On Wed, May 31, 2006 at 10:44:30AM -0700, Ian Clarke wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 31 May 2006, at 04:48, Matthew Toseland wrote: > >> > >>>My original objection stands to FCP connection management: 99% of > >>>the > >>>usage of FCP connection management will be for grotesque hacks which > >>>produce bogus opennets without the right topology. > >> > >>That demonstrates a rather condescending view of client authors. > > > >This demonstrates established fact. Have a look at the numerous key > >exchange boards on Frost, the 2+ ubernodes, #freenet-refs, the clients > >complaining about overload because they used a script to add 300 > >(disconnected) peers... > > Has it occurred to you that people are resorting to these because a) > there is no better alternative and b) we actually advise them to use > IRC *BECAUSE* we know there is currently no better alternative?
You think so? I haven't met one person who said "I wanted to add my friends to freenet but it was too hard". I have met many who said "I wanted to add my friends to freenet but they didn't want to", and a lot more who said "I don't have any friends". Have you met anyone in the first category? There is no alternative to IRC, because they can't persuade their friends to join them on Freenet, and that wouldn't help them anyway because their friends aren't _on_ Freenet. Not because it is difficult for them to add their friends. Having said that, an IRC plugin[1] might be an efficient way of connecting two people who already have nodes and who installed the plugin for their favourite IRC client. The hard part is invites. Short of email integration (which isn't possible because most people use webmail), I'm not clear how to do invites to people not already on the network. And this is a problem that needs to be solved. Adding FCP connection management will not - on its own - solve the problem. FCP connection management may be a good idea - it is reasonably easy - but without some idea of what it will be used for, I don't see that it is a priority. [1] A real IRC plugin. As opposed to the hack that some people have which lurks on #freenet-refs and adds every reference to their routing table. Resulting in their node being overloaded. > > Unless it is easy for people to connect to their friends the whole > Darknet concept will fail. Implementing support for third-party apps > to do this is absolutely *critical* to achieving this. The fact that > people are currently resorting to ugly kludges just underlines the > need for proper support for connection management in FCP, coupled > with guidelines as to how people can use it in a way that won't hurt > the network topology. This seems to be the core of your argument, but I don't understand it. Ugly kludges happen because people don't have friends. Not because it is hard for them to connect to their friends. Join #freenet-refs sometime and ask people. The current ugly kludges are caused by a) the fact that the network is small, and b) the fact that there is no opennet. But in the long run people will have friends already on Freenet, and FCP support for connection manipulation will help them to connect to these. So we need it - eventually. What we need now is a good invite mechanism - something that doesn't require the plugin to be on both ends. So far the best invite mechanism known is "get them to install freenet, and exchange references manually". This sucks. Whatever happens it must occur in real time because of NATs. If we can solve *this* problem, that would be a major contribution. Anyone have any ideas? > > Ian. -- Matthew J Toseland - toad at amphibian.dyndns.org Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: <https://emu.freenetproject.org/pipermail/tech/attachments/20060531/de864ebb/attachment.pgp>
