[hashcash] Re: hashcash v1 questions

On Tue, Jun 01, 2004 at 06:01:14PM +0000, Justin wrote:
> On 2004-06-01T05:28:37-0400, Adam Back wrote:
> > I think we are already close to having a line-length issue, especially
> > if we include sub-puzzles.  Practically you want sub-80-chars.
> 
> Isn't an 80 character limit unnecessary?  [...]  I have a handful of
> messages with 90+ character Exchange-generated MID headers that
> aren't line-wrapped.

I just took a random sample.  They mostly looked like below 80, plus a
few a-little-but-not-too-much longer.

> Is the length of the random field fixed in v1?  In v0 it was fixed at 96
> bits, right?

The current v1 code does not fix it.  We have discussed on the list if
it would be simpler / beneficial perhaps to fix it to 96 bits.

v0 never did fix the random field.  The early implementations used a
64-bit field (which is a bit small once you get to larger collisions
because v0 combines counter and randomness field); later v0
implementations changed to 96-bit.  But all v0 implementations would
accept essentially any length (subject to some sanity check of overall
stamp size).

> > Well if the postage generation is done in the background probably
> > doesn't matter so much.  (Though there is some user expectation
> > potential problem -- eg dialup, I want to know when its gone so I can
> > hangup).
> 
> Well that's a problem, but minting could also take much longer than
> expected with sub-puzzles.  

Much less likely though.  8 or 16 sub-puzzles greatly reduces the
variance.

Adam

Other related posts: