Linux-Misc Digest #86, Volume #20 Thu, 6 May 99 17:13:11 EDT
Contents:
updating glibc ("R.Y.L. Siem")
Re: different timestamp on nt/linux !! ([EMAIL PROTECTED])
Re: GNU reeks of Communism (returning to %252522GNU Communism%252522) (Marco
Anglesio)
Re: GNU reeks of Communism (Marco Anglesio)
Re: Boycott Intel on your own webpage (Andrew Comech)
Re: A Simple Question (Paul Kimoto)
CLASSPATH in RedHat Linux 5.2 ([EMAIL PROTECTED])
Re: GNU reeks of Communism
Re: GNU reeks of Communism (Andrew Carol)
Re: GNU reeks of Communism (Ian Woollard)
meta key doesn't work in xterms with Redhat-6.0 (Michael Keightley)
Re: The GNU Fragrance of Sharing vs. the Stench of Greed (was: GNU reeks of
Communism (really)^ (Robert Krawitz)
Problem unmounting file systems (SIMPLE) (Sellaro)
Re: Q:compiling c++ codes that contains templates (Josh Joyce)
Re: How can X be so slow? ("Mattias Dahlberg")
Re: Is Unix a single user operating system? (Bill Unruh)
simultaneous pap users (brad)
Re: GNU reeks of Communism (Marco Anglesio)
Re: CLASSPATH in RedHat Linux 5.2 (Sam Alexander)
Re: kernel 2.2 (Sellaro)
----------------------------------------------------------------------------
From: "R.Y.L. Siem" <[EMAIL PROTECTED]>
Subject: updating glibc
Date: Wed, 5 May 1999 21:37:16 +0200
I have a SUSE Linux 6.0 version and want to update my glibc to 2.0.7.
When I try to do this with rpm, the machine returns with all conflicting
fileerrors.
Can anyone tell me what to do ?
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: linux.samba
Subject: Re: different timestamp on nt/linux !!
Date: Thu, 06 May 1999 19:04:52 GMT
Ich bin ein Software-Insekt "J�ger-Totschl�ger" Suchen der Pest, das mein NT
4.0 SP-4 (128-bit) Arbeitsstation-System zu automatisch gestattet und die
meisten jederen *.exe unkontrollierbar ver�ndert Datei Datum/Zeit amtierend
da draussen auf einem Samba-Aufschl�ger, jedesmal wenn Ich der *.exe
einfachlich einf�hre. Ich habe viele Artikel um diese Art von Ding nicht
gesehen, und Ich verd�chtige, dass es ein Symptom von mindestens einige
NT/Samba-Entwicklung- Umgebungen auch sein mag. Unser Haupt-Programmierer,
mit meinem Chef, finden Sie es Mittel, zu denken, dass Ich bin verruckt weil
Ich in dem QA Stab bin�, wenn auch sie vor meinem PC stehen und das Problemen
mit ihren eigenen Augen sehen! -Ein �usserstes Symptom: Wenn Ich ein neues,
Einzel-Icon in dem Hintergrund einfachlich schaffe, um ein *.exe zu laufen.
da draussen auf einer Samba Aufschl�gers-Laufwerk, der *.exe Datei
Datum/Zeit-Stempel wird ver�ndert, um meine �rtliche Uhr- und vor Versuchen
zu folgen der *.exe zu laufen. (Noch das Datum/Zeit ver�ndert NICHT bei
Schaffen des Icon �ber Forscher richtig Maus- clicking(?), eine Abk�rzung
-tats�chlich unheimlich zu Schaffen!) So, was tun, denken, Sie? -Versuch die
einfach Icon Sch�pfung testet, mit einige *.exe ablegt, besonders Dateien in
Entwicklung, wie auch Versuch-Lauf einige Einzels *.exe Dateien und aufpassen
Sie der vorher/nachdem Datum-Zeit auch. Mitteilte mir... Danke!
--------------------- Deutch  Englisch -------------------- (Ha! -So
much for translation-ware.)
I'm a software bug "hunter-killer" looking for the pest that allows my NT 4.0
SP-4 (128-bit) Workstation system to automatically and uncontrollably change
most any *.exe file's date/time sitting out there on a Samba Server whenever I
simply initiate the *.exe. I haven't seen many articles about this kind of
thing, and I suspect it may also be a symptom of at least some NT/Samba
development environments. Our main programmer, along with my boss, find it
more expedient to think I'm 'cracked' because I'm in the QA staff..., even
though they stand in front of my PC and see the problem with their own eyes!
