[Q] Viruses Under UNIX?
Wayne Lee (Leewa)
To anyone,
I was wondering if UNIX platforms are vulnerable to "Virus"
programs similar to those that can affect DOS/Windows platforms? I have
been told that writing a memory-resident virus program for a UNIX platform
is not likely because of the way UNIX's memory management is done. UNIX
is more vulnerable to "Trojan Horses" and other things in this genre from
what I have been told. How does this compare to Win95 and WinNT? Even
though NT and 95 have different memory management schemes, are they
vulnerable to virus attacks that you see on DOS/Windows 3.1 platforms? If
someone can also give me a brief explaintion of the differences in the way
UNIX, WinNT, Win95 and Win3.1 manage memory, this would be appreciated
too.
Thanks,
Wayne
Mica Currie
Wayne Lee (Leewa) (pha...@neosoft.com) wrote:
: To anyone,
: Thanks,
: Wayne
In one word. NO. They are not vulnerable in the way that dos and windows are. DOS/Windows is an inheirantly open system with no security. Since *NIX's are by their nature intended to be secure systems, any time a place was found where bits of code could be used to harm or damage the system, the system was fixed to prevent the security problem. This is not to say that there is not holes. Merely that they tend to be patched very quickly once found making any sort of exploit with that hole limited to those wh
o haven't fixed the bug, or who are unaware. Good sysadmins don't remain unaware for long. In contrast DOS/Win is full of holes and their attitude is to detect exploit code 'virii' then eliminate the harmful code. Needless to say the *NIX approach works much better.
Unix is less vulnerable to trojan horses as long as the code is not run as root. Any user who runs a trojan horse program, since the program is limited to that users access, is the only one affected whereas with dos win31/95 the whole system can be lost. depending on how you are running WinNT it will either be like win31/95 if your not using ntfs on the system partition and is not as vulnerable if you are.
As far as memory.
DOS gives you 640k directly addressable memory.
After that you need to run device drivers to provide the
system calls to address more memory.
Win31 uses the same principle as dos except that they add the
swap partition which extends the physical memory via a
file on the harddisk.
I don't know how win95/winnt deals with memory however
in theory winnt provides the ability for protected memory
spaces. In which each processes can either use a shared
memory pool (like win31) or a protected memory. With protected
each process gets a block of memory that it has exclusive rights
to and no other process can read or write to it. As well
this process is not permited to read or write outside of this
protected block. As well winnt and 95 use swap space to increase the
amount of physical memory.
Finally *NIX's work by granting applications virtual protected
memory. In which each application is given a virtual address space
which is infinite in capacity. (theoretically) when the process
requests memory it allocates memory from the system pool and maps
it to the processes virtual address. This provides
memory security as no process can ever attempt to read anothers memory
without explictly coding support in both processes so that they can
share memory.
ttyl...Mica
cosmos
In comp.security.unix Mica Currie <mi...@vaughan.fac.unbc.ca> wrote:
: In one word. NO. They are not vulnerable in the way that dos and windows are. DOS/Windows is an inheirantly open system with no security. Since *NIX's are by their nature intended to be secure systems,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
hold on, time out, stop the presses...
since when has UNIX been "intended" to be a secure OS???!!
you sir, are very misinformed and need to go back and reread
your textbooks... ;)
seriously, security was not an intent/initial design of UNIX,
it's an open system designed for programmers with some light
security thrown in.
only now are teams like OpenBSD etc making security a focal point
in UNIX design and implementation.
as far as patches....quick is not what id use to label most
OS vendor's response to major security flaws. *BSD and linsux
are quick do to their open source tree, that's about it.
--cosmos
Mica Currie
cosmos (cos...@NOSPAM.enteract.com) wrote:
: In comp.security.unix Mica Currie <mi...@vaughan.fac.unbc.ca> wrote:
: : In one word. NO. They are not vulnerable in the way that dos and windows are. DOS/Windows is an inheirantly open system with no security. Since *NIX's are by their nature intended to be secure systems,
: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
: hold on, time out, stop the presses...
Just as a point of clarification. Unix was built as a multi user system which required security in the form of process control, memory, and filesystem to be stable. This is my justification. I do not mean by this statement that *NIX in anyway shape or form meant to be a "completely secure" by design. It was however written to prevent users from screwing each other up, otherwise it's use of a multi user system would be limited to small groups of people who get along with each other.
ttyl...Mica
slidge
In article <6atda4$i...@eve.enteract.com>, cosmos <cos...@NOSPAM.enteract.com> wrote:
>In comp.security.unix Mica Currie <mi...@vaughan.fac.unbc.ca> wrote:
>: In one word. NO. They are not vulnerable in the way that dos and windows are.
> DOS/Windows is an inheirantly open system with no security. Since *NIX's are
> by their nature intended to be secure systems,
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
> hold on, time out, stop the presses...
>
>
> you sir, are very misinformed and need to go back and reread
> your textbooks... ;)
>
I think he was indicating that there are certain principles with Unix that you
don't find in DOS/Windows. For instance, in DOS/Windows you don't have the
idea of a superuser, nor can you prevent people from tinkering with your files
by establishing permissions as to who can or cannot play with them. That's
the impression that I was getting.
