On Thu, Jan 04, 2001 at 01:49:47AM +0100, Oskar Sandberg wrote:
> On Thu, Jan 04, 2001 at 11:28:32AM +1100, Zem wrote:
> <>
> > It's not just Whiterose. A new developer needs to learn at least the
> > basics of the Freenet protocol before they can tackle Whiterose or
> > Fred. The major documents are incorrect or out of date. Anyone asking
> > questions about the current protocol is referred to the Fred source
> > code, or expected to trawl through the mailing list archives for
> > important decisions buried in flame wars.
>
> Actually, I have been trying to tell people with questions to ask right
> here, or via email or on irc if they want. Just be specific (there is
> most certainly good documentation of the general concept - Theo's paper
> for example).
>
> Ask questions, I'll answer, and then somebody can cut it together into a
> more comprehensive documentation. I know it isn't rational, but I find it
> easier to right 500 words in reply to an email than those same words off
> my head into a documentation.
>
> > The mainstream Freenet developers have the advantage of an early start.
> > Even Freenet 0.2 was relatively simple, as Liberator demonstrates. 0.3
> > is more complex - key exchange, crypto, multiple keytypes - yet has less
> > documentation.
>
> But it is fairly abstracted and layered. You don't need to understand the
> progressive hash to work on the key exchange, or vice versa. You don't
> have to understand the presentation protocol to work on the datastore.
But you do have to know all of this stuff to write a full client
library, which is the problem that I am running into with
Freenet-Python. There isn't sufficient documentation of the Freenet
0.3 protocol, and I'm not familiar with all the Fred internals and
such (which precludes reading the Fred source code).
--
Travis Bemann
Sendmail is still screwed up on my box.
My email address is really [EMAIL PROTECTED]
PGP signature
