Linux-Misc Digest #193, Volume #19 Fri, 26 Feb 99 19:13:13 EST
Contents:
Re: Best Free Unix? (why FreeBSD?) (NF Stevens)
Re: Linux is not even in Windows 9X's class. (Rick Onanian)
Re: Hard disk duplication?? (Dan Mack)
primary disk fail ?? ([EMAIL PROTECTED])
Re: Linux/FreeBSD compatability (Was Re: Best Free Unix? (why FreeBSD?)) (Matthias
Buelow)
Re: Turning 'ping' off, please help. (Lew Pitcher)
Re: domain name reg and IP setup (Alex Kamantauskas)
Microkernels are an abstraction inversion (Francois-Rene Rideau)
Re: Linux is not even in Windows 9X's class. (William Wueppelmann)
Re: RH vs SuSE ("Robert C. Paulsen, Jr.")
Re: Linux is not even in Windows 9X's class. (The Ghost In The Machine)
Re: Linux Programs (jik-)
Re: Best Free Unix? (why FreeBSD?) (NF Stevens)
Re: Can Linux run on NT for stability? (Yan Seiner)
Re: Dawicontrol DC-2974 compatibility (Holger Matthoefer)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (NF Stevens)
Crossposted-To:
comp.unix.questions,comp.unix.advocacy,comp.unix.misc,comp.unix.bsd.freebsd.misc
Subject: Re: Best Free Unix? (why FreeBSD?)
Date: Fri, 26 Feb 1999 21:17:02 GMT
Zenin <[EMAIL PROTECTED]> wrote:
>John Girash <[EMAIL PROTECTED]> wrote:
> >snip<
>: As for the bit about the infection of "derived works": assuming you only
>: use pristine(*) GPL code and want to simply link it to your own, that's
>: rather trivial in terms of legality.
>
> The LGPL is *not* the same as the GPL. The LGPL isn't nearly as
> evil or infectious as the GPL.
Since neither LPGL nor GPL is in any way evil I suppose your statement
is (from an axiomatic logical point of view) true. However I'd seek help
for that bigotry if I were you.
>
>: Your work is not a derived work, it's merely interfacing with some GPL'd
>: routines.
>
> If it's under the GPL and not the LGPL, this is not the case. Even
> the LGPL gets sticky.
>
> Also, what if I see 10 lines of code from a 10k line GPL project
> that I would like to use in my 10k line app? Under the GPL, *my*
> code must carry the full GPL license.
What if you see 10 roses in a display of thousands that you'd like to
have? Do you have to pay for them, or is it ok to take them without
asking?
>
> To use those 10 little lines of GPL code I must give up my right to
> license *my* code how I see fit. That is why the GPL is a virus.
And taking 10 roses from a display of thousands without asking can
get you a criminal record.
Norman
------------------------------
From: Rick Onanian <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Linux is not even in Windows 9X's class.
Date: Fri, 26 Feb 1999 16:13:30 -0500
I know, not good to reply to one's self..I just remembered something,
tho. Hrmph.
Rick Onanian wrote:
> > > > 6. One word: Microkernel
> > >
> > > Can mean a performance penalty.
> >
> > Waiting 15 minute for my kernel to compile for a new driver is a performace penilty
>
> Why in the world would you bother compiling? Modules do everything now.
Modules, which, BTW, you can load and unload without rebooting. I
recompiled a kernel once. Never had to reboot after that. In windows,
everytime I make a minute change (IP address, which version of a driver,
etc.), I'm required to reboot. In Linux, you can reconfigure pretty much
the whole system without rebooting - devices, addresses, etc.
Therefore, the once that I compiled my kernel cost me 15 minutes. The
15-30 times I rebooted the Windows machine cost me 2 minutes each.
--
rick - a guy in search of raw (ISO) cd images of SuSE and Slackware
===============
My opinions don't exist, and as such, are not anyone elses. I do not
represent
anyone, not even myself, and especially not my employer. Cows go moo.