F.Y.I. - An extreme symptom: When I simply create a new, single icon in the
background in order to run an *.exe out there on a Samba server's drive, the
*.exe file's date/time stamp is changed to follow my local clock -and prior to
trying to run the *.exe! (Yet the date/time does NOT change when creating the
icon via Explorer right mouse-clicking to Create a Shortcut -really weird!)
So, what do you think? -try the simple icon creation test, with a few *.exe
files, especially files in development, as well as try running some single
*.exe files and watch the before/after date-times too.
Let me know...Thanks
Servas!
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Marco Anglesio)
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.advocacy,gnu.misc.discuss
Subject: Re: GNU reeks of Communism (returning to %252522GNU Communism%252522)
Reply-To: [EMAIL PROTECTED]
Date: Thu, 06 May 1999 15:10:00 GMT
On Wed, 05 May 1999 23:02:49 -0700, jik- <[EMAIL PROTECTED]> wrote:
>Boy, you must not read much....." not because literacy is in and of
>itself an intrinsic good. To an individual, being able to read really
>isn't,"
I'm not sure. Why don't you open a dictionary, look up "intrinsic", and
report back to the class.
Sarcasm aside, does the ability to read give you pleasure, or does the
wide availability of books which you can read give you pleasure? I suspect
that it's the latter. While it may seem like a trivial distinction, it
really isn't; books are produced, like any consumer product, to satisfy
demand.
m.
--
,--------------------------------------------------------------------------.
> | The further I get <
> Marco Anglesio | from the things I care about <
> [EMAIL PROTECTED] | The less I care <
> http://www.the-wire.com/~mpa | how much further away I get <
> | --Robert Smith <
`--------------------------------------------------------------------------'
------------------------------
From: [EMAIL PROTECTED] (Marco Anglesio)
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.advocacy,gnu.misc.discuss
Subject: Re: GNU reeks of Communism
Reply-To: [EMAIL PROTECTED]
Date: Thu, 06 May 1999 15:20:13 GMT
On 06 May 1999 07:44:46 -0700, Mike Coffin <[EMAIL PROTECTED]> wrote:
>If ESR is a typical libertarian, he insists that everyone should be
>left alone unless they try to bludgen someone else into doing
>something they don't want to do. His philosophy is the precise
>opposite of "ethics from the barrel of a gun".
What about freedom to act (as opposed to freedom from action)? I'm curious
about the libertarian stance on that.
m.
--
,--------------------------------------------------------------------------.
> Marco Anglesio | Love is a perky elf dancing a merry <
> [EMAIL PROTECTED] | little jig and then suddenly he turns <
> http://www.the-wire.com/~mpa | on you with a miniature machine gun. <
> | --Matt Groening, Life in Hell <
`--------------------------------------------------------------------------'
------------------------------
From: [EMAIL PROTECTED] (Andrew Comech)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.hardware
Subject: Re: Boycott Intel on your own webpage
Reply-To: [EMAIL PROTECTED]
Date: 6 May 1999 13:39:22 -0500
On Thu, 06 May 1999 10:31:39 -0500, Erica Vogle wrote:
>
>
>Andrew Comech wrote:
>>
>> On Wed, 5 May 1999 22:16:58 -0400, Bill Frisbee wrote:
>> >
>
>> >Last time I checked you all had IP addresses, MAC addresses
>> >and some like Sun boxes and several other Uber
>> >processor machines had serial numbers.
>>
>> You _think_ there is no difference between the above numbers and the
>> PSN thing. I _think_ you are wrong. Let me try to justify my point of view:
>> MAC addresses are only relevant when we are talking about MAC users, IP
>> numbers are usually dynamically-assigned, SUN boxes are not a common thing
>> at home, and "who's Uber?".
>
>When he says MAC addresses, he doesn't mean MACintosh, he means MAC
>address, as in ethernet address, the address that IP addresses are
>finally translated to. And he is right, since MAC addresses are more or
>less unique. A site could theoretically track these just as they could
>do with the hostid on a Solaris machine, or the Processor# in a PIII.
Oops.
Anyways, IP addresses are dynamically assigned when you dial up from home,
so who cares (although a netmask 255.0.0.0 would be good).
>This is not a new concept by any stretch of the imagination. It's only
>a big deal because the media has begun harping on it as the downfall of
>civilization, which has gotten the less-than-informed to take up arms
>against Intel. And you know what? It's not that different than when a
>site throws a cookie on your machine, so the next time you visit that
>site, they know who you are and can base questions, offerings, news,
>etc. on it.
