[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ethmac] sorry



18-May-00

   Hi,

          In resepct to crypt the information, first the application can do
it and while in software it is slower it is compatibale. if you do it
hardware there are few issue you need to think of, and they are what field
can you encrypt as if you encrypt for example the IP header you will make a
total mess.
Also if you use some sort of proprietry crypt you will need to have this
piece of hardware anywhere which make it somewhat limitation, on the other
hand you might want simple to use a well known one like DES ot Tripel-DES
which are I belive the most popular or maybe RSA.

Another issue is what to crypt, are you going to crypt everything ? this
mean that if you want to pass a certain information to someone that don't
have this piece of hardware you will not be able, so maybe only certain
type of application and this mean more complex solution as you know need to
check the application in the packet and so on.

have a nice day

   
At 09:46 PM 05/19/2000 +0700, you wrote:
>Hi all,
>
>I'm really sorry I can't focus on the design since last week until end
>of may, because I have final exams.
>
>BTW, I'm now finding informations about development of network interface
>cards, about the speed or other features needed for next conditions of
>computer networks/communications. I need these informations for
>introduction/backgrounds section of the final project report.
>
>There is one suggestion interesting to me that may be we need a feature
>for data security in NIC (e.g. a kind of cryptography), because it says
>that until now data in computer network involving NIC are easily
>"sniffed". And it is important especially for VIP and private
>communications. 
>what do you think about it? or may be you have any other interesting
>information.
>
>regards,
>
>Mahmud 
>
>