Cryptography-Digest Digest #191, Volume #11 Thu, 24 Feb 00 06:13:01 EST
Contents:
Re: Passwords secure against dictionary attacks? (Jens Haug)
Re: Passwords secure against dictionary attacks? (Jens Haug)
Re: Processor speeds. ("Clockwork")
Some of the U.S. government programs in business enterprises are ("Markku J.
Saarelainen")
Re: Crypto enthusiasm (wtshaw)
Assistance is needed :) (Nemo psj)
Re: EOF in cipher??? (lordcow77)
Re: Swapfile Overwriter: R.I.P. (Jim)
Goodbye Swapfile Overwrite? (Jim)
Re: How Useful is Encryption as Long as NSA Exists? (Jim)
Crypto Speeds... (John)
Re: OAP-L3 Encryption Software - Complete Help Files at web site (lordcow77)
Re: Crypto Speeds... (Runu Knips)
Re: Passwords secure against dictionary attacks? (Tom Holroyd)
Re: Passwords secure against dictionary attacks? ([EMAIL PROTECTED])
Re: e-payment suggestion ("Dr.Gunter Abend")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Jens Haug)
Crossposted-To: comp.security.misc,alt.security.pgp
Subject: Re: Passwords secure against dictionary attacks?
Date: 24 Feb 2000 07:19:20 GMT
Reply-To: [EMAIL PROTECTED]
In article <MSUs4.45$VM6.2124@burlma1-snr2>, Barry Margolin <[EMAIL PROTECTED]>
writes:
> Going back to the original poster's question, I have this comment: using
> two words (assuming all the characters are significant) is certainly
> *better* than using one word. But as others have pointed out, there are
> better password schemes than that.
Our computer centre suggest choosing a phrase an using every first
(or second, or last or whatever) character for the password. For
example: My grandma takes 5 glasses of whine per evening = Mgt5gowpe,
who could guess that password?
Jens
------------------------------
From: [EMAIL PROTECTED] (Jens Haug)
Crossposted-To: comp.security.misc,alt.security.pgp
Subject: Re: Passwords secure against dictionary attacks?
Date: 24 Feb 2000 07:22:38 GMT
Reply-To: [EMAIL PROTECTED]
In article <891bds$8pp$[EMAIL PROTECTED]>, [EMAIL PROTECTED] writes:
> QWERTY offsets are not very secure. A typcial dictionary
> attack interation would go: 1) Dictionary, 2) Reverse Dictionary, 3)
> QWERTY Offset Dictionary, 4) Alpha offset Dictionary,
>
> If bullwinkle is in my dictionary, interation number 3 would get you.
>
> I used to use QWERTY offsets. Not any more.
>
> As to the original posting on concatenating dictionary words. That too
> can be weak. However, since the concatenation permutations far exceed
> the QWERTY offset, I would dare say that concatenation is more secure
> than QWERTY.
We try to crack our users' passwords every now an then. Once the
cracker program found out one consisting of two greek words which
make no sense together. :-0
Don't use *any* word in *any* language!
Jens
------------------------------
Reply-To: "Clockwork" <[EMAIL PROTECTED]>
From: "Clockwork" <[EMAIL PROTECTED]>
Subject: Re: Processor speeds.
Date: Thu, 24 Feb 2000 07:33:52 GMT
The previous post was an unfinished/draft email. The final message was not
posted for some reason. I apologize for the obvious misspelling. Weird.
Clock.
------------------------------
From: "Markku J. Saarelainen" <[EMAIL PROTECTED]>
Crossposted-To:
alt.politics.org.cia,alt.2600,soc.culture.usa,soc.culture.russian,soc.culture.nordic,soc.culture.soviet,soc.culture.israel
Subject: Some of the U.S. government programs in business enterprises are
Date: Thu, 24 Feb 2000 07:43:10 GMT
Some of the U.S. government programs in business enterprises are
actually quite idiotic. These programs are geared toward the
development, improvement and maintenance of the U.S. National Security
in military, commerce, finance and other areas (actually the Secret
Service is also playing some major part). Often these programs and
individuals who are running these programs are very incompetent. I could
run any of these people any time. They are especially amusing, when they
do not know that you are deceiving them and you know that you are within
a system or a program (that you are a mole ) Amusing, isn't it?.
In August, 1995 one of my U.S. ex-relatives who is in NY mentioned that
one Peace Corp volunteer went to the CIA after his service in Peace Corp
in Ghana. (He also said that he was participating in some special
activity and program that I would have been too scared of learning :)
what a joke !). He said that this new CIA person in Ghana would have
been a mole - and of course the person had already been identified by
Ghanian and other people. What a joke, obviously the person did not
understand the word "mole". The mole is really a person who is there and
never would be identified as an individual or an entity belonging to
another agency or company. The mole just is there. It is like all those
Kirjuris around the world. You are just there and every now and then
pass some intelligence. So did I capture this person too? Well, that was
already in 1995. Just a question !
