[pacman-dev] MD5/SHA* why?

[Xavier]

On Thu, Jul 05, 2007 at 03:42:42PM -0700, Jason Chu wrote:
> 
> We are at an inroads in hashing algorithm theory.  All the current hashing
> algorithms have flaws.  It's also likely that any new hash algorithms will
> have flaws as well.
> 

Maybe the information I had is already outdated, since all this stuff
moves pretty quickly :)
What are the flaws of all the SHA-224/256/384/512 hashes ?
see this for example :
http://en.wikipedia.org/wiki/SHA-1#SHA_sizes
Or are these the new algorithms ? They could indeed have flaws as well,
but still say more secure than the current ones, even after flaws
are found.

> If we just trusted md5s or sha1s, then it would be less secure and more
> complicated, but because we look at both md5s and sha1s *together* that
> things improve.
> 

I'm not convinced that
1) md5 or sha1 alone aren't enough secure (for our use case)
2) combining md5 and sha1 is better than eg SHA-256

> An analogy, think of two sheets with holes in them.  You can look through
> each sheet and see the light on the other side, but if you lay the two
> sheets on top of each other a lot less light is visible.  Because we're
> considering both hashing algorithms they cover some of the other's
> failings.
> 

In that case, you move both holes so that they match (with padding) :)
But yes, that's still the general case, not pacman one.

> I'm all for making less complication though... maybe a more abstract hash
> API?
> 

If we need to keep several hashing algorithm, I think this would be great.