Talk:Peer-to-peer/Archive 1

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 1 Archive 2

Merger proposal

Please see talk:File sharing for a proposed split/merge of this with file sharing and move out of the common file trading content into its own article, distinct from the generic concepts. Jamesday 16:32, 20 Mar 2004 (UTC)

Lists for eDonkey and ed2k should be merged.

Kademlia is not a protocol, but an algorithm. Different protocols based on that algorithm are already listed on the Kademlia page.

BitTorrent is not a network, it's a pair of protocols (tracker and peer). Each file or set of files being shared has its own network. (They may happen to share trackers, but they're not connected.)

I agree with the bittorrent opinion, BitTorrent is usually cited as a file-distribution system, although the activities that occur are similar. BitTorrent uses peer-to-peer technology, but not in the way it's listed under, someone clean it up please. zdude255

Does this article even make sense? Peer-to-peer is not the same thing as distributed filesharing. True, distributed filesharing is a very common use of P2P networks, but the two shouldn't be sysnonymous, especially not in a wikipedia article where folks that know better are supposed to inform folks who don't. How about something noting the the internet is a P2P network, or was, before ISP's began blocking ports and NATing connections? Then you could mention that some common uses of P2P networks are distributed filesharing, and lay out what the historical progression is from client-server to P2P (i.e. dialup/timesharing to the internet)? In most places, letting the distinction go unaddressed is fine, but on Wikipedia, I think we ought to do better, and be as precise as possible. drakaan

Peer-to-Peer is a networking protocol. Rewording is necessary -

Peer-to-peer is not a networking protocol. It is a term for describing or classifying networking protocols according to some of the characteristics they exhibit. Haakon 18:59, 4 March 2006 (UTC)
I agree with Haakon; to be more exact, I would say peer-to-peer is a network design principle, or perhaps a network topology. A good counterpart to the term peer-to-peer is Client-server, which in Wikipedia belongs to Category:Network architecture. The Client-server article begins with "Client server is network architecture which..." - this is a good start and I think Peer-to-peer should start the same way and belong to the same category. In an encyclopedic article we should definitely not mix up peer-to-peer principle and peer-to-peer filesharing. Ahtih 23:59, 5 December 2006 (UTC)
Unless others object, I am going to start changing this article over time towards describing P2P as network architecture. This will involve numerous changes, such as restructuring sections and moving most of filesharing content into filesharing-specific articles. Ahtih 23:55, 18 March 2007 (UTC)
I agree with Haakon and Ahtih. The current revision has only partially made this distinction between P2P and file sharing -- the section on legal controversy, for example, has no place in this article.
A minor point: I'm not sure that "network architecture" is the best way to describe P2P. Better choices would be "distributed systems architecture" or "distributed system design principle". For example, BitTorrent is definitely a distributed system but it might not be a network, depending on your definition. To me, network implies a distributed system whose purpose is to provide a communication infrastructure or service. Although BT nodes communicate, they do not provide a general communications service -- the only thing you can do is download a file, rather than ask BT to please send message M to node X for you.
Also, I see the term used much more often in its adjective form rather than its noun form.
Here is a candidate definition (working from Ahtih's much-improved current revision): "A peer-to-peer system is a distributed system whose component nodes participate in similar roles." (Take the distributed system article with a grain of salt as it is confused with distributed computing, in a similar way as this article has confused P2P and P2P file sharing.) --Nethgirb 01:01, 19 March 2007 (UTC)
In my eyes the term network is broad enough to cover BitTorrent etc as well, but I see your point and changed the article accordingly. I think there is no need to strictly replace all occurences of network though. Ahtih 23:29, 19 March 2007 (UTC)
I agree, it doesn't have to be so strict. There are certainly senses of the term "network" which apply to BitTorrent. My perspective is mostly from the academic computer science literature, but even there, it's not so precisely defined. --Nethgirb 00:03, 20 March 2007 (UTC)

The use of the term "first generation" and so on, referring to Napster and it successors, should (at the least) have a disclaimer indicating that this is merely a popular usage among the "file sharing" crowd (or better yet, it should be jettisoned as the uninformed misinformation that it is). That was by no stretch of the imagination the first generation of peer-to-peer networking systems, and not even the first to be used widely outside of academia and research. Apple's LocalTalk, Artisoft's LANtastic, Microsoft's (still in use) workgroup-based "file and printer sharing", and a dozen other widely-used peer-to-peer networking packages pre-dated Napster and the whole music-and-movie bootlegging movement by a decade or more. Tverbeek 02:05, 24 Oct 2004