However, I go along with what you say:
Russell Conner
In one word:
MAYBE.
Certianly any Intel based unix, such as SCO,LINUX, etc. is because some
attack the boot sector, hence the name boot sector virus.
Granted, it would be harder to do so, but it is possible.
Most likely if you boot into several OS'es on one machine.
Or if you booted off a contaminated diskette.
> In one word. NO. They are not vulnerable in the way that dos and windows
are. DOS/Windows is an inheirantly open system with no security. Since
Tom Holub
): In one word. NO. They are not vulnerable in the way that dos and windows are. DOS/Windows is an inheirantly open system with no security. Since *NIX's are by their nature intended to be secure systems,
) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
) hold on, time out, stop the presses...
)
) since when has UNIX been "intended" to be a secure OS???!!
Only two posts into the monthly troll thread and we're already to this point.
Very impressive.
-Tom
Kuntal M. Daftary
On 30 Jan 1998, cosmos wrote:
> : In one word. NO. They are not vulnerable in the way that dos
> and windows are. DOS/Windows is an inheirantly open system with no
> security. Since *NIX's are by their nature intended to be secure
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> since when has UNIX been "intended" to be a secure OS???!!
well, compared to DOS, unix is comparatively secure. it has concepts of
users/login/password, each user with his own privileges. it has concept of
system calls, user memory space and ssytem memory space, it have concepts of
protected memory space and root privileges.
not too bad, for basic security
:-)
Kuntal Daftary
1.408.527.9789
daf...@cisco.com
cosmos
In comp.security.unix Kuntal M. Daftary <daf...@cisco.com> wrote:
: On 30 Jan 1998, cosmos wrote:
: well, compared to DOS, unix is comparatively secure. it has concepts of
: users/login/password, each user with his own privileges. it has concept of
: system calls, user memory space and ssytem memory space, it have concepts of
: protected memory space and root privileges.
: not too bad, for basic security
: :-)
true. however, the way broad assumption that i gleamed from the original
post was way off base...of course, the poster clarified the original post
and put it into a bit better context.
and btw, compared to DOS, my cats litterbox has a pleasant odour. ;)
--cosmos
Gary L. Burt
One other point to consider. UNIX can not be expoited as easily by a binary
executable, since there are so many different underlying type of hardware.
This sures as a layer of protection (to a degree). However, it can be
penetrated by using shell scripts. That was how the Internet worm was
supposed to have worked.
Gary
Casper H.S. Dik - Network Security Engineer
[[ PLEASE DON'T SEND ME EMAIL COPIES OF POSTINGS ]]
"Gary L. Burt" <glb...@toad.net> writes:
>One other point to consider. UNIX can not be expoited as easily by a binary
>executable, since there are so many different underlying type of hardware.
>This sures as a layer of protection (to a degree). However, it can be
>penetrated by using shell scripts. That was how the Internet worm was
>supposed to have worked.
While one of the mechanisms used by teh work was a script (the sendmail
DEBUG one) the other used a stackoverflow in fingerd. This was done
using Vax assembly and was very much architecture dependent.
The sendmail one worked againsts Sun, and combined with the Vax fingerd
exploit, this covered a very large percentage of all internet systems at the
time. (A Sun fingerd exploit was possible, but none was provided with the
worm)
If the internet would once again become a mono culture then the chances
of a worm will be much better once again.
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
idup...@students.uiuc.edu
just wondering why do you use a '*' in *NIX ?
can I say UNIX ?
please explain .. sorry if this is a very simple question.
thanks.
mi...@vaughan.fac.unbc.ca (Mica Currie) writes:
> Wayne Lee (Leewa) (pha...@neosoft.com) wrote:
> : To anyone,
>
> : I was wondering if UNIX platforms are vulnerable to "Virus"
> : programs similar to those that can affect DOS/Windows platforms? I have
> : been told that writing a memory-resident virus program for a UNIX platform
> : is not likely because of the way UNIX's memory management is done. UNIX
> : is more vulnerable to "Trojan Horses" and other things in this genre from
> : what I have been told. How does this compare to Win95 and WinNT? Even
> : though NT and 95 have different memory management schemes, are they
> : vulnerable to virus attacks that you see on DOS/Windows 3.1 platforms? If
> : someone can also give me a brief explaintion of the differences in the way
> : UNIX, WinNT, Win95 and Win3.1 manage memory, this would be appreciated
> : too.
>
> : Thanks,
>
> : Wayne
> In one word. NO. They are not vulnerable in the way that dos and windows are. DOS/Windows is an inheirantly open system with no security. Since *NIX's are by their nature intended to be secure systems, any time a place was found where bits of code could be used to harm or damage the system, the system was fixed to prevent the security problem. This is not to say that there is not holes. Merely that they tend to be patched very quickly once found making any sort of exploit with that hole limited to those wh
Timothy J. Luoma
In <kv7k9bg...@ux5.cso.uiuc.edu> idup...@students.uiuc.edu wrote:
>
> just wondering why do you use a '*' in *NIX ?
