Any GIMP users (Linux)

J. Clarke wrote:

Sounds to me like you're grasping at a straw to find something to complain
about.

Has there ever been a case of a Windows system being compromised by
extracting the password list via a buffer overrun then cracking the
password list using a rainbow table attack?

As far as I know, no. It has never been necessary to
go to all that trouble.

The basic problem is that Windows was *designed* to be
insecure. The notion was to enable outsiders, including
Microsoft, to download and run programs on a machine without
user permission.

This was meant as a feature. It also meant that companies
could enforce rights management again without user permission.

This was known at the time to be an *awful* design, but
Microsoft knew better.

Ever since Microsoft has been attempting to build security
into a badly designed original OS.

One hopes that the next version will be better.


I'll also add that depending on users using long random
sequences of characters as passwords and similar good
ideas (seriously) are doomed to failure because user's
simply won't do that.

A good security model has to take that into account too.

--- Paul J. Gans

"J. Clarke" writes:

Måns Rullgård wrote:

"J. Clarke" writes:

Måns Rullgård wrote:

"J. Clarke" writes:

Måns Rullgård wrote:

"David J. Littleboy" writes:

"Richard Kettlewell" wrote:
"J. Clarke" writes:

I'm not sure why you're on about this though. Any decent system
locks out attempted logins after a few failures, so rainbow table
attacks don't work for gaining access.

This kind of brute-force attack does not attempt to login, but works
directly on the encrypted passwords.

How do you get access to encrypted passwords without logging in?

You might be able to trick some server into sending them, for instance
using a buffer overflow attack.

At that point you've already busted the server so why do you need them?

Perhaps all you can use that bug for is reading files already on the
system. Getting the passwords will allow you full access.

Sounds to me like you're grasping at a straw to find something to
complain about.

Has there ever been a case of a Windows system being compromised by
extracting the password list via a buffer overrun then cracking the
password list using a rainbow table attack?

I don't know of any cases where it was actually used. However, I do
recall there being a bug in the form handling in some web browser
whereby it could be made to send any file as a POST upload.

Any file, or just those to which the user had access?

No difference on most windows machines.

--
Måns Rullgård

"David J. Littleboy" writes:
"Richard Kettlewell" wrote:
"J. Clarke" writes:

I'm not sure why you're on about this though. Any decent system
locks out attempted logins after a few failures, so rainbow table
attacks don't work for gaining access.

This kind of brute-force attack does not attempt to login, but works
directly on the encrypted passwords.

How do you get access to encrypted passwords without logging in?

Steal a backup tape. Exploit a bug which gives you RO access to the
relevant file. Bribe or otherwise coerce someone. Crack a different
system where someone uses the same password. Take a copy before you
got fired. If it wasn't possible, nobody would need to invent
password encryption schemes...

--
http://www.greenend.org.uk/rjk/

Måns Rullgård wrote:

"J. Clarke" writes:

Måns Rullgård wrote:

"J. Clarke" writes:

Måns Rullgård wrote:

"J. Clarke" writes:

Måns Rullgård wrote:

"David J. Littleboy" writes:

"Richard Kettlewell" wrote:
"J. Clarke" writes:

I'm not sure why you're on about this though. Any decent system
locks out attempted logins after a few failures, so rainbow table
attacks don't work for gaining access.

This kind of brute-force attack does not attempt to login, but
works directly on the encrypted passwords.

How do you get access to encrypted passwords without logging in?

You might be able to trick some server into sending them, for
instance using a buffer overflow attack.

At that point you've already busted the server so why do you need
them?

Perhaps all you can use that bug for is reading files already on the
system. Getting the passwords will allow you full access.

Sounds to me like you're grasping at a straw to find something to
complain about.

Has there ever been a case of a Windows system being compromised by
extracting the password list via a buffer overrun then cracking the
password list using a rainbow table attack?

