Re: Is RC4 a good choice?

From: Mrsjunecarey (mrsjunecarey_at_aol.com)
Date: 08/05/03 

Date: 04 Aug 2003 22:50:52 GMT

>Subject: Re: Is RC4 a good choice?
>From: daw@mozart.cs.berkeley.edu (David Wagner)
>Date: 8/1/03 7:33 PM GMT Dayligh

>Right. That's not exactly "nobody has helped". And, frankly,
>you've gotten a lot of useful feedback, but from where I sit, it
>seems to have fallen on deaf ears.

Actually, the only person that really provided serious help was Bob Jenkins.

And I asked him for his help via email, not via USENET "sci.crypt".

>
>I'll give you some feedback, too, as long as you promise not to take it
>personally. (It is not intended to be taken personally; it is intended
>as constructive criticism of Leopard, which is different.)

Well, frankly, you are coming across as somewhat patronizing.

>
>Here's my feedback:
>
>You're going about this the wrong way. Cipher design requires deep
>analysis, and that's what you're missing.

Wrong.

I have analysed my current offering, L14, very deeply indeed.

>
>In general, the whole approach of trying lots of modifications to RC4
>strikes me as a bad idea; as you're discovering, RC4 variants are highly
>susceptible to serious security problems.

You make it sound like I am randomly modifying RC4 with no apparent direction
or motivation.

That is not the case at all.

Do you know what the word "patronize" means ?

Quantcast