> can I say UNIX ?
>
> please explain .. sorry if this is a very simple question.
Unix (it's not an acronym right?) is a trademarked OS, I believe.
The Unix-variants (FreeBSD, Linux, and so on) are not exactly Unix,
technically speaking.
That's why people use the * to stand for "all the various Unix-like OSes
which are out there"
TjL
Jim Dennis
Mica Currie (mi...@vaughan.fac.unbc.ca) wrote:
:: I was wondering if UNIX platforms are vulnerable to "Virus"
:: programs similar to those that can affect DOS/Windows platforms? I have
Wayne,
This question has been discuss several times with
vitriolic fervor.
My own short comments:
It is technically feasible to write programs
that infect other programs under Unix. In
fact it is technologically quite easy to write
such a program.
Unlike DOS (or MacOS) there is no trick at
all to writing a "resident" program (TSR)
under Unix. Any normal Unix user can write
simple shell scripts and invoke them with
the 'nohup' command or (in some shells)
simply run them in the background (usually
by appending an '&' ampersand to the command
line). However, these programs won't be
able to interact (or interfere) with other
users unless those users want to do so
(via named pipes/FIFO's, sockets, temp files
or some other IPC mechanism).
The fact that Unix is a *multi-user* OS
(by design) means that it's multi-tasking
model tends to protect the memory and files
of one user from processes run by most other
users.
This design facet, and various conventions employed
by most system administrators and most software
vendors makes Unix far less vulnerable to
virii than most PC's.
These last two comments are the important ones. Unix
was designed to be a multi-user OS. It was not designed
to be a "secure" OS -- but it implemented a relatively
minimal, lightweight, security model which is suitable
for most reasonable purposes (particularly for relatively
non-hostile groups of users to share the resources of a
given system).
The degree to which these features are "lightweight" and
"minimal" and to which the application of modern Unix
systems are "suitable" to the environments in which they
are commonly used -- is the crux of the whole comp.unix.security
newsgroup. Some versions of Unix have implemented much
more extensive security features than the early designs
called for.
WRT the conventions of sysadmin's and vendors: The fact
that most Unix software is installed by "root" and run
by "users" (non-root accounts) and generally not "writable"
by "users" is the key to questions about virii. The only
"live" Unix virus reports that I've ever heard from the field
were in the Bliss incident last year. This was a simple
demonstration virus that could (and did) infect Linux systems
where the user's (sysadmin's in this case) weren't following
the normal system conventions.
I found the messages regarding Bliss in one of the Linux
alert lists and forwarded them to a friend of mine who
is the chief researcher in computer viruses for McAfee.
(He actually happened to be at my house when I encountered
it). I pointed out that McAfee's existing anti-virus product
for Linux (which is designed to scan your FTP and NFS servers
for PC virii) might be able to detect this "native Linux"
virus.
(Yes, I and many at McAfee were flamed repeatedly for
"making a media fiasco" of the whole event. Yes, Bliss
had a "please remove me switch." Yes, there is ongoing
controversy about "what is a virus" etc. Yes, the
few people who got infected were in gross violation of
proper Unix system administration conventions. No, I
don't want to get into a raging debate regarding this
incident).
This is the only "incident" of which I'm aware. There
have been other feasability demonstrations. (I heard that
one was written as a Bourne shell script). The total
impact of Bliss was negible -- I doubt more than a couple
dozen people where ever infected.
The point is that virii don't propagate well under Unix.
Besides the issues that I've already discussed there is
the matter of communications media. It is relatively
rare to use diskettes with Unix systems. (Other than
PC's and Macs most Unix workstations don't even have floppy
drives). It is also increasingly rare to see tape used
for software installation. This leaves CD-ROM and
network interfaces as the dominate means to install software.
The problem of infecting a CD-ROM should be obvious.
The fact that most TCP/IP networks not particularly good
virus media is less obvious. If I put a package up on
an NFS or FTP server so that others can install it -- they
almost never have write access back to it. Since that
vast majority of users will come to the central distribution
sites (write protected) and not get their software through
secondary (user level) sources -- there is little opportunity
for viral spread (barring specific security vulnerabilities
in the server).
As Mica said -- there is no significant risk of virus
infection under Unix. What little exists can be reduced
greatly by following the conventional administrative
practices that are common to professional SA's.
--
Jim Dennis,
Starshine Technical Services http://ww.starshine.org
STOCK Frédéric
Wayne Lee (Leewa) wrote:
>
> To anyone,
>
> I was wondering if UNIX platforms are vulnerable to "Virus"
> programs similar to those that can affect DOS/Windows platforms? I have
> is not likely because of the way UNIX's memory management is done. UNIX
> is more vulnerable to "Trojan Horses" and other things in this genre from
> what I have been told. How does this compare to Win95 and WinNT? Even
> though NT and 95 have different memory management schemes, are they
> vulnerable to virus attacks that you see on DOS/Windows 3.1 platforms? If
> someone can also give me a brief explaintion of the differences in the way
> UNIX, WinNT, Win95 and Win3.1 manage memory, this would be appreciated
> too.