Truly, I disabled cookies. Not that I care much, but...
Cookies themselves do not mean that much: this is far from a 100%
identification, especially if you fake your email address (and disable finger).
At the same time, PSN is close to 100% in more than 50% cases (all intel
users who did not know how to disable PSN).
>Just because, as you state, SUN boxes are not a common thing at home,
>doesn't mean that they aren't a common thing at businesses, where the
>real battle over tracking serial numbers would be fought. Trust me, Sun
>Workstations are all over the place, and you know what? Businesses
>don't mind that they have a burned in serial number.
I do not care much about businesses; as long as I am not affiliated with
one which would force me to work with a camera aimed at me or alike. But,
as you say, businesses may use Suns as well, so they are not that affected.
>From my point of view,
intel's PSN _primarily_ affects privacy of _private citizens_, and
it would be the _best_ aid (among cookies and whatever) in tracing them
up to their desk.
Let me also answer other raised issues:
This idea "stick to Linux and you are free from PSN identification" is...
stupid. Linux is not _that_ popular so far, and we could face the mentioned
problem that some internet services are just not accessible from machines
with disabled PSN. They would not care much about losing Linux users as a
whole, as do not care the manufactures of winmodems.
Also, there are more and more applications for Linux; just you wait for
an IE (or is it there already?) and others... Or are you going to
answer me that I (and everybody else) should carve out PSN lines from the
source code?.. Again, there are "non-free" applications, when the source code
is not available, and there could be more of those...
Also, there are all those java things and plug-ins, and I wonder whether one
may use them to turn the PSN on. So far, Linux is probably safe; next year it
will not be.
That is, this would be jungles of methods and and contra-methods which enable
or disable PSN, where only brave [hackers] are able to overcome PSN in their
computers. Do we want to face all that in a year or two, or do we just keep the
voice up trying to avoid PSNs completely?
a.
--
Looking for a Linux-compatible V.90 modem? See
http://www.math.sunysb.edu/~comech/tools/CheapBox.html#modem
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Subject: Re: A Simple Question
Date: 6 May 1999 11:55:08 -0500
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, jason wrote:
> One thing I find annoying about grep is that when given just one
> filename, it doesn't report the filename being matched at the
> beginning of the line. This is only annoying when used in find
> operations like the one above, since in most cases you want to know
> the filename it matched. For example,
>
> % grep foo file1
> This sentence contains foo.
> % grep foo file1 file2
> file1: This sentence contains foo.
> file2: Be somebody or be somebody's fool.
$ grep foo /dev/null file1
will work. (/dev/null will never match, and grep will see >1 filename.)
--
Paul Kimoto <[EMAIL PROTECTED]>
------------------------------
From: [EMAIL PROTECTED]
Subject: CLASSPATH in RedHat Linux 5.2
Date: Thu, 06 May 1999 16:32:30 GMT
Hi,
I am running RedHat Linux 5.2. I have defined some Java packages. When I
tried to compile my Java program by accessing those Java packages, it said
"class not found". I have used "export CLASSPATH=..." in either command line
or in ".bashrc" or ".bash_profiles", BUT it still didn't work. Can anyone
help? Thank you for your time.
Jason Lee
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] ()
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.advocacy,gnu.misc.discuss
Subject: Re: GNU reeks of Communism
Date: Thu, 6 May 1999 12:39:13 -0700
On Thu, 06 May 1999 11:47:50 -0700, Andrew Carol <[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>, Matthias Warkus
><[EMAIL PROTECTED]> wrote:
>
>> You need to read more history. Do you know how long it took to develop
>> Windows 95 or even patch the minor bits of better functionality and
>> more annoyances in that made Windows 98?
>>
>> The development of Gnome took only 18 months. 'Nuff said.
>
>Now compare the amount of software which requires Windows 95 and which
>requires Gnome. Notice how the Windows product still has an advantage
>which can be measured in ratios of thousands to one?
That's also something that has little to do with the
relative merits of the two platforms or whether or
not either are commercial ventures.
>
>This doesn't make Windows "better", it just is. The average consumer
>simply doesn't care. When there are more people installing Gnome than
>Windows X, come back and tell me.
This is the kind of gibberish that constitutes that
unsurmountable entry barrier that requires a complely
non-commercial product to combat.
It's !DOS.
