Re: [waste-discuss] More development instead of more details
- On Sat, 29 Nov 2003, Closedshop wrote:
> HiOr Third, by getting the necessary information (id, key, address)
> as discussed, we could summarize a little bit?
> There was bootstrapping to waste as a friendlist, not as a network
> discussed. Only the user can add buddies, not the network itself should
> be able to do that. So it is a list, not a network.
> Users on this list need a bootstrapping.
> There were two machanism discussed, a g2-network with DHT for the search
> of waste-keys and their actual ip adress, or, second, the bootstrap from
> a w-cache.php, which every user needs to upload to his website. The url
> to the cache is added to the waste-key.
through some other means such as email.
Fourth, though related, is that this unformation could be in a
file with a certain mime type that could just be read by the
program, rather than any copying and pasting to generate and load
> This decision has to be made.Right, that's one way. Another way is to start with a new
> Then waste should be hybrid with gnutella2.com, to introduce magnet uri
> hashing for shared media.
> This means there are three networks/protocols/3-in-one-clients needed:
> - Waste-List
> - Gnutella2-network for normal publich filesearch
> - a new NEO-Boot Network based on a separate used g2-network with DHT to
> search for waste-keys and get the actual ip adress returned. This is
> ponly used for bootstrapping and to find the buddies Ip adress, if you
> know his Waste-key.
> So the working plan would be:
> Take waste, add 2 times a gnuc-dna to it, make webcaches for the
> separate g2-boot-network and add magnet uri hasing to waste-list and
client, not use any of ths waste code or protocol, but just the
basic idea, which is one of the best things about waste (the
other being that it actually exists).
> This client then could have an option to connect to other peers only byRight!
> peers, which are a decentral proxy for other peers. The proxy-location
> is transmitted from peer to peer, by peers identifying itself as a
> n-peer or a p-peer (normal and proxy peer). Proxy-Peers connect 90 %
> only to proxy-LEAF-peers as well. But if proxy-peers download for
> themself, the only connect to proxy-peers.
> Right ?