It's true that P2P is nowadays used or rather confused with file sharing over a peer-to-peer network. The term itself is much older and peer-to-peer protocols/services exist since ages. Personally, I recall to have the term when I used OS/2 although I don't remember which service was meant there. Though, most of those examples you've given are not very wide-spread nor representative, in my opinion. NNTP (Usenet) would be a better example and DNS has some peer-to-peer characteristics as well. One difference may be that older peer-to-peer services/protocols ran on servers and were installed by admins i.e., the servers of the networks were connected in a peer-to-peer fashion but there was still a client-side and a normal user would use the client-side only. The now popular P2P systems are ran by simple users on non-server systems. -- 17:38, 10 Jun 2005 (UTC)
To go back and find the first instance of Peer-to-peer file sharing you have to look at specific things. #1 was the network used routing like the internet? #2 did the network have home users? #3 did the software handle requests for lists of files located at the node? #4 did that node independently respond to those requests and send the file back through the network following the same routing system as the request was made? #5 Was the network large enough to support P2P for users and not an experiment? #6 was an end user able to easily install it? #7 was it widely distributed around the world? This constitutes the first use of peer-to-peer file sharing. In that way the first use of this was WWIVnet which was the 3rd largest user end network in the world at one time if you want to include the Internet as a user end network. Yet it was inaccessible to the end user until around the mid 90's. Which may make WWIVnet the largest decentralized network in the world that was accessible to the end user or home user. --Deathmolor (talk) 16:28, 14 February 2009 (UTC)

External links

There are way too many extraneous external links in this article. It needs serious trimming by someone who knows what is relevant and what is not. Rl 18:31, 16 May 2005 (UTC)

No kidding. I trimmed it down seriously. In my opinion, we should only link to resources that are essential for a deeper understanding of the subject. The point is not to provide a comprehensive collection of links to everything related -- for that, we have dmoz. So first, I removed everything only pertaining to file sharing, since this article is about peer-to-peer and not file sharing. Then, I removed the links to special-interest or peculiar sites, or to pure company sites. I think the sites we are left with are good. Sites can be added if they are relevant. Haakon 19:46, 28 October 2005 (UTC)
I find this Article is lacking structure. I like the history part, however , I think after that the paper should go from possible applications (including, but beyond file sharing), legal and security problems to the technical realisation in nowadays P2P systems and upcoming research.muellerw 18:37, 9 November 2005 (UTC)

Pro P2P Business Models

It may be relevant to add information regarding Pro P2P business models where P2P technology is a key factor in reducing distribution costs while retaining a new revenue structure.

—Preceding unsigned comment added by CortlandKlein (talkcontribs) 00:59, 27 May 2005 (UTC)

Wikipedia p2p?

Isn't Peer-to-peer a good idea for a non-commercial project like wikipedia? I have recently seen that project Gutenberg employs this technique. I like the idea. It cuts the costs and thus the dependencies on donations.

— Preceding unsigned comment added by (talk) 12:46, 27 May 2005 (UTC)


What about Dijjer integration via the dijjer:// pseudo-protocool? Unify all traffic in HTTP.

— Preceding unsigned comment added by (talk) 02:02, 23 December 2005 (UTC)

Why not its own category???

Why did Peer-to-peer lose its own category? File sharing is only one aspect of what peer-to-peer networks and programs are used for.

I agree. --Aravind 15:01, 30 May 2006 (UTC)

Have some info

Although using p2p networks as a source of music sharing has gone down to 41%, people are finding ways to keep sharing. 23% of the U.S. is still practicing swapping music without paying. A recent survey of the same people was taken and found that: 19% share though other people's MP3 player, 28% said email and instant messaging and 4% through blogging. It was a survey of 1500 people. Done in Jan-Feb 2005 and again in 2005.

