you got me wrong. i used to agree with moak in the past, but i think he has gone too far. concerning greed, it is not a problem as long as the network is not hurt.

i can't see anyone hurting the network. i didn't believe the rumours that Phex/XoloX hurt the network with their automated queries, and i don't believe that LimeWire hurts the network with their clustering either.

however, i do agree with you were bearshare is concerned. i am convinced that clustering clients based on features is a good thing. i am however not convinced that clustering based on vendors is a good thing. but that is exactly what you are doing: by blocking specific vendors, you are in effect clustering specific vendors, and that is exactly what vinnie does. that is why Mr. Gone called you a Hypocrite.

further, i dislike how Bearshare completely stopped supporting 0.4 connections (hurting the network), because that is not "preferencing by feature", that is "blocking by feature". and i won't tolerate these spy packets (as they are a security issue). noone really knows what they are, and i don't believe vinnie that they contain only version numbers. it would be much easier to call home to bearshare.com.

but i don't advocate all the community to go and block bearshare. even if all bears were clustered together, that wouldn't necessarily mean that it hurts the network. if you disagree than please explain me why. think of the gnutella network as a large sphere and the bearshares as a small sphere inside the big one: that's what bearshare does. it could make it hard for non-bearshares to connect because Bearshare will not attempt their connection requests. that's the only issue i see with it.

Adam: is there a reason to completely stop support for 0.4 clients? how much of the network still uses 0.4, do you have any accurate statistics? i can't believe that this step was necessary, why didn't limewire use its good relationship to vinnie and tried to convince him to go on with 0.4 support?