With my smiling face, I can do whatever I ever like to do. I have the
diplomatic immunity !
Best regards,
Markku
------------------------------
From: [EMAIL PROTECTED] (wtshaw)
Subject: Re: Crypto enthusiasm
Date: Thu, 24 Feb 2000 01:05:17 -0600
In article <[EMAIL PROTECTED]>, Mok-Kong Shen
<[EMAIL PROTECTED]> wrote:
> wtshaw wrote:
> >
> > This morning, I awakened with thoughts of all that I might get done in a
> > crypto way today. The result on best will be that of the various areas,
> > I'll just get a little done, however. But, I pick the topic...that's
> > freedom. Here are the options:
> .
> If one instead tries to pursue a number of options parallelly, which
> one could very easily be tempted to do, the result is more often than
> not like chasing the same number of hares.
Herding cats, perhaps.
> Once I wanted to learn
> two foreign languages parallelly, the result was extremely
> disappointing. Your dozen or so of options all seem to be quite
> promising. Why don't you just randomly choose one and DO it? (Or do
> you need an ideal source of 'true' randomness to help you make the
> choice in a theoretically impeccable way?) Good luck to you!
>
Things are getting better and I am not falling farther behind each day as
I was for a while. Directed efforts should show themselves more and more
now.
--
Regarding healthcare, when GWB became govenor, Texas was 43 in
the nation, now we are 49th. And, I need not tell you about his
bloody support of the death penalty. Reformer?
------------------------------
From: [EMAIL PROTECTED] (Nemo psj)
Subject: Assistance is needed :)
Date: 24 Feb 2000 08:29:20 GMT
Hello all this is my first time posting here so i hope i dont sound silly.
Alright here we go, I need to find some source for RC4(good source not from
the planetsourcecode.com site) or any other yet un cracked encryption
algorythm. The reason is that i have made over the past 8 months a what i
belive to be a really solid and good encryption algy. The password in the algy
is sent through an cypher code i wrote up to even make the process stronger.
This pasword is sent through the cypher each time it encrypts a letter, so each
letter in theory is encrypted a lil differently. Now the only problem is that
my cypher has a tendency to repeat the passwords it creates every so often and
this causes some patterns to break out in the file. Now i thaught hmm what if
i used an proven secure encryption algy to encrypt the password insted of the
cypher? Well this would be great for one thing it would stop the pattern
problem and add a whole new set of complexities to the algy. If i could get
the source and implement it i could finish up this progam and finally get some
sleep ;). Thanx to anyone that can help.
-Pure (yes its late here and my spelling and typing is of the lowest quality)
------------------------------
Subject: Re: EOF in cipher???
From: lordcow77 <[EMAIL PROTECTED]>
Date: Tue, 22 Feb 2000 12:08:05 -0800
K&R is not the end all of the C language. The current ISO C
Standard mandates in 5.2.4.2.1 _Sizes of integer types
<limits.h>_ that a char must be at least 8-bits. In the event of
a disreprency between K&R and the ISO Standard, the ISO Standard
prevails if we are to be discussing the standard C language.
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
------------------------------
From: [EMAIL PROTECTED] (Jim)
Subject: Re: Swapfile Overwriter: R.I.P.
Date: Tue, 22 Feb 2000 20:30:09 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 22 Feb 2000 02:19:53 GMT, [EMAIL PROTECTED] (Steve K) wrote:
>On Mon, 21 Feb 2000 03:01:38 GMT, [EMAIL PROTECTED] (Dave
>Hazelwood) wrote:
>
>>Go get Scramdisk....it has one to wipe both freespace and
>>the swapfile.
>
>Got Scramdisk & like it a lot. The "wipe swapfile slack" dingus is
>probably OK, but the manual only says:
>
>"Any slack left when the swapfile contracts may contain data,
>overwriting the slack securely clears any data in it."
But it doesn't wipe the swapfile itself, does it?
I have a fixed length swapfile, so I expect the slack isn't
a problem.
>But I still like having a little dedicated app that eats the entire
>swapfile, with multiple overwrites. This is not possible during a
>Windows session, hence the "slack" that is overwritten by Scramdisk,
>not the whole "swap file". Anything in there that is "in use" by an
>application, won't be touched. That's probably nothing to panic
>about, but some crypto nuts just hate loose ends... especially when
>there are easy ways to avoid leaving them.
Swapfile Overwriter is fine, or use Norton's WipeFile in DOS to
wipe the swapfile as many times as you need. The latter writes
ones then zeros for as many times as you specify, then writes
Hex F6 once and verifies the write.
