, we can test your hypothesis. If JFK was not shot, then he must
have been alive after the event, no? In that case the body buried
might not have been his. So  go and do a dna test. They just did
it for the russian royal family remains, it should be a doddle for
jfk. 

But there are simpler tests. Just ask lots of people if they saw jfk
alive after the event. And ask the people who filmed him being
shot how shot he was. You can also ask his wife how she managed to get
legal permission to remarry! Whilst these are not conclusive tests,
they can take the probability of error down to negligable values.
I think he's dead, with a very high degree of certainty.


Peter

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


** 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 by posting to comp.os.linux.misc.

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

Reply via email to