Re: Hashing a procedure object reliably

On Sun, Apr 13, 2025 at 07:03:56PM +0200, Maxime Devos wrote:

[...]

> > Typical hashing of the non-cryptographic kind aren't designed to
> > virtually eliminate hash collisions [...]

Nit: given a "reasonable" hash, the collision probability should be
the same for crypto or non-crypto hash (for the same width, for "random"
input).

The special thing about cryptographic hashes is that it is supposed
to be hard to create a thing which hashes to a given value (preimage
attack) and to create two things which hash to the same value (it's
hard to "engineer" a collision). So it's about deliberate engineering,
knowing how the hash function works.

-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQRp53liolZD6iXhAoIFyCz1etHaRgUCZ/v4rAAKCRAFyCz1etHa
RquMAJ9GfCDsNfyQU12D1j9xjiuWbut6DQCeKHd5NkifGiSaZttTFPjBJRQ1JRo=
=WL4S
-----END PGP SIGNATURE-----