[deletia]
It was an assinine criticsm in 1981, it was still
assinine in 1988 and is no less so in 1999.
--
Microsoft subjected the world to DOS until 1995. |||
A little spite is more than justified. / | \
In search of sane PPP Docs? Try http://penguin.lvcm.com
------------------------------
From: Andrew Carol <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.advocacy,gnu.misc.discuss
Subject: Re: GNU reeks of Communism
Date: Thu, 06 May 1999 09:47:25 -0700
In article <7KfY2.2162$[EMAIL PROTECTED]>, Christopher
Browne <[EMAIL PROTECTED]> wrote:
> Note that a "paranoid chip" that *permanently* removes your access to
> software at any provocation is a giant Denial of Service attack just
> waiting to happen...
No. There is simply nothing in the software model to even _attempt_ to
get the key. It is not in memory or in any software accessable
register.
The *permanently* would only refer to physical attack such as the
package being opened, chilled well below zero, etc.
I can't conceive of a denial of service attack on that type of setup.
--- Andrew
------------------------------
From: Ian Woollard <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.advocacy,gnu.misc.discuss
Subject: Re: GNU reeks of Communism
Date: Thu, 06 May 1999 18:31:07 +0100
Reply-To: [EMAIL PROTECTED]
Andrew Carol wrote:
>
> In article <[EMAIL PROTECTED]>, Jim Richardson
> <[EMAIL PROTECTED]> wrote:
>
> > Securing data is one thing, securing an executable
> > is a whole nother world. Since it is dynamic, and controllable, the cracker
> > can run the equivilent of dictionary attacks on the system, He doesn't need
> > to know the code, just be able to replicate the hardware signals at some
> > point.
>
> Dictionary attacks on 128 bit blocks would be almost unimaginable.
> This system does not have some "hardware signal" which says go or
> no-go. It simply decodes and runs what it gets.
Check.
> Of course, he could record the 'decoded' instructions as they enter the
> main part of the CPU, but they would be in some form of uMicrocode and
> may not translate easily back into assembly.
I think that they would just get back their normal machine code-
decryption doesn't perform any decompression.
> This would require the
> complete reverse engineering of a modern large scale 64 bit CPU without
> access to any Intel documentation other than the programmers model.
People may be just be able to run chips with their top off under an
electron microscope to read off the unencrypted stream as it goes past.
> Oh well....
--
-Ian ([EMAIL PROTECTED])
2 Secrets to success: 1. Don't tell everyone what you know.
------------------------------
From: Michael Keightley <[EMAIL PROTECTED]>
Subject: meta key doesn't work in xterms with Redhat-6.0
Date: 06 May 1999 17:54:33 +0100
Just installed Redhat-6.0. The Meta key no longer works in bash in xterms
(i.e. the Alt Key of PC keyboards). It works ok in Redhat-5.2. The
xterms seem to be producing 8-bit characters now, which I definitely
don't want. Any idea how to fix it? (Using the Escape key isn't the answer!)
Michael
--
_________
Michael Keightley <[EMAIL PROTECTED]> Tel: +44 131 220 4491
Systems Manager Fax: +44 131 220 4492
Quadstone Limited WWW: http://www.quadstone.com
------------------------------
From: Robert Krawitz <[EMAIL PROTECTED]>
Crossposted-To:
talk.politics.misc,comp.os.linux.advocacy,alt.activism,alt.society.liberalism
Subject: Re: The GNU Fragrance of Sharing vs. the Stench of Greed (was: GNU reeks of
Communism (really)^
Date: 03 May 1999 10:56:10 -0400
Bill Bonde <[EMAIL PROTECTED]> writes:
> Loren Petrich wrote:
> >
> > In article <[EMAIL PROTECTED]>,
> > Chris Costello <[EMAIL PROTECTED]> wrote:
> >
> > > The GPL is a crock. It forces openness. That's not freedom.
> >
> > The way that anti-slavery laws make one not free to own slaves?
> >
> Is it wrong to keep software proprietary?
Is it wrong to say "You're welcome to use my source code, but you have
to play by the same rules I'm playing by"?
--
Robert Krawitz <[EMAIL PROTECTED]> http://www.tiac.net/users/rlk/
Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail [EMAIL PROTECTED]
"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton
------------------------------
From: Sellaro <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Problem unmounting file systems (SIMPLE)
Date: Thu, 06 May 1999 14:05:29 -0300
I apologize for posting such simple question, but the manuals did not
work in this case :)
/usr/local is mounted in a separeted partition. KDE is installed on it.
