Re: XORShift PRNG as a diffusion structure

From: Tom St Denis (tomstdenis_at_yahoo.com)
Date: 03/24/04 

Date: 24 Mar 2004 09:40:21 -0800

Mok-Kong Shen <mok-kong.shen@t-online.de> wrote in message news:<c3rkec$p79$06$1@news.t-online.com>...
> Mok-Kong Shen wrote:
>
> > Paul Pires wrote:
> >
>
> >>
> >> Bull***.
> >
> >
> > Ahh, you claimed to have put me on your kill-file and
> > you could still read my post??? A liar, in addition to
> > being a pseudo-grammarian and nonsense-poster!!!
>
> I confounded you with Joe Peschel above. But you are
> posting nonsense to the group, wasting the bandwidth and
> polluting the atmosphere of the group all the same
> like him and Tom St Denis.

You want to talk about waste? Seriously? Let's try just in this
thread alone...

In http://groups.google.com/groups?q=g:thl1804188767d&dq=&hl=en&lr=&ie=UTF-8&selm=c3n3r6%24ln3%2401%241%40news.t-online.com

You basically asked for a comparison between AES and the new idea
[which is hard to come up with on it's own].

In http://groups.google.com/groups?q=g:thl1804188767d&dq=&hl=en&lr=&ie=UTF-8&selm=c8016437.0403221604.3c3a14da%40posting.google.com

I digressed a bit and showed the CSQUARE design against a bitsliced
cipher [I was trying to reasonably interpret your question]. I gave
concrete numbers showing that the transform Choi was takling about
results in a lower trail weight and requires more rounds for equal
security [er...privacy].

Admitedly I should have used AES for comparison but I like the CSQUARE
results and they're not exactly "new" [the paper has been up for over
a month].

Then you asked: http://groups.google.com/groups?q=g:thl1804188767d&dq=&hl=en&lr=&ie=UTF-8&selm=c3o1d5%24of9%2403%241%40news.t-online.com

Which basically asked the question I had just finished answering.
This isn't the first time you've done this before [or the first time
I've pointed this out for that matter]. You being your typical lazy
jack-ass mean little minded self decided to take a normal civil
discussion and twist it into a useless endless debate in which you'd
play the "guy who doesn't trust others" and I'd play the "guy trying
to answer the question".

Then you asked: http://groups.google.com/groups?q=g:thl1497522181d&dq=&hl=en&lr=&ie=UTF-8&selm=c3p2sj%24mt9%2404%241%40news.t-online.com

Where you asked "how much diffusion do you get from it?" Which is not
only a trivial question [anyone into crypto should be able to answer]
but totally moot. Diffusion != sufficient. You need branch not
diffusion. Not only have I been raving about wide-trail design
philosophy for the longest time it has existed since before I even got
to this scene [re: Daemens Ph.D].

So really this is just an example of one of your "moot questions" that
you often deploy to either change the topic or ask "insightful"
questions to prolong the thread. Because lord knows, if we aren't
talking in a MKS thread then your life is incomplete!

So you wonder why I get all "vulgar" it is because you basically make
it a pain for people to hold discussions here. You exploit the
genuine kindness of people into endless two party discussions that
never really go anywhere because you deny what people say, you ignore
most of what they provide, you misinterpret the little stuff you do
read and you rarely seek the references people point out.

You don't talk rationally in this group whether on purpose or not I
don't really care. If you're just some punk 12 yr old kid then why
not just wait until you can actually understand an "old person"
conversation a bit better. If English isn't your first language then
take some college prep classes. If you're mentally ill then just
leave the f'ing group already and if you're just being an ass [most
likely] then why not give it quits for a while? You've had a long run
as village idiot and you should make room for other more enlightened
people to take the spot.

Whatever you do, don't play the "victim" here. People do genuinely
answer your questions. They only get angry when you take advantage of
that and make it more hassle then it's worth.

Tom