---
Looking for a 1968 Camaro SS convertible, black interior, beat-up
rustbucket
that is in need of a lot of restoration and TLC. Must be cheap...I'm
broke.
---
Reply to me at either thc <at sign here> psynet <dot> net or
rick <at sign> mail <dot> artmold <dot> com
------------------------------
From: Dan Mack <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.help,linux.redhat.install,linux.redhat.misc
Subject: Re: Hard disk duplication??
Date: Fri, 26 Feb 1999 15:48:09 -0600
Bruce Kline wrote:
>
> Hi,
>
> I think the use of dd should work here as an image
> copier to a partition that is the exact same size.
> I have never done it.
>
> As long as people are looking here I want to ask the
> additional question of is there a way to duplicate a
> DOS/WIN partition with dd, and have it be able to boot?
>
> I am very curious as to whether this is do-able!?
Yes, dd will work fine. We used to do this at the University. We had
20 dual-boot Linux/Win95 desktops. We made one, and then we cloned the
disks using dd, something like (don't dd the partitions, dd the whole
disk):
original drive is hda
new drive is hdb
<boot single user>
# dd if=/dev/hda of=/dev/hdb bs=1024k
For a 500M drive it took us about 20 minutes per disk. It worked great
but I would only do this if you know that all your hardware is
identical. It is also a good idea to fsck the drive's partitions after
you dd it:
# fsck /dev/hdb1
# fsck /dev/hdb2
<etc..>
Becuase we found that the new partitions were dirty. We actually
mounted the new partitons up on the clone machine and changed system
files and such before putting the disk back in the destination machine.
Hope this helps, I've also done this on SunOS 4 & 5 and IRIX with no
real problems or issues. It's not for everyone but it can save you a
lot of time.
Dan
--
- -- -- --- --- --- ---- ---- ----- ----- ----- ----- ------
Dan Mack ([EMAIL PROTECTED]) <http://thistle.cray.com/~mack>
Information Services 651-683-7288
Silicon Graphics / Eagan, MN - -- --- --- -----
------------------------------
From: [EMAIL PROTECTED]
Subject: primary disk fail ??
Date: Fri, 26 Feb 1999 21:43:39 GMT
This would be fine... "primary master hard card fail"... except it only
happens SOME of the time... the disk is OK if I boot to it with rescue disk,
etc.
It seems like a power thing too... if I open the case, and reconnect the
power cable to CD, it works fine... however, if just turn power off and on,
it works also... so, probably not a cord problem.
Is it possible this is a virus for linux ?
It is definitely new... Ive run this linux for months with no probs.
stu7
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Crossposted-To:
comp.unix.questions,comp.unix.advocacy,comp.unix.misc,comp.unix.bsd.freebsd.misc
Subject: Re: Linux/FreeBSD compatability (Was Re: Best Free Unix? (why FreeBSD?))
From: [EMAIL PROTECTED] (Matthias Buelow)
Date: 26 Feb 99 19:39:59 GMT
In article <7b6skh$27ts$[EMAIL PROTECTED]>,
Robert Sexton <[EMAIL PROTECTED]> wrote:
[...]
Great Article.
>UNIX(TM) is a trademark of somebody, but I forget who currently owns
>it :-)
The Open Group. :)
------------------------------
From: [EMAIL PROTECTED] (Lew Pitcher)
Subject: Re: Turning 'ping' off, please help.
Reply-To: [EMAIL PROTECTED]
Date: Fri, 26 Feb 1999 20:02:53 GMT
On Fri, 26 Feb 1999 17:19:35 GMT, Chad M. Townsend <[EMAIL PROTECTED]> wrote:
>If I _do not_ want my computers to return a ping (besides turning them off)
>how to I disable ping?
I'd say that one way would be to install an IP firewall
(ipfwadm or ipchains) and disable ICMP echo replies using the firewall rules.