--
Jim,
nordland at lineone.net
amadeus at netcomuk.co.uk
------------------------------
From: [EMAIL PROTECTED] (Jim)
Subject: Goodbye Swapfile Overwrite?
Date: Tue, 22 Feb 2000 20:32:16 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 22 Feb 2000 02:19:53 GMT, [EMAIL PROTECTED] (Steve K) wrote:
>On Mon, 21 Feb 2000 03:01:38 GMT, [EMAIL PROTECTED] (Dave
>Hazelwood) wrote:
>
>>Go get Scramdisk....it has one to wipe both freespace and
>>the swapfile.
>
>Got Scramdisk & like it a lot. The "wipe swapfile slack" dingus is
>probably OK, but the manual only says:
>
>"Any slack left when the swapfile contracts may contain data,
>overwriting the slack securely clears any data in it."
But it doesn't wipe the swapfile itself, does it?
I have a fixed length swapfile, so I expect the slack isn't
a problem.
>But I still like having a little dedicated app that eats the entire
>swapfile, with multiple overwrites. This is not possible during a
>Windows session, hence the "slack" that is overwritten by Scramdisk,
>not the whole "swap file". Anything in there that is "in use" by an
>application, won't be touched. That's probably nothing to panic
>about, but some crypto nuts just hate loose ends... especially when
>there are easy ways to avoid leaving them.
Swapfile Overwriter is fine, or use Norton's WipeFile in DOS to
wipe the swapfile as many times as you need. The latter writes
ones then zeros for as many times as you specify, then writes
Hex F6 once and verifies the write.
--
Jim,
nordland at lineone.net
amadeus at netcomuk.co.uk
------------------------------
From: [EMAIL PROTECTED] (Jim)
Subject: Re: How Useful is Encryption as Long as NSA Exists?
Date: Tue, 22 Feb 2000 20:29:59 GMT
Reply-To: [EMAIL PROTECTED]
On Mon, 21 Feb 2000 21:05:03 GMT, [EMAIL PROTECTED] wrote:
>Maybe a drug dealer living in Costa Rica, who has fled the U.S. years
>before, is using encryption software from a non-U.S. country in some of
>his daily operations, thinking that nobody is listening, when the U.S.
>actually has the key. What protection is a safe with an infinite number
>of combinations if your enemy has the secret code? Code breaking ceases
>to become an issue.
Because of that (almost) infinite number of combinations, he still has
to search for the right one you've used that day for that message.
That could, conceivably, take an (almost) infinte amount of time.
>Furthermore, even if the drug dealer in question is lucky enough to be
>using crypto from a country that has not been pursuaded by the U.S. to
>cough up the key, he is probably still using Windows, and might still
>have some of his data vulnerable.
That's all so very complicated! There are much more cost-effective
ways of getting hold of your keys, from plain theft through torture,
to hacking. If they want them, they'll get them - without having
to break any ciphers.
>Even though all this is a big "what if" scenario, can someone (say,
>living outside of the U.S.) using a Windows operating system be
>positively sure that the U.S. cannot decrypt his encrypted
>communication or the encrypted information inside his computer, except
>by guessing the password (which is the most difficult way)? how about
>access his/her files through Microsoft?
Well how? If I were that paranoid I'd purchase and use a security firewall
to block any attempt to compromise my computers. Suppose I use an off-line
computer to store all my sensitive stuff and to do my crypto on? How then
would they get to my keys apart from theft etc.?
A back-door in PGP? I don't think so...anyway we've done that thread to
death too many times already...!
--
Jim,
nordland at lineone.net
amadeus at netcomuk.co.uk
------------------------------
Subject: Crypto Speeds...
From: John <[EMAIL PROTECTED]>
Date: Tue, 22 Feb 2000 13:07:19 -0800
What would be an average or "acceptable" speed for encryption on
a pentium 3 processor at 500mnz?
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
------------------------------
Subject: Re: OAP-L3 Encryption Software - Complete Help Files at web site
From: lordcow77 <[EMAIL PROTECTED]>
Crossposted-To: talk.politics.crypto,alt.privacy
Date: Wed, 23 Feb 2000 16:07:56 -0800
Wagner does not use the word "unbreakable" once in his entire
posting. I don't know where you're getting this stuff from. He
asserted the existence of certain algorithms which were provably
secure *under a specific security model*.
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
------------------------------
Date: Thu, 24 Feb 2000 11:05:04 +0100
From: Runu Knips <[EMAIL PROTECTED]>
Subject: Re: Crypto Speeds...
John wrote:
> What would be an average or "acceptable" speed for encryption on
> a pentium 3 processor at 500mnz?
Well a general question with useless details (why such a specific
processor?) results in a general answer. For any operation, if the
user waits...
