# Re: simple math question

*From*: tom@xxxxxxx*Date*: Thu, 16 Apr 2009 16:10:22 -0700 (PDT)

On Apr 16, 1:38 pm, pubkeybreaker <pubkeybrea...@xxxxxxx> wrote:

"Implementing AES" is not doing cryptography. It is writing code

from a well documented cookbook.

Um, optimizing it means you have to understand the math of the design

and perform design tradeoffs to exploit it's structure. We can't all

be rocket surgeons like you.

Multi-precision arithmetic is part of cryptography.

Hi, I'm Tom St Denis, author of LibTomMath and TomsFastMath, a text on

the subject, and I don't have an advanced degree in mathematics. Can

we put the ego stroking aside?

As for why trig is needed: Try analyzing anything to do with

elliptic curves without knowledge of periodic functions.

I've implemented performance ECC as well. Using algorithms for which

I haven't seen implementations of.

It shows a fundamental lack of background mathemtics. The ability to

work with trig functions; to understand them, to understand their

relation to general functions of a complex variable; to understand

groups on the unit circle, etc. etc. etc. is a requirement.

Depends on what you're doing. I can fully understand differential

cryptanalysis AND not know anything about trig. Are you saying

cryptanalysis is not part of cryptography?

We are not discussing the implementation of crypto methods. That

can be done cookbook fashion from well-documented sources. We are

discussing the ability to design and analyze ALGORITHMS. Someone

who can't (or hasn't) learned trig, will not be able to do the

former.

You're both very condescending, and no offense, but full of shit.

Implementing optimized crypto is not a matter of copy/paste from

"cookbooks." If it were you'd see more LibTomMaths floating around

[or TFM or LTC or ...]. It takes some brains to put it together, but

you don't have to be a math PhD to be capable. Not only that but when

I was referencing a lot of standard material like HAC, I had to

correct things [and people helped] that were wrong with them [or just

plain missing].

I'd say you need far more algebra than trig.

But that's just being pedantic. My point is you don't have to be you

to be a "cryptographer." I implement algorithms, I've developed

theory, and designed algorithms. And I don't hold an advanced degree,

and am nowhere near a math genius. I think we can all agree there is

a large range of different tasks in the crypto field involving

different levels of education and specialty.

You've really got to stop being so dismissive.

Tom

.

**Follow-Ups**:**Re: simple math question***From:*pubkeybreaker

**References**:**simple math question***From:*Antony Clements

**Re: simple math question***From:*tom

**Re: simple math question***From:*pubkeybreaker

**Re: simple math question***From:*tom

**Re: simple math question***From:*Thomas ***in

**Re: simple math question***From:*tom

**Re: simple math question***From:*pubkeybreaker

- Prev by Date:
**Re: Cryptography Today.** - Next by Date:
**Re: simple math question** - Previous by thread:
**Re: simple math question** - Next by thread:
**Re: simple math question** - Index(es):