>
> Thanks,
>
> Wayne
Hello,
I've read all the answers.
There's less probability to get viruses on *UNIX* systems, by their
conception than DOS/Windows (security access, security by files, ...).
But, simply visit http://www.rootshell.com , and you'll learn many
things about *UNIX* systems.
It's about systems failures, but imagine using it to get viruses !!!
Bye.
--
°° °°
Frédéric STOCK
°° °°
Daniel Brown
In article <6b0af5$oj4$2...@ha2.rdc1.nj.home.com>, Timothy J. Luoma wrote:
>In <kv7k9bg...@ux5.cso.uiuc.edu> idup...@students.uiuc.edu wrote:
>> just wondering why do you use a '*' in *NIX ?
>> can I say UNIX ?
<clip>
>Unix (it's not an acronym right?) is a trademarked OS, I believe.
>
>The Unix-variants (FreeBSD, Linux, and so on) are not exactly Unix,
>technically speaking.
>
>That's why people use the * to stand for "all the various Unix-like OSes
>which are out there"
That, and because it's trademarked, and for some people it's an actly of
diety-worship to not fully spell the name (some usenetters use "Ghod", and
"Y*hwey"[1] (sp) is said to be written on the Wailing Wall in Jerusalem).
And no, Unix isn't an acronym -- it's a play on "Multics", the name of
another OS in use at Unix's creation. Aside from multi-user abilities,
the two were relatively unrelated.
[1] Yahwey (sp), which is supposed to be the modern translation for the
name for God, in the Judeo-Christian sense. Formerly, this was Jehovah.
--
Daniel Brown | Linux: The choice of a GNU generation #
cmbrown at aa.net | Real Hackers are Zen programmers: #
Seattle, Washington | http://www.ccil.org/jargon/ #
====================^=======================================~
NOTE!!! Change the FROM: address when sending mail!
Pavel V. Zaitesev
Well,Wayne...
Any operating system is suceptible to viruses. See DOS as a Stoneage type
of civilization, people could die from flu much easier than now, as now
society has many protections against it, but traces of virus can appear
thruaout the world. Same on UNICES. See , *NIX is built so , that it can
be shared in betweenof many users, so In case I have a virus in my piece
of multiuser system , It wouldn't spread to anywhere else. Mostly viruses
are not likely to work under Unixes, unless they are complied and written
for a certain flavour.... while for dos, it just needs to get out... and
then you are at the power of the virus' creator.
In short:
Viruses are possible on Unixes, but due to very nature of unices, they
are like a different animals, so they don't get sick from the same virus.
In DOS they are portable and therefor can easely wipe put to danger data
on many PCs. Get a Linux , get a life... thats what my friends and I say!
Pavel.
/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\_/-\
Hey... don't get mad, get even.
Phone Number is avail. at your request.
Linux:Windoze from Bill:BC++:WIN32API:OWL:ORACLE:XWin wins:g++:libc:TCP/IP
IRC ii:inetd:sendmail:joe:/etc/default/uugetty.ttyS3!!!
! PANIC !: Segment Fault
Kaz Kylheku
In article <6b0af5$oj4$2...@ha2.rdc1.nj.home.com>, Timothy J. Luoma <> wrote:
>In <kv7k9bg...@ux5.cso.uiuc.edu> idup...@students.uiuc.edu wrote:
>>
>> just wondering why do you use a '*' in *NIX ?
>> can I say UNIX ?
>>
>
No, it is the designation of a compatibility standard for constructing
a certain type of OS.
>The Unix-variants (FreeBSD, Linux, and so on) are not exactly Unix,
>technically speaking.
Only due to the technicality of not being branded as such.
Branding requires $$$. Also, it requires careful adherence to the most
minute requirements, obviously.
I don't think that anyone in the freeware world really cares about the
branding these days. The free UNIXes are far more useful out of the box;
the first thing you do with a commercial UNIX workstation is endow it
with enough freeware so that it approaches the usability of FreeBSD or Linux.
Kaz Kylheku
In article <6at46s$5pp$1...@uuneo.neosoft.com>,
Wayne Lee (Leewa) <pha...@neosoft.com> wrote:
>To anyone,
>
> I was wondering if UNIX platforms are vulnerable to "Virus"
>programs similar to those that can affect DOS/Windows platforms? I have
The short answer is that whereas it is possible to write software which
attaches itself to UNIX binaries and spreads, doing so is merely a futile
academic exercise because UNIX systems prevent you from writing to an
executable to which you do not have write permissions.
A reasonable UNIX user does not log in as root to do his or her ordinary
work, and does not allow system binaries to be modified by anyone other
than the superuser.
Thus the virus hacker faces the formidable task of not only writing
self-replicating code, but also endowing such code with the ability to
automatically hack root first.
Furthermore, the virus writer's problem is confounded by the portability
of UNIX; viruses tend to be written in assembly language, or are otherwise
``close to the machine''. The mechanisms of infection exploit particular
executable file formats.