When I have to shutdown the system, /usr/local is not unmounted because
it's in use. I think this happens because the system is running in level
4, and I'm using KDM as login manager.
Can anybody help me to solve this simple problem?
T.I.A.
--
~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~
"Eu sou um imortal. Nao tenho onde cair morto." - Olavo Bilac
~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~
Sellaro
Network and System Administrator
Computer Science Dept.
Federal University of Ceara - Brazil (UFC)
PGP KEY AVAILABLE UPON REQUEST
------------------------------
From: Josh Joyce <[EMAIL PROTECTED]>
Subject: Re: Q:compiling c++ codes that contains templates
Date: Thu, 6 May 1999 16:50:16 GMT
D. Vrabel wrote:
>
> On Thu, 6 May 1999, Richard wrote:
>
> > "D. Vrabel" wrote:
> > >
> > > On Thu, 6 May 1999, Richard wrote:
> > >
> > > > As title says: How do I compile c++ codes that contains
> > > > templates?
> > > > I know that I have to use the flag -frepo when doing compilation.
> > > > But I just couldn't link the object files after that.
> > > -frepo is an switch for older gcc's. Leave it out and you'll be able to
> > > compile template stuff.
> > >
> > Well, I found this flag (-frepo) when I failed to do separate
> > compilation
> > of a template class. If I put the definitions of member functions
> > together with
> > the template class in one file, the program compiles, but if I separete
> > them
> > to one header file and one .cc file, the compilation would fail with
> > some message
> You can't do this. The compiler needs to see all the template
> code so it can generate the code for referenced templates for each source
> file. A seperate process (collect2 I believe) runs before linking to
> remove duplicate instatiations of a template. ie You can't seperatly
> compile a template because it doesn't know what types to generate the code
> for. If you take a look at the STL headers you'll see that all the code
> is in the headers.
When egcs gets support for the export keyword, it should be possible.
Josh
------------------------------
From: "Mattias Dahlberg" <[EMAIL PROTECTED]>
Subject: Re: How can X be so slow?
Date: Thu, 6 May 1999 18:18:09 +0200
>1) Window manager might be slower, or configured differently.
Same window manager and the same desktop environment.
>2) Less mature X server.
Well, that of course could be it. Anyone else experiencing slow i740
performance?
>3) RAM Starvation. How much RAM do you have on your box at home?
Nope. At home 128MB and at work 64MB. So everything suggests X should be
faster at home, but it's so much slower...
Matt
------------------------------
From: [EMAIL PROTECTED] (Bill Unruh)
Crossposted-To: comp.os.linux.development.apps,comp.unix.bsd.freebsd.misc
Subject: Re: Is Unix a single user operating system?
Date: 6 May 1999 17:18:50 GMT
In <[EMAIL PROTECTED]> "Rolf Marvin Be Lindgren" <[EMAIL PROTECTED]>
writes:
>right, and a file can belong to at most one user and one group, and on
>some unices there's even a limitation to how many groups a user can
>belong to.
Yes, a more detailed ACL would be useful sometimes. However yo useems to
be of the opinion that if an OS does not have just the thing that you
want, then it must be single user. It is funny how a single user OS can
then have 50 people all running under the single user OS and not getting
in each others way. If you want to call such a system single user, I
guess you may. You are free to invent your own language which
communicates to noone else if you wish.
>under VMS, OPERATOR can read and write to all user's directories. under
>UNIX, only root can do that.
So, you have some attachment to the word "OPERATOR"? I would call any
user that can write and read everything root.
You seem to attach great meaning to arbitarty labels people attach to
things. Would it make you happier is we all started calling root on a
linux system OPERATOR?
>whenever a user came to us with a magtape, under VMS it was no problem
>to dump the contents to the user's home directory. under UNIX, we had
>to use the only machine that could read the tape station, find a
Lets, see, under VMS, a machine that has no tape drive can read tapes?
Hmm it is a good OS!
>directory with enough room, _ask root to mount that machine's disk to
On VMS you can put more data on a disk than it can hold? Again a great
system!
On VMS any user can put on any tape and dump garbage into anyone elses
directory? And this is useful?
>the user's machine so the user could read his tape dump_...
I wonder why I can allow a user to read his own tapes under Unix? I must
have been dreaming!
>all of this convinced me that an operating system that does not support
>an OPERATOR concept is fundamentally single-user. there muse be a user
It does have one. It is called root.
>midway between user and root.
