1995-07-19 - Re: RC4 crack

Header Data

From: Simon McAuliffe <Simon.McAuliffe@Comp.VUW.AC.NZ>
To: cypherpunks@toad.com
Message Hash: 9fe93a85ae1407c1f7272a9694a9524262bc8d3f8cb0090dfa27faca3454bc1f
Message ID: <199507190045.MAA28819@lido.comp.vuw.ac.nz>
Reply To: <199507182047.QAA03926@bb.hks.net>
UTC Datetime: 1995-07-19 00:45:53 UTC
Raw Date: Tue, 18 Jul 95 17:45:53 PDT

Raw message

From: Simon McAuliffe <Simon.McAuliffe@Comp.VUW.AC.NZ>
Date: Tue, 18 Jul 95 17:45:53 PDT
To: cypherpunks@toad.com
Subject: Re: RC4 crack
In-Reply-To: <199507182047.QAA03926@bb.hks.net>
Message-ID: <199507190045.MAA28819@lido.comp.vuw.ac.nz>
MIME-Version: 1.0
Content-Type: text/plain


shamrock@netcom.com (Lucky Green) wrote on Tuesday, 18 Jul 1995:

> >As to the problem of not having 24-7 connectivity, you could either
> >use the WWW page, or the server (it will do this anyway) will keep
> >re-trying to get a socket connect to the master until it suceeds, so
> >when you next go on-line ...  wham it gets through again as
> >connectivity is resumed and says whatever it has been trying to say.
> >We need it to retry in case of network out (or horror) big master
> >falling over, until it gets resumed.
> 
> On many machines that will mean that it will try to initiate a connection
> to the host. Please allow for a manual connect option.

While we're suggesting features, how about including something (on the
networked version) which performs a quick sanity check on any clients
so we know they're not bogus, ie send one or more known plaintext/ciphertext
pairs with corresponding keys to verify the correctness of the compilation.






Thread