>been told that writing a memory-resident virus program for a UNIX platform
>is not likely because of the way UNIX's memory management is done. UNIX
Right, UNIX does not have memory-resident programs. But modern UNIXes do have
loadable device drivers, which are the closest thing UNIX has to MS-DOS TSR
programs. YOu could write a malicious UNIX device driver that messes with the
kernel. But again, you face the problem of permissions (goading the superuser
into loading your evil driver), and portability problems.
Part of the reason why you don't see UNIX viruses is thanks to the superior
levels of clue possession on the part of people who run UNIX boxes; they
refuse to accept foreign binaries, and don't run suspicious programs as root
Also, UNIX freeware comes in source code form, whereas the PC world is
crawling with virus-infested, binary-only ``shareware'' programs.
/dev/zero
To an extent, yes it is possible. Any user can write a shell script that
copies and appends itself to other files that in that user's directory,
but the only way for the program to affect other users is if those users
make there files world-writeable (most reasonable Unix users would agree
that that would be a pretty mindless thing to do unless it's for a good
reason). The only person who has access to EVERYTHING is the root user,
and s/he could write a program which would have the same priveleges as
him/her that could infect other programs on the system, but most
administrators are pretty trustworthy, and I don't see why a person with
their own Unix box would want to do that to their own system for purposes
other than proving that it could be done.
The only major incident that I can think of offhand in which a malignant
program infected a large number of machines was the incident with the RTM
worm back in 1988, but that program (which was a worm, not a virus)
exploited a flaw in the version of Sendmail that was used by Sun3 and VAX
machines of that time, and thus it wasn't a virus per se. Look in some
newspapers from around the 5th of November of that year (I think, I'm not
certain of the date) for more information. There was also the Bliss
experiment last year, in which a virus was written and spread on a few
Linux machines just to prove that Unix wasn't completely impenetrable.
But anyway, if you're asking this experiment because you want to get a
Unix box but you're afraid of viruses, you really shouldn't have anything
to worry about, since Bliss incident was only an experiment and the RTM
worm exploited a security hole in Sendmail which has long since been
fixed. If you are intent upon getting a Unix box, I would recommend
FreeBSD (www.freebsd.com). You can buy a 4-CD set from Walnut Creek
CD-ROM for $39.95 (that's for the cost of putting the OS on the CD, not
the OS itself :). Go to www.cdrom.com for details.
-Eric-
---------------------------------------------------------
| "Avoid all needle drugs - | Eric DeArment |
| The only dope worth | e...@efn.org |
| shooting is Richard | e...@m-net.arbornet.org |
| Nixon." -Abbie Hoffman, | Team Ada "member" & |
| "Steal This Book" | Socialist |
---------------------------------------------------------
/dev/zero
"Unix" isn't an acronym now, but it used to be. Originally, it was
spelled "UNICS," which stood for "UNiversal Computer System" :)
Eventually, they took the acronym's meaning away and replaced the "CS"
with an "X"
Toby Creek
/dev/zero wrote:
>
> "Unix" isn't an acronym now, but it used to be. Originally, it was
> spelled "UNICS," which stood for "UNiversal Computer System" :)
> Eventually, they took the acronym's meaning away and replaced the "CS"
> with an "X"
I've seen this stated before, but it is not correct. UNIX was a weak
pun on Multics, a long forgotten dead project OS at AT&T.
Toby
--
Toby Creek - UNIX System Administrator
Alcatel Telecom, 2912 Wake Forest Road, Raleigh, NC
Phone +1 919 850 6465 / Fax +1 919 850 6703
mailto:cr...@aur.alcatel.com
Eye Vee
In article <34E33C31...@aur.alcatel.com> Toby Creek <cr...@aur.alcatel.com> writes:
>/dev/zero wrote:
>>
>> "Unix" isn't an acronym now, but it used to be. Originally, it was
>> spelled "UNICS," which stood for "UNiversal Computer System" :)
>> Eventually, they took the acronym's meaning away and replaced the "CS"
>> with an "X"
>
>I've seen this stated before, but it is not correct. UNIX was a weak
>pun on Multics, a long forgotten dead project OS at AT&T.
>
>Toby
From _Unix for the Impatient_ by Abrhams and Larson
The first version of UNIX, called Unics, was writen in 1969 by Ken
Thompson at Bell laboratories in Murray Hill, New Jersey. It ran on a
Digital Equipment PDP-7 computer. Bell Labs had been involved...in the
deveolpment of the Multics system, a large, highly capable time sharing
system...Thompson and his colleagues admired the capabilities of the
Multics system but felt it was far too complicated.
You decide......
(is Ken Thompson still alive?? If so, what's his e-mail address?? ;-)
-=scott=-
--
well | Scott Ivancic call...@en.com
whatever | http://www.en.com/users/callmeox
nevermind | (..and Kam Fong as Chin Ho)
Neil.Frankli...@ccw.ch
/dev/zero <e...@efn.org> claimed:
>"Unix" isn't an acronym now, but it used to be. Originally, it was
>spelled "UNICS," which stood for "UNiversal Computer System" :)
>Eventually, they took the acronym's meaning away and replaced the "CS"
>with an "X"
Error.