Info from:

Source: P2P Use Is Down, But Piracy Has New Outlets, Study Says Antony Bruno Billboard; Apr 9, 2005; 117, 15; JuniorQuest Magazines pg. 8

— Preceding unsigned comment added by (talk) 22:56, 5 June 2005 (UTC)

Ambient network link

Hi! I added "Ambient network" as a link to this page's "See also". Ambient network is basically a network infrastructure that uses the peer-to-peer paradigm to allow the merging of network functions (i.e. connection, but also other functions). It seems to be the future, at least for me. --Msoos 12:46, 31 July 2005 (UTC) moo


A clear definition of the (first, second and third) generations of P2P-clients would be pleasant.--DXL 20:54, 11 December 2005 (UTC)

These "generations" are a file sharing phenomenon and does not apply to peer-to-peer in general. They are treated in the file sharing article. Haakon 21:10, 11 December 2005 (UTC)
I see, thanks for the information. --DXL 14:24, 13 December 2005 (UTC)

Networks, protocols and applications

This section is sick. I still don't know what to do with it. It is equal to having a software written in C language section in C programming language article. I was going to add DALiWorld because it is very different from many other p2p uses, but currently it makes no sense to do so. --Easyas12c 00:11, 28 December 2005 (UTC)

This article needs some serious cleaning...

The article is fairly well done until the 'Networks, Protocols and Applications' section where it balloons into a horrendous mess. I cleaned up the multi-network clients some, but it'll be a whole host more work for the networks...

—Preceding unsigned comment added by FrYGuY (talkcontribs) 07:59, 31 December 2005 (UTC)

Gnutella: Acquisition, BearShare, BetBug, Cabos, Gnucleus, Grokster, iMesh, gtk-gnutella, Kiwi Alpha, LimeWire, FrostWire, MLdonkey, mlMac, Morpheus, Phex, Poisoned, Swapper (this link seems to need some work; it doesn't actually take you a description of the p2p program as the other links do, but instead links you to Lilo and Stitch), Shareaza, XoloX

—Preceding unsigned comment added by Pictish (talkcontribs) 06:22, 1 February 2006 (UTC)

Hybrid P2P

Can someone explain what "Route terminals are used addresses, which are referenced by a set of indices to obtain an absolute address." under Hybrid P2P means in the article?

— Preceding unsigned comment added by (talk) 14:27, 10 January 2006 (UTC)

Mixed P2P

Do we really want separate classifications for 'Pure P2P', 'Hybrid P2P' and 'Mixed P2P'.

The current classification is very unclear. I think what is really wanted here is just two classifications - 'Pure P2P' should be one of them, and 'Hybrid P2P' - being Pure P2P with centralised elements, should be the other. Comments? --Matthewleslie 14:38, 1 March 2006 (UTC)

Structure and Unstructured

No mention is given to the classification of peer to peer architectures into structured and unstructured. This is a commonly used classification, to differentiate between search networks like gnutella, and adressable networks like Chord or CAN. This should really be added? --Matthewleslie 14:41, 1 March 2006 (UTC)

I just added a brief structured vs unstructured classification. Most of the structured p2p need wiki pages. I am working on it. --Aravind 15:00, 30 May 2006 (UTC)

These descriptions are relatively poor and have misstatements of fact regarding scoped flooding and other techniques used for unstructured networks. Strongly recommend a literature review...these assertions are outdated and proven wrong.

Application of P2P Network

(Illegal) File-sharing seems to dominate any discussion of P2P networks, however these can be put to other less controversial uses. I'd like to make (as far as possible) a comprehensive list: So far the list I've got (far from comprehensive is):

File sharing
Grid applications (there will be sub-applications here)
Digital Cinema

Please add to this list and then it could be appended to the article.

Does anyone know the point of the useless TV reference in the "Application of P2P Network outside Computer Science" section? No explanation is offered for that bullet item so I would recommend deleting it unless someone wants to explain what specific applications they are referring to. amRadioHed 04:26, 18 March 2007 (UTC)

The Creative Commons License?

