Editing Gnutella (section)

== Gnutella2 ==
{{Main|Gnutella2}}
The Gnutella2 protocol (often referred to as G2), despite its name, is not a successor protocol of Gnutella nor related to the original Gnutella project,<ref>[http://www.mp3newswire.net/stories/2003/gnutella2.html Gnutella vs. Gnutella2, Part 2] {{Webarchive|url=https://web.archive.org/web/20120728110155/http://www.mp3newswire.net/stories/2003/Gnutella2.html |date=2012-07-28 }} Thomas Mennecke for [[Slyck.com]], February 25, 2003.</ref> but rather is a completely different protocol that [[fork (software development)|forked]] from the original project and piggybacked on the Gnutella name. A sore point with many Gnutella developers is that the ''Gnutella2'' name conveys an upgrade or superiority, which led to a flame war. Other criticism included the use of the Gnutella network to bootstrap G2 peers and poor documentation of the G2 protocol. Additionally, the more frequent search retries of the [[Shareaza]] client, one of the initial G2 clients, could unnecessarily burden the Gnutella network.<ref>[http://www.slyck.com/news.php?story=100 Gnutella vs. Gnutella2, Part 1] {{Webarchive|url=https://web.archive.org/web/20160305053853/http://www.slyck.com/news.php?story=100 |date=2016-03-05 }} Thomas Mennecke for [[Slyck.com]], February 25, 2003</ref>

Both protocols have undergone significant changes since the fork in 2002. G2 has advantages and disadvantages compared to Gnutella. An advantage often cited is that Gnutella2's hybrid search is more efficient than the original Gnutella's [[query flooding]].  However, Gnutella replaced query flooding with more efficient search methods, starting with Query Routing in 2002.  This was proposed in 2001 by Limewire developers.<ref>[http://wortschatz.uni-leipzig.de/~fwitschel/vorlP2P/literatur/search_methods.pdf A comparison of Peer-to-Peer search methods] {{webarchive|url=https://web.archive.org/web/20070926110918/http://wortschatz.uni-leipzig.de/~fwitschel/vorlP2P/literatur/search_methods.pdf |date=2007-09-26 }} Dimitrios Tsoumakos, 2003</ref><ref>[http://www.ic.unicamp.br/~celio/peer2peer/gnutella-related/query-routing-protocol.htm Query Routing for the Gnutella Network proposal] {{Webarchive|url=https://web.archive.org/web/20150924032530/http://www.ic.unicamp.br/~celio/peer2peer/gnutella-related/query-routing-protocol.htm |date=2015-09-24 }} Christopher Rohrs, 2001-12-18</ref><ref>[http://www.ic.unicamp.br/~celio/peer2peer/gnutella-related/query-routing-protocol.htm Query Routing for the Gnutella Network 1.0] {{Webarchive|url=https://web.archive.org/web/20150924032530/http://www.ic.unicamp.br/~celio/peer2peer/gnutella-related/query-routing-protocol.htm |date=2015-09-24 }} Christopher Rohrs, 2002-05-16</ref> An advantage of Gnutella is its large user base, which numbers in the millions.<ref name="population"/>  The G2 network is approximately an [[order of magnitude]] smaller.<ref>[http://crawler.trillinux.org/ G2 Network Statistics] {{Webarchive|url=https://web.archive.org/web/20100501141118/http://crawler.trillinux.org/ |date=2010-05-01 }} Population ~200,000 January 2010.</ref>  It is difficult to compare the protocols in their current form.  The choice of client, on either network, probably affects the end user just as much.