<= 0.1s - excellent, no need for optimizations anymore
0.2..0.3s - good, optimizations not necessary except if this is the
response time for really minor operations (for example,
inserting a character into a text file, or displaying
a menu).
0.5..1s - okay if the operation is nontrivial (loading a large
document etc).
2..5s - okay for longer jobs
>5s - bad, display something like a progress bar and inform
the user (a) whats going on (b) how long he has to
wait (c) the computer isn't crashed, but is doing
something
In other words, what do you want to know ? There is symmetric
encryption, and there is asymmetric encryption. For each class,
there are many different algorithms, where the faster ones are
often less secure and the slower ones are sometimes more
secure. If you use GnuPG or OpenSSL you will have very good
performance, near the best you can get, except if you really
compare them with full assembly implementations.
------------------------------
From: Tom Holroyd <[EMAIL PROTECTED]>
Crossposted-To: comp.security.misc,alt.security.pgp
Subject: Re: Passwords secure against dictionary attacks?
Date: Thu, 24 Feb 2000 19:02:38 +0900
On Tue, 22 Feb 2000, Ilya wrote:
> crypto/life cyber@machine green-dog Loud!Music
Say "passphrase" instead of "password" and be free. Assuming your crypt
implementation doesn't truncate at 8 characters (many still do -- trash
yours if it does). Then set your passphrase to something like
"in the summer avec vous"
or
"frequently backwards hotdogs"
or anything you care to type. Go ahead and use spaces, quotes, etc. You
want your passphrase to have high entropy (hard to guess) but still be
easy to remember (which high entropy 8 character passwords are not).
After setting your passphrase, check that just typing in the first 8
characters doesn't work. If it does, complain to the authorities that
they need to upgrade. There's no excuse for limiting users to 8 character
passwords.
And remember, if there's a packet sniffer watching you, you lose whenever
you type your passphrase in the clear, no matter what it is. So use an
encrypted channel or SRP. Also remember, many exploits bypass the
authentication method completely, so even if you have a 'secure'
passphrase, your system may be wide open to, e.g., buffer overflows.
Dr. Tom Holroyd
"I am, as I said, inspired by the biological phenomena in which
chemical forces are used in repetitious fashion to produce all
kinds of weird effects (one of which is the author)."
-- Richard Feynman, _There's Plenty of Room at the Bottom_
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.security.misc,alt.security.pgp
Subject: Re: Passwords secure against dictionary attacks?
Date: Thu, 24 Feb 2000 10:23:30 GMT
My 2p worth...
I have a (very) poor memory for cryptic passwords, so I use a few
techniques to construct memorable passwords for myself that are
relatively safe from dictionary attacks. A common one is to use a
memorable sentence such as:-
I need a hard-to-crack password
Taking first letters of the above could give you:-
Inah-2-cp
Another (more difficult to describe) method I use is to look at a
labelled object that you can always return to (if you forget the
password). An example on my desk is for instance my telephone, which
also has labeled function keys or perhaps a calender, or a poster or a
phone list or...... use your imagination. Extract words from the source
using a diagonal or vertical line... For example, down the left hand
side keys of my phone I have
Goodbye3Skip6Delete9Call#Stop>>>
That could translate into a password of
G3S6D9C#S>
or
>S#C9D6S3G
or
e3p6e9l#p>
or......
These passwords are very far from being dictionary words, but allow me
to easily remember them by just glancing at my phone - of course I do
have to remember which set I used, but for some reason I find that
easier than remembering purely random characters. Of course if
everybody starts using their phone in this way I guess these
common combinations may eventually filter into the cracker
dictionaries...
Regards
Ross
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Dr.Gunter Abend" <[EMAIL PROTECTED]>
Subject: Re: e-payment suggestion
Date: Thu, 24 Feb 2000 11:56:06 +0100
Mike Rosing wrote:
> Check out http://www.zero-knowledge.com/
> You'll find out more about secure money transfers than you
> want to know.
Thank you for this link -- I already looked at this site earlier.
The complicated, challenging problem of *anonymous* e-cash is,
of course, very interesting -- but, I merely had the question,
why personalized payments still are that risky. I dislike to
transmit my credit card number to any possible payee. Most
payments connected with the internet are not safe enough.
If I'd buy some second-hand equipment, some software, anything
in the order of 50$ or less, I need not trust the partner that
much as to give him my credit card number!
I didn't intend to invent another anonymous e-cash system,
however, my proposed "cheque" not necessarily reveales my true
identity. Only my bank knows my name and account number. The
police might access these data -- it is not truely anonymous.
Ciao, Gunter
------------------------------
** 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 sci.crypt) via:
Internet: [EMAIL PROTECTED]
End of Cryptography-Digest Digest
******************************
