Prev: Chua's treatment of Wolfram's result
Next: What could WYLBNIMGTHAD possibly mean?
From: Thomas on 27 Jul 2010 00:35
> I'm sure most people can start rightaway from the brain to think of
> (understand, analyze, attack,...) it. But I think most people must
> read some documents (pseudo-code, formulars, schemata, implementation
> notes,...) in order to implement anything.
Probably, but I think it would be a thousand times easier to read/
visualize if you could just upload the clean PDF somewhere and hand
out the link, rather than having to decrypt an insanely long usenet
post with linefeeds and ascii math. That would take about five minutes
at most (converting everything to LaTeX and PDFing it). Could you
please do it, then eventually submit a pseudocode (later) ?
> Anyway you're right that pseudo-code may not be the best way to ease
> memory. Schemata are.
Depends on people. Personally I do better with pseudocode (I can write
code in my head so I'm used to it).
From: Mok-Kong Shen on 27 Jul 2010 04:02
Tran Ngoc Duong wrote:

> I'm sure most people can start rightaway from the brain to think of
> (understand, analyze, attack,...) it. But I think most people must
> read some documents (pseudo-code, formulars, schemata, implementation
> notes,...) in order to implement anything.
>
> Anyway you're right that pseudo-code may not be the best way to ease
> memory. Schemata are.

I agree with Thomas that I personally prefer pseudo-code. I meant
however that, if the algorithm is of such a nature that the pseudo-code
as such is comparatively easy to be kept in memory, especially when no
or at most a couple of particular numbers need to be kept in mind, then
that would be an ideal situation of a simple algorithm.

M. K. Shen


From: Tom St Denis on 27 Jul 2010 06:45
On Jul 26, 4:35 pm, Tran Ngoc Duong <tranngocdu...(a)gmail.com> wrote:
> On Jul 26, 7:54 pm, Tom St Denis <t...(a)iahu.ca> wrote:
>
> > On Jul 25, 2:56 pm, Tran Ngoc Duong <tranngocdu...(a)gmail.com> wrote:
>
> > > THE BLOCK CIPHER NSABC
> > > (Public domain)
>
> > I haven't read your spec because it was line-wrapped as you posted it
> > to usenet [hint: get a website to post this, or turn it into a PDF and
> > post it ...]
>
> Thank you for the hint. I'll add schemata, vectors, reformat and post
> it again to Usenet.
>
> > But
>
> > - Lack of distinct pseudo code
> > - Lack of vectors
>
> > Makes me very dis-interested in pursuing it further.
>
> I'm sorry for dumb question but as English is not my nature language,
> what is "distinct pseudo code"?

Write out the pseudo-code like an algorithm as opposed to dropping
pseudo-code throughout the document. It wasn't apparent from skimming
your document that you had any description of the algorithm in
mechanical terms.

Tom
From: Tran Ngoc Duong on 27 Jul 2010 10:08
On Jul 27, 5:45 pm, Tom St Denis <t...(a)iahu.ca> wrote:
> Write out the pseudo-code like an algorithm as opposed to dropping
> pseudo-code throughout the document.  It wasn't apparent from skimming
> your document that you had any description of the algorithm in
> mechanical terms.
>
I'm sure the pseudo-code is written the best way (structured and
formal, but still readable) we can do. It wasn't apparent, that's
right. But perhaps because it is so simple that the description, in
English, is not long enough to draw your attention.

And the formatting is not the best :( I'll re-post it soon. Many
thanks for consideration.
From: Tran Ngoc Duong on 27 Jul 2010 10:22
On Jul 27, 11:35 am, Thomas <thomas.benet...(a)yahoo.fr> wrote:
> > I'm sure most people can start rightaway from the brain to think of
> > (understand, analyze, attack,...) it. But I think most people must
> > read some documents (pseudo-code, formulars, schemata, implementation
> > notes,...) in order to implement anything.
>
> Probably, but I think it would be a thousand times easier to read/
> visualize if you could just upload the clean PDF somewhere and hand
> out the link, rather than having to decrypt an insanely long usenet
> post with linefeeds and ascii math. That would take about five minutes
> at most (converting everything to LaTeX and PDFing it). Could you
> please do it, then eventually submit a pseudocode (later) ?

A TeX version will come later. When we write in TeX, we must surely
take the opportunity to re-typeset it with standard math symbols
(rather than C language ones). For now I can only offer a pure-ASCII
version, for which we've decided because we thought there aren't many
math stuffs in it. Thanks.
First  |  Prev  |  Next  |  Last
Pages: 1 2 3
Prev: Chua's treatment of Wolfram's result
Next: What could WYLBNIMGTHAD possibly mean?