I do not call something which can read and write anywhere "midway
between user and root" I call it root.
Now, I do agree that more complex ACLs would be nice in Unix/Linux. A
single group ownership of a file is not always sufficient to allow that
file to be treated in the way one wants. But that will be true of any
ACL system-- I can always come up with real life scenarios where the ACL
will be insufficient to capture exactly what I want to allow to be done
with the data. (eg, I might want to have byte level control over the
contents of the file, so that only certain people can read or alter
certain sections of the file, while still leaving it as a single file to
those that have the appropriate permissions. Of course this would lead
to an ACL structure associated withthe file that was much longer than
the file, but maybe that is what I want. I should therefor say that no
multiuser system exists because none offers me that level of control? Or
should I say "I will that the Unix access control structure were richer
than it is"?)
>--
>Rolf Lindgren http://www.uio.no/~roffe/
>[EMAIL PROTECTED]
------------------------------
From: brad <[EMAIL PROTECTED]>
Subject: simultaneous pap users
Date: Thu, 06 May 1999 17:12:59 GMT
I'm running pppd under redhat 5.1 using PAP for the authentication for
dial-in users. Is there a way to prevent having multiple concurrent logins
from the same user? In other words if I've defined a user named joe in
pap-secrets, I want to prevent having 2 separate users dialing in and both
using the username joe to login. If a user joe has already logged in, then I
don't want anyone else to be able to login as joe.
Thanks,
Brad
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Marco Anglesio)
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.advocacy,gnu.misc.discuss
Subject: Re: GNU reeks of Communism
Reply-To: [EMAIL PROTECTED]
Date: Thu, 06 May 1999 17:59:48 GMT
On 06 May 1999 10:45:16 -0700, Mike Coffin <[EMAIL PROTECTED]> wrote:
>Libertarians believe that one's right to swing a fist stops at the
>next guy's nose. They believe you can act however you like as long as
>you don't whack someone else in the process.
Pretty poetry (Mill's always been one of my favourites - although I
daresay he would have personally disapproved of libertarians :) ) but not
really answering the question that I asked.
I'll lay a basis for the question: there is freedom to act - freedom to,
as Mill said, swing one's arm so long as it doesn't run into your nose.
And there's freedom from action - if your arm-swinging prevents me from
doing something, should I have the right to prevent you from doing it? The
latter is the question I was really asking.
marco
--
,--------------------------------------------------------------------------.
> Marco Anglesio | Alcohol, hashish, prussic acid, <
> [EMAIL PROTECTED] | strychnine are weak dilutions. <
> http://www.the-wire.com/~mpa | The surest poison is time. <
> | --Ralph Waldo Emerson <
`--------------------------------------------------------------------------'
------------------------------
From: Sam Alexander <"salex"@psw(takethisout).com>
Subject: Re: CLASSPATH in RedHat Linux 5.2
Date: Thu, 06 May 1999 12:49:02 -0400
[EMAIL PROTECTED] wrote:
> Hi,
>
> I am running RedHat Linux 5.2. I have defined some Java packages. When I
> tried to compile my Java program by accessing those Java packages, it said
> "class not found". I have used "export CLASSPATH=..." in either command line
> or in ".bashrc" or ".bash_profiles", BUT it still didn't work. Can anyone
> help? Thank you for your time.
When you add the CLASSPATH variable to your .bashrc or .bash_profile, you must
shutdown Netscape and either log out and back in, or type:
source .bashrc
or
source .bash_profile
then reload Netscape for the changes to take affect. If you are loading Netscape
from a button or action, I would log out and log back in to make sure all shells
get updated.
The problem is that if you just add it to one of these files before doing the
above, the shell that called Netscape doesn't know about this variable, so
Netscape will not know about it either.
Sam Alexander
>
>
> Jason Lee
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Sellaro <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: kernel 2.2
Date: Thu, 06 May 1999 14:07:25 -0300
Monica wrote:
> How can I upgrade a kernel 2.0.32 Red hat to a kernel 2.2? I have a CD with
> this kernel but I dont't know what I must do.
You can find useful information reading the Kernel-HOWTO. If you have some specific
problems, please drop them here ;)
--
~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~
"Eu sou um imortal. Nao tenho onde cair morto." - Olavo Bilac
~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~
Sellaro
Network and System Administrator
Computer Science Dept.
Federal University of Ceara - Brazil (UFC)
PGP KEY AVAILABLE UPPON REQUEST
------------------------------
** 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
******************************