There's probably other ways as well.
Lew Pitcher
System Consultant, Integration Solutions Architecture
Toronto Dominion Bank
([EMAIL PROTECTED])
(Opinions expressed are my own, not my employer's.)
------------------------------
From: [EMAIL PROTECTED] (Alex Kamantauskas)
Crossposted-To: comp.os.linux.networking,comp.os.linux.help,comp.os.linux.setup
Subject: Re: domain name reg and IP setup
Date: 26 Feb 1999 22:10:24 GMT
> But surely this is a domain hosting service - correct me if I am wrong ?
>
The InterNIC, or Network Solutions, Inc., serves two functions - they
maintain the files that update the various root servers, and they act as
registrar for the COM, ORG, and NET domains.
> The info that I require is to know howto set up my own IP www homepage
> and control it from my own box, I know I need to contact the respecive
> authorities
> Who are they ?
Depends on what domain name you want to be registered under - InterNIC, as
mentioned, runs the .com, .net, .org. However, they cannot register you
under .co.uk.
> What hardware do I need - leased line or normal daily used telephone
> line ?
>
> I have set up PPP and DNS this works fine for my IPS but what if I would
> like to host my own server.
>
If you will be hosting your own DNS server, than you will want a line that
is up 24/7.
--
alex kamantauskas
tugger networks
[EMAIL PROTECTED]
------------------------------
From: Francois-Rene Rideau <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.linux.advocacy
Subject: Microkernels are an abstraction inversion
Date: 26 Feb 1999 16:31:46 +0100
[EMAIL PROTECTED] (Christopher B. Browne) writes:
> The theory is that the microkernel approach should be:
> a) More supportive of multiprocessor systems; [...]
> b) More stable. [...]
> c) More portable. [...]
> d) Easier to mess with. [...]
> Practice provides some downsides:
> [All the contrary, and much worse]
Your message sums up the hype and reality very well,
but it lacks a *general explanation*.
And explanations are precisely what is needed to avoid further mistakes!
> There are lots of bits of counterexamples; see VSTa, QNX, L4, Fiasco
> as examples of "workarounds" for some of the downsides. None have
> "leapt out" to replace Linux or *BSD or Hurd in terms of mindshare,
> for whatever reasons.
For a good reason: the whole idea of a microkernel is EVIL, and its *only*
possible valid justification is in proprietary black-box kernel components.
Microkernels are an ABSTRACTION INVERSION between a high-level concurrent
agent programming model and a low-level resource management implementation.
They introduce overhead without simplifying any functionality:
complexity is just moved out of the "kernel", not at all reduced;
the only global effect of microkernel design is increasing overhead
in inter-service communication.
If you're any interested in Micro-kernels, please read the TUNES Glossary:
http://www.tunes.org/papers/Glossary.html#microkernel
[ "Far�" | VN: Уng-V� B�n | Join the TUNES project! http://www.tunes.org/ ]
[ FR: Fran�ois-Ren� Rideau | TUNES is a Useful, Nevertheless Expedient System ]
[ Reflection&Cybernethics | Project for a Free Reflective Computing System ]
Don't have good ideas if you aren't willing to be responsible for them.
-- Alan Perlis
------------------------------
From: [EMAIL PROTECTED] (William Wueppelmann)
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Linux is not even in Windows 9X's class.
Date: Fri, 26 Feb 1999 20:10:16 GMT
In our last episode (Thu, 25 Feb 1999 18:44:23 -0800),
the artist formerly known as Ryan Cumming said:
>Matthias Warkus wrote:
>
>> > 1. Better GUI
>>
>> Hmph. Depends on how you define a GUI and what kind of GUI you are
>> using on Linux.
