On Mon, Aug 04, 2003 at 09:11:39PM -0700, Tracy R Reed wrote: > On Mon, Aug 04, 2003 at 08:16:39PM -0500, Tom Kaitchuck spake thusly: > > Ok, why does everyone act as though NGrouting is going to be the Holy Grail. > > Because we are all frustrated that performance has only gotten worse over > the last year and are really really really hoping that things will get > better. We have nothing else to pin our hopes on but NGR.
And bugfixes! See the recent changes relating to NIO and message sending times. But NGRouting *is* a good idea, it should solve the load balancing problem, something we have been very weak on so far, improve performance, adapt to network topology and maybe even allow routing to take place in a guerilla-cell-style network. > > > So for the time being, will everyone just stop rejecting reasonable ideas > > simply because "NGrouting is comming". (Although this idea is clearly much > > more contrivertial than most.) > > What reasonable ideas have been rejected? Most of the ideas proposed > recently that I object to would not be needed if freenet worked properly > and are only a distraction from getting it working properly. NGR is a big > step in the direction of getting it working properly. It has been shown > that freenet performance is all about routing. Without effective routing > we are just another non-scalable broadcast search network that doesn't > find data. > > If NGR doesn't work I think something serious is going to have to be > proposed to enforce routing and strong specialization. > > -- > Tracy Reed > http://ultraviolet.org -- Matthew J Toseland - [EMAIL PROTECTED] Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so.
pgp00000.pgp
Description: PGP signature
