🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.security.misc
Article #891

Re: Security vs usefulness (was Re: reasons

#891
From: Olaf Titz
Date: Thu, 17 Dec 1992 11:25
85 lines
4831 bytes
In article <1992Dec16.173049.19678@oracle.us.oracle.com> mfriedma@uucp
(Michael Friedman) writes:

>>I really don't know about installations where the security people are
>>officers...
>
>Personally, I don't think that you know about any installation where
>there is any need for security, period.

You surely may be right. There is no need for this kind of "security"
they're talking about here, *here*. I'll state more precisely.

>>You run into this as soon as you start thinking, "the less I allow to
>>the users the less damage can/will they do", which is wrong, imho.
>>But if the changes involve (e.g.) hiding an
>>entire subnetwork behind a firewall, this *does* affect the users. And
>...
>
>Olaf, do you really believe that an academic site should put computers
>that students can access on the same network as the ones containing
>grades, financial data, and medical records without putting in major
>firewalls between the networks?

No, but:

1. "Students can access": You risk of getting me angry if you put up
the implication "student => security risk". I really do not want to
fire up a flame war, and I well recognize that there are some people
who are students and who are a security risk, especially when it comes
to networks (don't tell me, we had an incident like this just
yesterday :-( ) But the kind of thinking, "we put students on this
machine, therefore we have implemented stiff security, tell me if
something doesn't work anymore"(*) - the kind of labeling ANY student
a potential network disrupter and malicious cracker *in the first
place* - this discriminatory thinking against students IN GENERAL, is
a major problem, imho. It is "guilty-until-proven-innocent" thinking.

(*) This is an actual case. Exactly this message has been issued to
all NON-student users of the system I'm on now, about a year ago. And
they still wonder why students who accidentally heard of this message
feel pissed off.

2. Firewalls where firewalls are due, but not between two neighbouring
departments at the same university (also an actual case).  (Don't get
me wrong - not a medical department; rather two branches of the
Computer Science Dept.)  Sites that *have* sensitive data are to
protect *themselves*, and if you boil it down to firewalls, then they
should firewall themselves.  Turned around: Does a machine whose only
work is to maintain a database about medical records for exclusive use
in that hospital have to be hooked on a world-wide network which is
known for its lack of inherent security, which is known for many
people being around who can be malicious (NOT ALL OF THEM BEING
STUDENTS) and for a general open-ness of services which implies that
there are many services most people don't know of? I doubt.  And I've
never said that all machines are to be treated equal regardless of the
data they work on. But just a little care has to be applied on either
side.

3. Disrupting users: yet another actual case. You wonder why that
program didn't compile and discover that <curses.h>, like almost any
other file on that machine, is mode 640 and you're not in the right
group, because all "security" on this Unix system is based on groups
(not a bad idea at all, but it can be implemented more or less
reasonably.) Security????
This is the kind of cases I thought of with my statement that a
certain misconception of "security", not security in general, serves
no good, especially in places that have no inherent need for enhanced
security.

I'm very concerned about security in general, e.g. for privacy
matters. Something is absolutely necessary in this field. But for
security concerns to be accepted by the users, the implementation has
to be done carefully so that legitimate use is NOT disrupted, and this
makes me worry insofar as I from my (admittedly limited) experience
know of more cases where this went wrong than where it went right.
Unfortunately.

Olaf


-- 
| Olaf Titz - comp.sc.student  |   o     | uknf@dkauni2.bitnet | old address |
| univ. of karlsruhe - germany |  _>\ _  | s_titz@ira.uka.de   | is still    |
| +49-721-60439                | (_)<(_) | praetorius@irc      | valid       |
  "My heart is human - my blood is boiling - my brain IBM" - Mr. Roboto
Technical Details
Message-ID: <1gpo44INNe27@iraul1.ira.uka.de>
Path: arc1-ca.pugleaf.net!sparky!uunet!mcsun!Germany.EU.net!ira.uka.de!uka!s_titz
References: <1992Dec14.211255.15839@lambda.msfc.nasa.gov> <1gkpsmINNn05@iraul1.ira.uka.de> <1992Dec16.173049.19678@oracle.us.oracle.com>

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!mcsun!Germany.EU.net!ira.uka.de!uka!s_titz
From: s_titz@ira.uka.de (Olaf Titz)
Newsgroups: comp.security.misc,alt.comp.acad-freedom.talk
Subject: Re: Security vs usefulness (was Re: reasons
Date: 17 Dec 1992 11:25:56 GMT
Organization: Fachschaft math/inf, Uni Karlsruhe, FRG
Lines: 85
Message-ID: <1gpo44INNe27@iraul1.ira.uka.de>
References: <1992Dec14.211255.15839@lambda.msfc.nasa.gov> <1gkpsmINNn05@iraul1.ira.uka.de> <1992Dec16.173049.19678@oracle.us.oracle.com>
NNTP-Posting-Host: irau31.ira.uka.de