>
>I dont't think any UI in Linux even touches Windows ease of use. Some of the new
>fangled desktop enviroments (Gnome, KDE, etc.) come close, but MS has done a good
>job of allowing the newbie be able to use the OS right away,
But is that ease of use or is that ease of learning? To me, ease of use
means being able to accomplish task x in an efficient manner. While you
only learn a system once, you use it every day, so to me, how effectively a
skilled person can use the system is a much more important metric than how
easily a rank novice can get started (though it's true that often people
don't bother to learn their system and stay at the level of rank novice
indefinitely; this is a separate problem though). Windows does not allow
this at all for many cases, and in others requires a knowledge of the
system similar in sophistication as that needed to use X effectively. Part
of the problem is Windows' lack of configurability: often you must do
things the One Microsoft Way if you want to do them at all, and the One
Microsoft Way is a way which, while easily grasped by novices, is hardly
easy to use. Of course, the other is the lack of a unifying design
philosophy (other than `Windows everywhere,' which leads to a lot of
changes to the system without reference to the overall design.
Also, Windows is seriously losing its newbie friendliness. Windows 3 was
certainly ugly, but it was clean and simple and used consistent metaphors
and controls. I haven't actually worked on it (as opposed to repaired
systems belonging to other people) for 4 years, but I can still get around
on it without a mouse attached to the computer. Windows 95 and 98 have
been successively more baroque, with extra controls, unusual items (e.g.
the system tray) which don't behave like the rest of the system, requiring
the user to learn a separate method for doing things, the unbelievably
twisted .inf files which install their programs by right-clicking (only)
the icon and selecting install, in complete defiance of every aspect of the
standard install procedure, and other nastiness (most of which is detailed
at iarchitect's Interface Hall of Shame).
--
William
It is pitch black. You are likely to be spammed by a grue.
------------------------------
From: "Robert C. Paulsen, Jr." <[EMAIL PROTECTED]>
Subject: Re: RH vs SuSE
Date: Fri, 26 Feb 1999 17:02:30 -0600
Jason S Hackney wrote:
>
> I am very new to Linux. Just installed two days ago. I saw a copy of
> SuSE in the store a while ago, but chose to install Red Hat 4.2 (which I
> bought about a year ago). I'm considering switching to SuSE because I
> don't care much for the software that shipped with RH 4.2. Can anyone
> tell me a bit about SuSE? Is it worth my trouble? Has anyone had any
> problems with it? I am also going to need to install my 3c905B NIC --
> I've had trouble getting RH to recognize it, but I think that's a user
> error on my part.
The 3C905B is supported by SuSE, and by just about any recent
distribution, I think.
>
> I would appreciate any advice availalble. I'm sure this has been
> discussed in the past so I don't mind if anyone emails me at
> [EMAIL PROTECTED]
>
> Thanks so much.
> -Jason
Be sure to get the new SuSE 6.0. Some places may still have 5.3 on the
shelf.
I have tried the following distributions (in this order):
Red Hat 4.2 (some time ago when this was the newest)
Slackware 3.6
SuSE 5.3
Red Hat 5.2
SuSE 6.0
So far, I like SuSE 6.0 the best by a pretty good margin. It's install
process seems to be the best organized and easiest to deal with. It has
the biggest selection of software included with the distribution. For
example, RedHat doesn't have KDE, StarOffice, nor even diald.
I have had three problems with SuSE 6.0:
1. The Applix demo gets an error trying to import Office95 files. SuSE's
web page acknowledges the problem but doesn't yet offer a fix.
2. smbumount fails. I just installed the 5.3 version and all is OK.
3. the egcs 1.1 compiler isn't configured right. I didn't bother trying
to figure it out. Instead I installed the 1.1.1 version from the
Cygnus/egcs site and it works OK.
--
Robert Paulsen http://paulsen.home.texas.net
If my return address contains "ZAP." please remove it. Sorry for the
inconvenience but the unsolicited email is getting out of control.
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Linux is not even in Windows 9X's class.
