April 16th, 2002
|
Gnutella Aficionado | | Join Date: March 13th, 2002 Location: Aachen
Posts: 832
| |
Quote: Originally posted by Archie
Personally I believe we may have problems with the lack of updates for it though. Gnucleus has been updated with ultrapeer support, but I can't use it properly since none of the morpheus/gnucleus 1.6.0.0 clients will become child-nodes to me. Very annoying; I hope they actively contribute to gnucleus rather than forking.
| Here's a posting of the GDF on that matter:
From: "Greg Bildson" <gbildson@l...>
Date: Sun Mar 10, 2002 2:37 am
Subject: Gnucleus and the future of Morpheus
One thing that developers should keep in mind is that it appears that Gnucleus is an interim solution for Morpheus. From all the information that I have, it appears that Streamcast will be switching to their own proprietary code base in 3 weeks or likely (much) more. I expect that Streamcast really doesn't want to maintain their client as open-source for the ongoing future. Now, they may have changed their mind but I doubt it. They are likely content for the time being to stick with Gnucleus because it works very well for them given the effort involved in customizing it. I also assume that Gnucleus will continue to have a large user base going forward.
Streamcast claims to be implementing all the basic Gnutella features as well as metadata searches and some other more recent proposals. One slight concern that I have is that they claim to be using slightly different metadata that includes some of their old fields from FastTrack but any improvement in the metadata area is probably worth it. I believe that they will be working on UltraPeers as well. Actually getting all this code right will be tricky for them. It is a big question mark as to how this will all fit into the Gnutella network in the future.
I think it would be very wise for them to release early betas to the GDF so that we can validate their implementation. They can probably get a lot of free work out of some of us.
Thanks
-greg |