Prev: Infinite One-Time Pad, is this product BS?
Next: reverse use of encryption and decryption
From: David Eather on 8 Jan 2010 11:29
WTShaw wrote:
> On Jan 5, 12:14 pm, unruh <un...(a)wormhole.physics.ubc.ca> wrote:
>
>> Secrecy of the cypher IS a form of security defense. But it should not
>> be relied on. IF you use your cypher only with one other person, it may
>> be a very good defense. If you use it with hundreds, it is a bad
>> defense, because someone will leak the details to your opponent-- by
>> accident or design.
>>
> Which takes you back to the key(s), then there are two good rules:
> 1) Protect the source of the key so that even if you get it by hook or
> crook any future keys are kept obscure.
> 2) Don't betray your key(s) with a from of encryption that utilizes
> the whole key. It is better to use an algorithm that never uses key(s)
> in a simpleton manner.
>
> Got it? Protect sources. Protect methods; let's call that the Valerie
> Protection Protocol

It's not a hard question is it?

Who is Valerie?

What did they do in this field? (I'd just like to know - see, I know of
Kerchhoff and his principle so I know why he is worth listening to and
what he said - I would like to know the same about Valerie)
From: WTShaw on 9 Jan 2010 17:57
On Jan 8, 10:29 am, David Eather <eat...(a)tpg.com.au> wrote:
> WTShaw wrote:
> > On Jan 5, 12:14 pm, unruh <un...(a)wormhole.physics.ubc.ca> wrote:
>
> >> Secrecy of the cypher IS a form of security defense. But it should not
> >> be relied on. IF you use your cypher only with one other person, it may
> >> be a very good defense. If you use it with hundreds, it is a bad
> >> defense, because someone will leak the details to your opponent-- by
> >> accident or design.
>
> > Which takes you back to the key(s), then there are two good rules:
> > 1) Protect the source of the key so that even if you get it by hook or
> > crook any future keys are kept obscure.
> > 2) Don't betray your key(s) with a from of encryption that utilizes
> > the whole key. It is better to use an algorithm that never uses key(s)
> > in a simpleton manner.
>
> > Got it? Protect sources. Protect methods; let's call that the Valerie
> > Protection Protocol
>
> It's not a hard question is it?
>
> Who is Valerie?
>
> What did they do in this field? (I'd just like to know - see, I know of
> Kerchhoff and his principle so I know why he is worth listening to and
> what he said - I would like to know the same about Valerie)

Pardon me but the question is funny given that a lady named Valerie
was outed as a source for partisan reasons. The lesson is simple,
don't do the obvious that somebody else wants, especially nor for
really bad reasons like giving away the store. Some principles are
rather hefty in the importance sphere. I bet some folks got it first
time; Greetings.
From: Theo Markettos on 15 Jan 2010 18:57
Mok-Kong Shen <mok-kong.shen(a)t-online.de> wrote:
> So it does mean: (1) if one has one's own algorithm, one should publish
> it (in order not to have the disadvantage of errors undetected), (2) if
> one takes an algorithm from others, one should only take one that is
> publicly-known. Taking both these together, it amount to the same as I
> wrote above in my humble view.

Look at it this way. A rough illustration:

Reverse engineering the cipher is an O(n) operation
Bruteforcing the cipher to determine the key is an O(x^n) operation

(Not necessarily the same 'n' in each case)

Once you've reverse engineered the cipher, that's valid for every instance
of that cipher. Once you've determined the key, that's valid for that
instance of the key only.

Pragmatically, publishing the cipher may not help you in detecting errors.
It only helps you if there are enough people listening who care enough to
cryptanalyse it and tell you the results. You might end up just helping
your adversary.

But if you intend to mass produce your cipher in any way, it will be reverse
engineered sooner or later. And you shouldn't just take a publically known
cipher, you should take one that has been well-analysed. Any fool can
publish a cipher, but it doesn't mean it's any good or any decent
cryptographer has looked at it. And if it hasn't been subjected to
plenty of scrutiny, it's sensible to assume that it's bad.

Theo
From: WTShaw on 21 Jan 2010 00:25
On Jan 15, 5:57 pm, Theo Markettos <theom+n...(a)chiark.greenend.org.uk>
wrote:
..
>
> But if you intend to mass produce your cipher in any way, it will be reverse
> engineered sooner or later.  And you shouldn't just take a publically known
> cipher, you should take one that has been well-analysed.  Any fool can
> publish a cipher, but it doesn't mean it's any good or any decent
> cryptographer has looked at it.  And if it hasn't been subjected to
> plenty of scrutiny, it's sensible to assume that it's bad.
>
> Theo

Good means well conceived, corrected until correct. It does not mean
that decent cryptographers are necessary, just good ones. Being
rather indecent is more fun, but we see lots of that here out of
contempt and/or boredom. The proof is in the pudding if it is not too
hasty. Frankly, describing a decent algorithm to an indecent model is
as effective as psychoanalysis in probing the mental content of the
subject if explained well to her and skillfully exposed in return.

Mass producing any cipher requires simple well placed proliferation
knowing that the ignorant masses can't tell the difference. Naughty
or nice, Sister Marie has explained before that goodness has nothing
to do with it...cryptic comments fore sure as you had to be there. To
be a good cryptographer means doing your best and not just guessing,
or guesting as the idiomatic case may be. Quoting and being quoted
are different stances be they not wide apart ones.
First  |  Prev  | 
Pages: 1 2 3 4
Prev: Infinite One-Time Pad, is this product BS?
Next: reverse use of encryption and decryption