Date: 26 Feb 1999 21:11:24 GMT
On Wed, 24 Feb 1999 18:43:17 -0800,
Ryan Cumming <[EMAIL PROTECTED]> wrote:
>iratheous wrote:
>
>> > This fun to reply to. Maybe we dont have all those speacialized graphics
>> >clogging up our memory, but when my software crash's, it the software. not
>> >need to reboot whole system. If you want my 234 reasons why linux is better
>> >than Winblows 98 just ask.
>>
>> How about 10 honest reasons why you think windows98 is better than linux?
>> That would be interesting to hear from a windows hater >:)
>
>OK. I think there are more and better resons that Linux is better, but here it
>goes: (This hurts)
>
>
>1. Better GUI
Define "better". Athena is extremely primitive, but works.
Motif is a bit better. Openlook works, but is largely gone now,
I think. Gtk++ seems to be state of the art,
but I can't say I've studied it in detail yet. Tcl/TK is
very easy to use, but the Tcl part of it is slightly klunky.
Windows is fixed, the standard, messy, and occasionally stupid.
Ncurses isn't graphical, but has a very light footprint and
isn't that hard to program in.
And there are even a few apps which don't really have a "GUI"
at all, but just open a window and do all the hard fiddlybits themselves.
>2. More software
Well, admittedly, that's debatable. There's a lot of freeware source code
out there, but equating that to shrinkwrapped Windows packages may
be like comparing a blueprint to a finished car, in some cases.
(Especially if the software is old and needs to be upgraded.)
But at least with source code one can conceptually read it.
(Side note: I think the gaming industry, at least with the 3-D
shootemups, is progressing towards data, as opposed to source code,
being the defining factor -- witness the success of Unreal, for
instance; the prime factor there isn't its game engine, but the
worlds using it. (I still see some ordering problems in there, BTW. :-) )
>3. More Hardware support
If the Hardware isn't something like a WinModem or a WinPrinter, maybe.
(Ugh.)
>4. Better gaming platform
>(shudder)
Again, define "better". I'll admit it's more stable; it also
doesn't have as big a footprint, making more resource available
to the game itself.
(Side note: I hate Windows95 with UnrealEd, written in (shiver)
Visual Basic. If I sneeze, it goes down, and about half the time it
takes my Win95 system with it.)
>5. More consistency (see my previous post)
Can't comment; haven't seen it.
>6. One word: Microkernel
Eh?
This is a funny concept. NT is also supposed to have a microkernel,
for example. I'm not sure this makes any difference, or not.
>7. No mounting
I think you're slightly confused; Windows may mount, but the user
doesn't do the mounting; the system does after it detects a disc change.
Assuming that the concept of "mounting" makes sense in the context
of Windows (or DOS!) at all.
Unix requires either explicit mounting, the usage of such tools as
mtools (mdir, mtype, etc.), or direct access to the raw device file.
Linux does, too...although one can put an entry into /etc/fstabs,
which allows a user to mount volumes. (Needless to say, this could
be a security risk if improperly done.)
>8. Better file locking
>9. More multithreaded apps
>10. Better user support
Unix was designed (or at least evolved) with multiuser in mind. Windows
NT...um...dunno if it's multiuser or merely multiprocess; it's a funny
operating system, especially if DCOM is thrown into the mix (does DCOM
run in its own process, in the first process requesting its loading,
just in a thread managed by an 'init'-type process, or in whatever
process is needing it at the moment (i.e., like Amiga dynamically
loadable and shareable libraries)? :-)
>
>(Bodnar42 collapses because he just went against everytihing he stands for)
>
Which means you're for less software, less hardware, worse gaming,
less consistency, monolithicity, explicit mounting, no or cruddy
file locking, no multithreads, and no users? :-)
Just curious....