It stood for Uniplexing Computing System.
It was a pun on MULTICS, Multiplexing (= Multitasking) Computing System.
--
Neil.Frankli...@ccw.ch, http://www.ccw.ch/Neil.Franklin/
for Geek Code, Papernet, Voicenet, PGP public key see http:
Mac, 95 and NT users are CLUEless (Command Line User Environment)
If I go missing, its once again my newsfeed that has craped
P Kumar
Toby Creek wrote:
>
> /dev/zero wrote:
> >
> > "Unix" isn't an acronym now, but it used to be. Originally, it was
> > spelled "UNICS," which stood for "UNiversal Computer System" :)
> > Eventually, they took the acronym's meaning away and replaced the "CS"
> > with an "X"
>
> pun on Multics, a long forgotten dead project OS at AT&T.
Hey, I read something "interesting" that said UNIX came from "eunuchs"
because it was "castrated Multics" <grin grin>
Donald Campbell
In article <34E33C31...@aur.alcatel.com>, Toby Creek
<cr...@aur.alcatel.com> writes
>/dev/zero wrote:
>>
>> "Unix" isn't an acronym now, but it used to be. Originally, it was
>> spelled "UNICS," which stood for "UNiversal Computer System" :)
>> Eventually, they took the acronym's meaning away and replaced the "CS"
>> with an "X"
>
>I've seen this stated before, but it is not correct. UNIX was a weak
>pun on Multics, a long forgotten dead project OS at AT&T.
>
I understand that they did produce Multics and it did/does run on a
couple of IBM mainframes, although it never escaped from the University
environment.
Regards
Donald Campbell
Matt Ryan
In article <i8AlpEAQGM60Ew$c...@suedon.co.uk>,
Donald Campbell <Don...@suedon.co.uk> writes:
>
> I understand that they did produce Multics and it did/does run on a
> couple of IBM mainframes, although it never escaped from the University
> environment.
>
I think you're wrong - see http://www.best.com/~thvv/sites.html
Matt.
--
Matt Ryan - Network Engineer ma...@planet.net.uk
Planet OnLine Ltd, The White House, Tel: +44 113 2345566
Melbourne Street, Leeds, LS2 7PS, UK Fax: +44 113 2240003
James Youngman
>>>>> "Donald" == Donald Campbell <Don...@suedon.co.uk> writes:
Donald> I understand that they did produce Multics and it did/does
Donald> run on a couple of IBM mainframes, although it never escaped
Donald> from the University environment.
Yes it does. Search the Web. The NSA still run Multics on at least
one machine.
Donald Campbell
In article <6cbsjs$mi4$1...@svr-c-02.core.theplanet.net>, Matt Ryan
<ma...@banana.planet.net.uk> writes
>In article <i8AlpEAQGM60Ew$c...@suedon.co.uk>,
> Donald Campbell <Don...@suedon.co.uk> writes:
>>
>> I understand that they did produce Multics and it did/does run on a
>> couple of IBM mainframes, although it never escaped from the University
>> environment.
>>
>
>I think you're wrong - see http://www.best.com/~thvv/sites.html
>
>
>Matt.
>
Interesting I have just had a couple of mails saying that "The NSA still
run Multics on at least one machine".
Regards
Donald Campbell
John Andrea
> I understand that they did produce Multics and it did/does run on a
> couple of IBM mainframes, although it never escaped from the University
> environment.
some military sites used (and continue to do so) Multics because it
is very secure. See http://www.best.com/~thvv/multics.html
Benjamin J. Robinson
In "[Q] Viruses Under UNIX," Toby Creek wrote:
>/dev/zero wrote:
>>
>> "Unix" isn't an acronym now, but it used to be. Originally, it was
>> spelled "UNICS," which stood for "UNiversal Computer System" :)
>> Eventually, they took the acronym's meaning away and replaced the "CS"
>> with an "X"
>
>I've seen this stated before, but it is not correct. UNIX was a weak
>pun on Multics, a long forgotten dead project OS at AT&T.
Although it didn't generate much at the box office, my understanding is
that Multics was more than just a "dead project OS." It supposedly
innovated such things as security levels (i.e., a privileged kernel
level vs. a non-privileged user level), among other things.
There was a commercial release of Multics. One of the seven dwarfs
(GE? Honeywell? Sorry, I forget) ran it.
-Benjamin Robinson
--
"The government cannot reduce the adult population to reading or viewing only
to what is appropriate for children." -- Bruce Ennis
My opinions don't represent, and are likely contrary to, those of my employer.
This message may or may not contain sarcastic content; your burden to decide.
Eppa
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Wasnt UNIX some sort of poor pun on unucs, the ancient noblemen <?>
with erm some _disadvantage_ 8^/ Referring to the idea that UNIX
was "missing something" ?
