Re: Meganet on Cryptogram again

From: Tom St Denis (tomstdenis_at_iahu.ca)
Date: 09/18/03 

Date: Thu, 18 Sep 2003 10:35:56 GMT

"Mxsmanic" <mxsmanic@hotmail.com> wrote in message
news:2ksimv820kj6vp5gdco1or9cb3s66qucfa@4ax.com...
> Tom St Denis writes:
>
> > Pick up your keyboard. Now let it go. Experience tells
> > me it will fall.
>
> You may infer that it will fall, based on past experience, but you don't
> really know until the keyboard is dropped.
>
> Of course, if I were using a light-based keyboard, your predication
> would fail.

Are you really this anal in person? I've already said there are exceptions
to the rules. e.g. "we're not talking in absolutes".

Maybe you should actually read what people write.

> > Did I have to personally inspect the keyboard to know that?
>
> To know what? You're making a prediction, not displaying any knowledge
> about my keyboard specifically.

So you're saying you never learn from experience? Well you go run with
scissors now.

> > Experience dictates that people who develop ciphers in
> > the manner of those people are almost always designed
> > poorly [e.g. rehashed password-XOR ciphers] or overly
> > complicated algorithms which can be reduced to more
> > trivial problems.
>
> So how do you spot the exceptions to "almost always" without actually
> examining the algorithms?

I personally don't care to. If the person doesn't care to propose their
design properly I don't care to read about it. The odds of their design
actually being unique and original [and useful] are so small that playing
Jedi Academy is a better use of my time [cuz my force powers are growing
strong]

> > The problem is to design a cipher that is secure you have
> > to follow academics.
>
> Why?

Because academics have attacks on ciphers you will want to know about.

> > Chances are if you have invented new theory you're not
> > going to bottle it up. You'll share the theory with others.
>
> That must be why the NSA publishes all the details of its research.
> That explains why all of Alan Turing's work on information theory is so
> freely available on the Web.

First the NSA is a bunch of weak-ass sissies. They attend public
conferences so they can leach original work then re-hash it interally to
cash a government pay-cheque.

> > True, but how does this reflect on whether the algorithm
> > is secure or not?
>
> It doesn't, any more than the speculations of people who consider
> themselves authorities in cryptology.

I'd say Wagner is qualified to make judgement calls in cryptography. Moreso
than you mr. "MixmasterBeatYo!"

> > My point is if it looks like snake oil, smells like snake
> > oil, it probably is snake oil.
>
> And if a message looks like random characters, and smells like random
> characters, it probably is random characters.

Um, wow, you're so earning the first slot in my killfile.

> > Um I'd say the crowd here denouncing the process???
>
> Denouncing them isn't going to stop them. Unless you have proof that
> their algorithm is insecure, it's just your word against theirs.

I don't get what your point is. The onus isn't on the public to prove that
a design is secure so until they do otherwise it's safe to say the design is
weak. Why is the design "weak" because chances are the design was made by a
person who hasn't read academia, doesn't know about cryptanalysis and
designed a cipher to solve a non-problem [re: privacy is easy,
authentication is harder]

Even if Meganet's crap is secure I wouldn't use it because I don't like them
[that and their design is inefficient].

Tom

Relevant Pages

  • Re: Great SWT Program
    ... The environment having more input options is necessarily no worse. ... Since one of the constraints is the skill of the designer, ... the best possible design using keyboard alone. ... I can also see dislike of the GUI "clutter" I call "providing context ...
    (comp.lang.java.programmer)
  • Re: Help! The keyboard I want is too good to be true!
    ... designing true pro keyboards that are up to todays possibilities. ... design - but sadly failed in the market. ... Is there anyone who I can hire to build me my customized keyboard ...
    (rec.audio.pro)
  • Re: iPhone
    ... justify the design, manufacturing and distribution overheads of more ... the bluetooth keyboard I bought with it. ... so that each customer would design his own ... very badly indeed with technology as a fashion item. ...
    (rec.arts.sf.fandom)
  • Re: Outside of calculator after a year of use
    ... But not at all is much slower. ... HP has convinced me with the HP 49G screen problems no IR port (they try to ... finally this HP49G+ monument to bad design it will be the last HP that I ... has been called and use it on improving the keyboard ...
    (comp.sys.hp48)
  • Re: VMPC isnt free
    ... Yah, tom the genius. ... for the sole fact they are not interested in academia. ... Admitedly I designed quite a few ciphers but I never claimed they were ... design over and over. ...
    (sci.crypt)