(To be fair, a lot of hackers [*] -- at least on the Amiga -- love to talk
directly to the hardware, which allows for a speed improvement if they
are good at it. In modern systems, however, there are drawbacks,
mostly because the hacker is going to have to work carefully with the
operating system he's hacking in so as to not confuse the device when
two requests (one from him, one from the OS) come in. Improperly done,
all sorts of DMA and interrupt silliness could ensue.)
[*] as opposed to crackers and malicious vandals. In a sense, a hacker
is a cracker, but in a good sense; he's exploiting the hardware
and the operating system to accomplish something useful, or at least
pretty (some of the Amiga eye candy was simply amazing, even without
acceleration) -- but to do that he may have to understand certain
undocumented features on occasion.
----
[EMAIL PROTECTED] -- and then there's CORBA...
------------------------------
From: jik- <[EMAIL PROTECTED]>
Subject: Re: Linux Programs
Date: Fri, 26 Feb 1999 15:07:10 -0800
> - they fit into the system properly (config files pre-configured and in
> the right place, bugs removed, etc). This is one reason why Debian is
> considered the most secure Linux distribution.
How does packaging something suddenly make it better? Any bugs that
were there, are there,...unless they were fixed by someone...but then it
wasn't the packager that did it, was it. Also, .deb packages have
nothing to do with how secure or unsecure Debian is.
The amount of bugs in the program is directly proprtional to the
programmers ability, and the time he/she had to squash bugs before the
release.
The security of the system is directly proportional to the software on
the system, and the administrater upkeeping that software.
Neither have anything to do with any package format.
------------------------------
From: [EMAIL PROTECTED] (NF Stevens)
Crossposted-To:
comp.unix.questions,comp.unix.advocacy,comp.unix.misc,comp.unix.bsd.freebsd.misc
Subject: Re: Best Free Unix? (why FreeBSD?)
Date: Fri, 26 Feb 1999 21:17:03 GMT
[EMAIL PROTECTED] (John S. Dyson) wrote:
>In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] (NF Stevens) writes:
>> jik- <[EMAIL PROTECTED]> wrote:
>>
>>>> You aren't seeing it from the point of view of the original author of the
>>>> GPL work. "If I had wanted you to use my code for something you would
>>>> sell for profit, I wouldn't have licensed it under GPL, would I?"
>>>
>>>That statement seems to be directly incompatable with FSF's definition
>>>of 'free' software.
>>>
>> A better wording which (as I interpret it) preserves the original author's
>> meaning would be "If i had wanted you to use my code in something...".
>> That is entirely consistent with the FSF's definition of free software.
>>
>Selective freedom. The code ain't free.
>
>Not only that, if you don't want people to use your code, then don't
>let people use it.
There are two uses of the word "use" here that you seem to be
confusing. You should consider improving your comprehension
skills.
Norman
------------------------------
From: Yan Seiner <[EMAIL PROTECTED]>
Subject: Re: Can Linux run on NT for stability?
Date: Fri, 26 Feb 1999 17:43:09 -0500
Huh? I think you're confusing hardware and software.
Yan
moi wrote:
> I have found my NT server very reliable, and want to run Linux on it. Is this
> possible? I know some people say Linux is stable, but I don't want to take a
> chance.
>
> moi
------------------------------
From: [EMAIL PROTECTED] (Holger Matthoefer)
Subject: Re: Dawicontrol DC-2974 compatibility
Date: Fri, 26 Feb 1999 16:06:14 +0100
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>,
Arthur Dent <[EMAIL PROTECTED]> writes:
> Does someone know what Host adapters have compatible drivers
> for the Dawicontrol Dc-2974 Pci ?
It uses an AMD chip (AM53/79C974); the drivers for the Tekram
DC390 should work also, but I didn�t get them to work when
installing Linux. Once Linux was running, I had no problems
compiling a new kernel with the latter. The Tekram driver should
be preferred, as it�s still being developped (in contrast to the
generic one which is essentially superfluous). and it now
officially takes care of all Controllers with the AMD chip.
Holger
------------------------------
** 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
******************************