Hell what do I know
Mark
- -
"There is hopeful symbolism in the fact that flags do not wave in a
vacuum." -- Arthur C. Clarke
http://www.aber.ac.uk/~mch6 - PGP5.0i Key
On 19 Feb 1998:
> In "[Q] Viruses Under UNIX," Toby Creek wrote:
> >/dev/zero wrote:
> >>
> >> "Unix" isn't an acronym now, but it used to be. Originally, it was
> >> spelled "UNICS," which stood for "UNiversal Computer System" :)
> >> Eventually, they took the acronym's meaning away and replaced the "CS"
> >> with an "X"
> >
> >I've seen this stated before, but it is not correct. UNIX was a weak
> >pun on Multics, a long forgotten dead project OS at AT&T.
-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
Charset: cp850
iQA/AwUBNOy9lPT5mk5IGzm5EQI9AwCg/hpiND2KnjyKUdoBbtSME/HSD98AoLCy
Q6yA/uwUaYaoq9xeoVh33kvM
=z2C/
-----END PGP SIGNATURE-----
Dave Turner
Eppa <mc...@aber.ac.uk> wrote:
>> In "[Q] Viruses Under UNIX," Toby Creek wrote:
>> >/dev/zero wrote:
>> >> "Unix" isn't an acronym now, but it used to be. Originally, it was
>> >> spelled "UNICS," which stood for "UNiversal Computer System" :)
>> >> Eventually, they took the acronym's meaning away and replaced the "CS"
>> >> with an "X"
>> >
>> >I've seen this stated before, but it is not correct. UNIX was a weak
>> >pun on Multics, a long forgotten dead project OS at AT&T.
>Wasnt UNIX some sort of poor pun on unucs, the ancient noblemen <?>
>with erm some _disadvantage_ 8^/ Referring to the idea that UNIX
>was "missing something" ?
In an AT&T Bell Laboratories Technical Journal (October 1984, Vol. 63. No. 8,
Part 2) Dennis Richie wrote an article titled "The Evolution of the UNIX
Time-Sharing System."
In it he describes the Labs use of Multics, the decision to stop Multics
development and the group of holdouts who started in 1969 to look for
an alternative to it. In 1969, Ken Thompson developed the game of
"Space Travel", first writen on Multics, later in Fortran for GECOS. It
cost about $75 for CPU time on the big computer. Before long, he
found a little-used PDP-7 and Thompson and Ritchie rewrote Space
Travel for it (it was assembled on under GECOS and the paper tape output
was carried to the PDP-7. Soon Thompson began to implement a file
system that had been designed earlier and then added additonal requirements
to make a working operating system, including processes. A few utility
commands and the shell soon followed. This was all done using the GECOS
cross-assembler. Once an assembler was completed the system became
self supporting.
"Although it was not until well into 1970 that Brian Kernighan suggested the
name 'UNIX,' in a somewhat treacherous pun on 'Multics,' that the
operating system we know today was born."
One of the references refers to "The Multiplexed Information and Computing
Service; Programmer's Manual" which is probably the source of the acronym
MULTICS.
Regards,
Dave
Bob Voll
On Mon, 16 Feb 1998 22:45:36, Donald Campbell
<Don...@suedon.co.uk> wrote:
> [...]
> I understand that they did produce Multics and it did/does run on a
> couple of IBM mainframes, although it never escaped from the University
> environment.
Aaaargh!
1. General Electric produced the 645 processor and its
associated Multics operating system in the mid- to late-1960s.
(GE's computer division was bought by Honeywell in '70 or '71;
the hardware was modified and became the 6180, still later the
DPS/8 SomethingOrOther--DPS8/70M?.) The most significant players
in the development were GE, AT&T, Honeywell, MIT and ARPA.
2. They were a commercial success: business, academic and
government sites. (Since these were very large, very powerful
systems with unprecedented security, both of the hardware and the
operating system, it _was_ sort of a niche product for those
organizations with special needs.)
3. You can still find them in use today. The NSA's National
Computer Security Center has one that is publicly known.
4. Multics was the first system to earn a B2 security rating.
--
Bob Voll 203 328 5162
WordPROse, Inc. 203 264 3910
Bruce Barnett
dmtu...@primenet.com (Dave Turner) writes:
> "Although it was not until well into 1970 that Brian Kernighan suggested the
> name 'UNIX,' in a somewhat treacherous pun on 'Multics,' that the
> operating system we know today was born."
The treacherous pun, for those who still don't get it, is that UNIX is
a "castrated version of MULTICS".
--
Bruce <barnett at crd. ge. com> (speaking as myself, and not a GE employee)
cr...@synapse-group.com
> The treacherous pun, for those who still don't get it, is that UNIX is
> a "castrated version of MULTICS".
The (ahem) "official version" is that UNIX was chosen because "MULTICS did
many things badly, and UNIX did one thing well." The "one thing well" was
probably the orbital simulator/space game that Thompson couldn't get to run
under MULTICS.
Would that all projects begun so frivolously should turn out so well....
Craig Mertens
craig<at>synapse-group<dot>com
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
Bill Vermillion
In article <i8AlpEAQGM60Ew$c...@suedon.co.uk>,
Donald Campbell <Don...@suedon.co.uk> wrote:
>In article <34E33C31...@aur.alcatel.com>, Toby Creek
><cr...@aur.alcatel.com> writes
>>> "Unix" isn't an acronym now, but it used to be. Originally, it was
>>> spelled "UNICS," which stood for "UNiversal Computer System" :)
>>> Eventually, they took the acronym's meaning away and replaced the "CS"
>>> with an "X"
>>I've seen this stated before, but it is not correct. UNIX was a weak
>>pun on Multics, a long forgotten dead project OS at AT&T.
