The new KSKs are functionally equivalent to KHKs. The difference is that the
data is signed with the keyword, so unless you know the original keyword, you
can't make a false response.
To request something indexed under KSK you still only need a keyword
("teletubby" ,"decss", "oskar's decssing teletubbies"), only the process of
creating the key from the keyword is a little more contrived.
We also said we would support hashing in a single keyword in with the Public key
for the SVKs, making it work for the update by enumeration. I think Scott wrote
the code for that but he hasn't committed it.
On Sun, 06 Aug 2000, Ian Clarke wrote:
>
> While we wait for AGL to document Oskar and Scott's recent rearrangement of
> the
> key-handling (!), which is not particularly easy to reverse-engineer through
> looking at the code - I think we need to think a bit harder about 0.3
>
> KHKs have now been removed, this is desirable, but the idea was that they
> would
> be replaced by fuzzy searching which won't make it in to the code until at
> least
> release 0.4.... this leaves a gap when there won't be any "guessable" keys
> (unless I have misunderstood something from examining the new Freenet.keys
> package).
>
> I guess this means that in-order that people will be able to play with Freenet
> usefully we will have to redo the key-indexes so that they take act like
> Yahoo,
> but for Freenet CHK, SVK or KSKs.
>
> Comments?
>
> Ian.
>
----------------------------------------
Content-Type: application/pgp-signature; name="unnamed"
Content-Transfer-Encoding: 7bit
Content-Description:
----------------------------------------
--
\oskar
_______________________________________________
Freenet-dev mailing list
Freenet-dev at lists.sourceforge.net
http://lists.sourceforge.net/mailman/listinfo/freenet-dev