I don't know of any cases where it was actually used. However, I do
recall there being a bug in the form handling in some web browser
whereby it could be made to send any file as a POST upload.

Any file, or just those to which the user had access?

No difference on most windows machines.

Only those on which users run as administrator.

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)

Richard Kettlewell wrote:

"David J. Littleboy" writes:
"Richard Kettlewell" wrote:
"J. Clarke" writes:

I'm not sure why you're on about this though. Any decent system
locks out attempted logins after a few failures, so rainbow table
attacks don't work for gaining access.

This kind of brute-force attack does not attempt to login, but works
directly on the encrypted passwords.

How do you get access to encrypted passwords without logging in?

Steal a backup tape. Exploit a bug which gives you RO access to the
relevant file. Bribe or otherwise coerce someone. Crack a different
system where someone uses the same password. Take a copy before you
got fired. If it wasn't possible, nobody would need to invent
password encryption schemes...

If someone is willing to go to that amount of trouble they're going to get
in no matter what you do. Salt or no salt is going to make no real
difference at that level.

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)

"J. Clarke" writes:
Richard Kettlewell wrote:

Steal a backup tape. Exploit a bug which gives you RO access to
the relevant file. Bribe or otherwise coerce someone. Crack a
different system where someone uses the same password. Take a copy
before you got fired. If it wasn't possible, nobody would need to
invent password encryption schemes...

If someone is willing to go to that amount of trouble they're going
to get in no matter what you do. Salt or no salt is going to make
no real difference at that level.

So the whole thing is just a ruse to keep underemployed cryptographers
off the street? Right...

--
http://www.greenend.org.uk/rjk/

Richard Kettlewell wrote:

"J. Clarke" writes:
Richard Kettlewell wrote:

Steal a backup tape. Exploit a bug which gives you RO access to
the relevant file. Bribe or otherwise coerce someone. Crack a
different system where someone uses the same password. Take a copy
before you got fired. If it wasn't possible, nobody would need to
invent password encryption schemes...

If someone is willing to go to that amount of trouble they're going
to get in no matter what you do. Salt or no salt is going to make
no real difference at that level.

So the whole thing is just a ruse to keep underemployed cryptographers
off the street? Right...

There are applications in which it is necessary to "keep ahead of the
Joneses" so to speak, but generally neither Linux or Windows would be used
in such situations.

--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)

"joe mama" wrote in
om:

hi,

i am muther-f'ing sick of windows, bill gates, and redmond, wa. as
well. i want to migrate over to linux, but need to know if the gimp is
even close to PS CS2 in quality. my main concern is being able to use
layers via PS, and curves. I don't use too many filters, and the soft
focus, Gausiann blur ones seem to be inthe gimp.

thanks for any help....

Hi,

A little late to the party, but check this one out: 32bit, layers and
colour management: http://www.koffice.org/krita/

On Sun, 14 May 2006 04:53:26 +0000 (UTC), Paul J Gans wrote:

"As for updates/upgrades and security -- far better than windows"

Never has been the case. At least if you disregard 95x, 98x+ products which
never had security in mind from its outset. In fact IP was not a default,
and was added later after the Internet took off. How many years did NT run
before its first breach? A breach enabled via a hacked Unix server/s.

This is incorrect. NT was good but never a real target for
hackers. Windows XP started out terribly and had a totally
insecure design. The idea that an outside agent can download
a program into your computer and run it without your intervention
is (a) a Microsoft idea incorporated in Windows and (b) a horrible
security hole that was *known* to be a horribe hole before Microsoft
used it.


As to haxoring Linux? LOL, from its inception it was rootable, yes, things
have tightened up considerably and now they tighten up the software as
well, why? In the hopes we won't have fun. If Linux users are like Windows
Users, and most are nowadays, they don't bother to get the security fixes
so their systems are as open as any others.

Wrong. The only real change in the Linux security model was to
turn on all security by default. That was done some time ago.
It is not easy to hack into a Linux system.

Further, if one does, it is usually not by exploiting a hole
in the system. That's basically not done. The reason is that
the few holes get fixed within hours, days at the most.

Most break-ins are due to guessable passwords and the like. Even
so, if you break into a user account, that does NOT give you access
to the entire system. You can screw the user but you won't bring
the system down.

Hope you're right, but I came across a message the other day from
a couple of months ago in alt.comp.hardware.amd.x86-64 that's a bit
chilling in its premise that new technology will be arriving that in
the right/wrong hands of a select few may give total access to any
computers not isolated from the internet or local networks. Here's
part of the msg:

=============================================

NNTP-Posting-Date: Sat, 25 Mar 2006 21:39:44 -0600
From: billy
Newsgroups: alt.comp.hardware.amd.x86-64
Subject: Will Intel catch up to AMD this year?
Date: Sat, 25 Mar 2006 19:40:35 -0800
Message-ID:

snippage

Intel quietly adds DRM to new chips
Friday 27 May 2005 - 11:02


http://www.digitmag.co.uk/news/index.cfm?NewsID=4915


Microsoft and the entertainment industry's holy grail of
controlling copyright through the motherboard has moved a
step closer with Intel Corp. now embedding digital rights
management within in its latest dual-core processor Pentium
D and accompanying 945 chipset.



Officially launched worldwide on the May 26, the new
offerings come DRM-enabled and will, at least in theory,
allow copyright holders to prevent unauthorized copying and
distribution of copyrighted materials from the motherboard
rather than through the operating system as is currently the
case.

While Intel steered clear of mentioning the new DRM
technology at its Australian launch of the new products,
Intel's Australian technical manager Graham Tucker publicly
confirmed Microsoft-flavored DRM technology will be a
feature of Pentium D and 945.

"[The] 945g [chipset] supports DRM, it helps implement
Microsoft's DRM ... but it supports DRM looking forward,"
Tucker said, adding the DRM technology would not be able to
be applied retrospectively to media or files that did not
interoperate with the new technology.

However, Tucker ducked questions regarding technical details
of how embedded DRM would work saying it was not in the
interests of his company to spell out how the technology in
the interests of security.

The situation presents an interesting dilemma for IT
security managers as they may now be beholden to
hardware-embedded security over which they have little say,
information or control.

Conversely, Intel is heavily promoting what it calls "active
management technology" (AMT) in the new chips as a major
plus for system administrators and enterprise IT. Understood
to be a sub-operating system residing in the chip's
firmware, AMT will allow administrators to both monitor or
control individual machines independent of an operating
system.

Additionally, AMT also features what Intel calls "IDE
redirection" which will allow administrators to remotely
enable, disable or format or configure individual drives and
reload operating systems and software from remote locations,
again independent of operating systems. Both AMT and IDE
control are enabled by a new network interface controller.

"We all know our [operating system] friends don't crash that
often, but it does happen," Tucker said.

Intel's reticence to speak publicly about what lies under
the hood of its latest firmware technology has also prompted
calls to come clean from IT security experts, including
Queensland University of Technology's assistant dean for
strategy and innovation, IT faculty, Bill Caelli.

"It's a dual use technology. It's got uses and misuses.
Intel has to answer what guarantees it is prepared to give
that home users are safe from hackers. Not maybes,
guarantees".

Caelli said it was "critical Intel comes clean" about how
the current DRM technology is embedded into the new CPU and
chipset offering.

Microsoft was unavailable for comment at press time.

=============================================

Sounds like nothing less than an undetectable back door buried in
the hardware, to allow corporate and gov't snoops to do more than
just administer networks, but to also allow clandestine remote
examination and possibly manipulation of the hardware and software
contents of computers. This new hardware could *really* be useful
to the "right" people if it is installed in voting machines.
Both MS and Intel have in the past been caught designing "stealth
snoop" features in their software and hardware.

"Say no more, say no more." -- John Cleese