>I understand that they did produce Multics and it did/does run on a
>couple of IBM mainframes, although it never escaped from the University
>environment.
Weren't those GE computers? For a company who only made computers
for 14 years, GE really had a lot of firsts -eg packet switched
networks, computer to computer communication via satellite, and our
checks still have the MICR bank encoding on the bottom that they
developed along with SRI. And the middle field of the password
file is still called the GECOS field.
--
bi...@bilver.magicnet.netREMOVETHIS | bi...@bilver.comREMOVETHIS
(Remove the anti-spam section from the address on a mail reply)
Bill Vermillion
In article <6chbuv$rlq$1...@maya.ece.ufl.edu>,
Benjamin J. Robinson <afn4...@freenet2.afn.org> wrote:
>In "[Q] Viruses Under UNIX," Toby Creek wrote:
>>>
>>> "Unix" isn't an acronym now, but it used to be. Originally, it was
>>> spelled "UNICS," which stood for "UNiversal Computer System" :)
>>> Eventually, they took the acronym's meaning away and replaced the "CS"
>>> with an "X"
>>
>>I've seen this stated before, but it is not correct. UNIX was a weak
>>pun on Multics, a long forgotten dead project OS at AT&T.
>
>that Multics was more than just a "dead project OS." It supposedly
>innovated such things as security levels (i.e., a privileged kernel
>level vs. a non-privileged user level), among other things.
>There was a commercial release of Multics. One of the seven dwarfs
>(GE? Honeywell? Sorry, I forget) ran it.
GE, which then was sold to Honeywell, which was then sold to ???
- a European manfacaturer. I want to say Bull, but I think that is
wrong.
For an interesting look at GE, see the IEEE book published last
year by IEEE press "The King of the Seven Dwarfs". GE's biggest
failure to succeed was a management that didn't think they should
be in computers, yet they had a lot of innovations that are still
with us today.
Mark Greene
Bill Vermillion wrote:
>
> In article <6chbuv$rlq$1...@maya.ece.ufl.edu>,
> Benjamin J. Robinson <afn4...@freenet2.afn.org> wrote:
> >In "[Q] Viruses Under UNIX," Toby Creek wrote:
> >>/dev/zero wrote:
> >>>
> >>> "Unix" isn't an acronym now, but it used to be. Originally, it was
> >>> spelled "UNICS," which stood for "UNiversal Computer System" :)
> >>> Eventually, they took the acronym's meaning away and replaced the "CS"
> >>> with an "X"
> >>
> >>I've seen this stated before, but it is not correct. UNIX was a weak
> >>pun on Multics, a long forgotten dead project OS at AT&T.
> >
> >Although it didn't generate much at the box office, my understanding is
> >that Multics was more than just a "dead project OS." It supposedly
> >innovated such things as security levels (i.e., a privileged kernel
> >level vs. a non-privileged user level), among other things.
>
> >There was a commercial release of Multics. One of the seven dwarfs
> >(GE? Honeywell? Sorry, I forget) ran it.
>
> GE, which then was sold to Honeywell, which was then sold to ???
> - a European manfacaturer. I want to say Bull, but I think that is
> wrong.
No, not wrong, entirely correct. And Bull, following the trend, sold the
hardware business a couple of years ago. However, can't remember to
whom. Any one else want to contribute?
>
> For an interesting look at GE, see the IEEE book published last
> year by IEEE press "The King of the Seven Dwarfs". GE's biggest
> failure to succeed was a management that didn't think they should
> be in computers, yet they had a lot of innovations that are still
> with us today.
>
> --
> bi...@bilver.magicnet.netREMOVETHIS | bi...@bilver.comREMOVETHIS
> (Remove the anti-spam section from the address on a mail reply)
--
Mark Greene My real address: mailto:gree...@hlthsrc.com
The above opinions are mine, not my employer's.
GLE
Mark Greene wrote:
> > >There was a commercial release of Multics. One of the seven dwarfs
> > >(GE? Honeywell? Sorry, I forget) ran it.
> >
> > GE, which then was sold to Honeywell, which was then sold to ???
> > - a European manfacaturer. I want to say Bull, but I think that is
> > wrong.
>
> No, not wrong, entirely correct. And Bull, following the trend, sold the
> hardware business a couple of years ago. However, can't remember to
> whom. Any one else want to contribute?
>
I think it was a Japanese company.
Fujitsu or NEC, or something.
--
------------------------------------------------------------------------
My real gle (Sorry for the inconvenience
email @ but I was getting far too
address mail.dotcom.fr much spam to handle it)
------------------------------------------------------------------------
- spammers, add us to your list:
root@localhost postmaster@localhost webmaster@localhost
abuse@localhost ab...@aol.com
ask...@microsoft.com pres...@whitehouse.gov