fish <fish at bovine.artificial-stupidity.net> writes:

> On 3 Nov 2002, Edgar Friendly wrote:
> 
> So I can't type.  sue me :)

I'm suprised that _everyone_ else on this thread has pointed out your
mistake (often derisively), and not even tried to answer the obvious
question.  Please try to be constructive, people.
> 
> > As for the opposite, private -> public, I'm of the opinion that
> > clients shouldn't expect any relation between the key a SVK is
> > inserted with and the key it needs to be requested with.  The node
> > returns the request key when the insertion completes, and I think that
> > should be used.  I believe that clients shouldn't dissect the insert
> > key and try to put together a request key before seeing what key the
> > node says to request.
> 
> See, that's a nice theory, but you need to know what your public key is in
> order to do any sort of linking.
> 
Many kinds of linking are possible, but you're right, you do end up
with the CHK <-> CHK linking problem; your pages need to satisfy a
dagraph property in terms of which link to others.  It's probably
reasonable to remember the SSK prefix, but when coding parts of liber,
Travis committed what I find to be quite sinful code, in doing stuff
like appending "PAgM" to the public key to make the SSK base.

> I have 5 different "bug reports" now which convince me that people are
> not, in fact, capable of dealing with this.
> 
>       - fish
> 
my design choice for liber is to have a long-lived config file that
holds on to an identity; the SSK public/private keypair (if I had my
way, it'd only hold the private key, but...)  That way people's
inserts are consistently identified by them with both a signature
(better than a KSK because it's not subvertible) and a plaintext name
(which is better than most CHKs, even though CHK at blah/filename.ext is
allowed by the specs).  

This allows for some sort of trust to keys; if it's a SSK I control,
and others have requested other keys of mine, they can reasonably
surmise that this new content is similar quality.

Hoping to spread the use of SSKs,
Thelema
-- 
E-mail: thelema314 at bigfoot.com                        Raabu and Piisu
GPG 1024D/36352AAB fpr:756D F615 B4F3 BFFC 02C7  84B7 D8D7 6ECE 3635 2AAB

_______________________________________________
devl mailing list
devl at freenetproject.org
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to