Hi Sean,

Yes, your "kind" of correct.. My explanation was a little simplified. But
for all practical purposes with a good hash you can just about say the
number is unique.

The hash lengths are commonly 128 bit (MD5) or 160 bit (SHA). With that
number of bits I'm pretty sure you can assign a unique number to EACH and
EVERY particle in the universe and still have some left over! I remember
reading this about UUIDs (Universal Unique IDentifiers, used in software
like COM and other object broker systems) which are 128 bit if I remember
correctly.

The "strength" of a Hash can be measured by the number of different messages
that hash to the same key and also how many of the bits change for each bit
of the message. How they determine this I don't know - I just use the
algorithms :)

It is possible to several messages to map to the same hash, but the hash
function is specifically designed so that the only possible way to workout
the "collisions" is by brute force. It is also far more improbable that two
"useful" messages will map to the same hash.

Cheers,
Ash.

---
Ashley Roll
Digital Nemesis Pty Ltd
www.digitalnemesis.com
Mobile: +61 (0)417 705 718




> -----Original Message-----
> From: pic microcontroller discussion list
> [mailto:PICLIST@MITVMA.MIT.EDU]On Behalf Of Sean H. Breheny
> Sent: Thursday, 9 May 2002 2:00 PM
> To: PICLIST@MITVMA.MIT.EDU
> Subject: Re: [OT]: Software Protection
>
>
> At 01:32 PM 5/9/02 +1000, you wrote:
> >If you aren't sure what a Hash (also called a message
> digest) is, it is
> >basically a way to take the contents of a message and
> compute a unique
> >(large) number. The idea is that no two messages will give
> you the same
> >number and you can not recover the message from the number.
> They are often
> >used to verify that the contents of a message have not been
> altered. Almost
> >like a CRC.. but MUCH bigger and better. Look for "SHA"
> (sometimes called
> >"SHS") and "MD5". These are the most common. I used SHA. SHA
> generates a 160
> >bit number, MD5 a 128 bit number.
>
> Hi Ash,
>
> I don't think you really mean unique, do you? If the number
> is shorter than
> the message then there MUST be more than one message that
> gives you the
> same number.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.