(referring to the new type of licenses which recognize individual authorship but not exclusive property rights, such as the GNU General Public License and the Creative Commons License)

Create Commons is not a single license but a set of licenses. Not all of them are copyleft licenses like the GNU GPL. "and some Creative Commons Licenses" is a more correct phrase. Sunil Mohan 18:35, 6 March 2006 (UTC)


It seemed a little odd to put the GNUNet logo at the top of the article. Wonderful though GNUNet no doubt is, does it really deserve such a prominent picture? —Preceding unsigned comment added by Matthewleslie (talkcontribs)

I thought exactly the same, but been too jaded to say anything. It's a strange and arbitrary choice of image. Haakon 09:12, 18 April 2006 (UTC)

Beyond a technical definition

This existing p2p page takes an exclusively technological definition of peer to peer, while the same distributed dynamic exists in every distributed system. Thus the peer to peer meme is now being generally used in the human organisational field and many other areas, most notably by Yochai Benkler and his Wealth of Networks on commons-based peer production. In my own work: I distinguish peer production, peer governance, peer property. Merging it with filesharing would make it even narrower still. So I think we should go in the opposite direction. 1) give a general abstract definition of peer to peer 2) then how it applies in the various field, including, but not exclusively, as a technical architecture. - Michel Bauwens,

— Preceding unsigned comment added by (talk) 12:45, 19 May 2006 (UTC)

i dont get it

How do people know about all these p2p file sharing networks? Even though they might not be public or non-public??? Call of duty 18:30, 21 June 2006 (UTC)

OS Downloads

Hey, is it true that entire OS', such as Windows XP can be downloaded off of a P2P? The Wretched 04:29, 12 July 2006 (UTC)

If it's made of bits, P2P can move it. Whether it works when you get it, is your problem, and irrelevant to this article.WolfKeeper 04:50, 12 July 2006 (UTC)
Yeah, about that, is there a way to make it work besides putting through optical disc authoring program? The Wretched 08:00, 13 July 2006 (UTC)

Do NOT Merge

P2P Streaming and Peer-to-Peer Networks in general are VERY different topics, and should not be merged into a single article.

— Preceding unsigned comment added by (talk) 19:23, 25 October 2006 (UTC)

Requesting correction of one sentence (EU Copyright Directive)

The article states that "The EU Copyright Directive ... prohibits peer to peer, claiming it is a violation of the directive". This sentence is really misleading and should at least be written: "The EU Copyright Directive ... prohibits distribution of copyrighted material". The distribution of personal material (my holiday photos, for instance) over p2p networks is NOT barred under European law: it would be insane. — Preceding unsigned comment added by (talk) 21:54, 2 November 2006 (UTC)

Peer-to-Peer goes to "Weekend at Bernie's"

I found that when I clicked on a "peer to peer" hyperlink from the Windows Vista article it goes to an article about "Weekend at Bernie's" titled Peer-to-peer However there is an article on Peer-to-peer that is redirected from P2P. Can someone somehow delete the "Bernie" article titled "Peer to Peer"? I don't know how. I checked and there is another article which is the same as the "Bernie" one but it is titled "weekend at bernies" (so the P2P article with the wrong title may be deleted) Sorry if this sounds confusing!

I can't reproduce this problem.... — coelacan talk — 22:47, 29 November 2006 (UTC)

it happened to me too. really weird...


I removed all text pertaining to Usenet since it is not a P2P protocol. 13:39, 11 January 2007 (UTC)

Usenet certainly is a P2P protocol, as per the definition given in the article. Therefore, I'm reverting your changes. Wrs1864 14:12, 11 January 2007 (UTC)

While SMTP is client server at the network periphery, it is P2P at the core - as all relaying MTAs such as Sendmail, Exim, Postfix etc. have both client and server functionality. I think it will be useful to include a note about this aspect of SMTP alongside Usenet. --Copsewood 15:16, 6 March 2007 (UTC)


Attention all artists better than me....this article needs pictures.... 18:43, 15 January 2007 (UTC)

Current image contain six computers connected each to each, which causes image of "star of david" appear. I think this could be offensive to people with other religious views. So i reccomend change this number to five, or seven. —Preceding unsigned comment added by (talk) 19:41, 19 October 2008 (UTC)

Oh come on! What, have religions claimed ownership of that specific shape? To claim that no one can now use a six-pointed star because it is "offensive to religions" is ridiculus! —Preceding unsigned comment added by (talk) 22:01, 4 March 2009 (UTC)

Comparison of networks table

We have Comparison of file sharing applications. I think we should also have Comparison of file sharing networks, like Peer-to-peer#Networks.2C_protocols_and_applications. The big column I want to see is number of users, but things like centralized/decentralized, encryption, etc. would be good, too. — Omegatron 22:31, 20 February 2007 (UTC)

I agree; it would be a nice addition to wikipedia :) - Rubikfreak 00:37, 18 May 2007 (UTC)

Is there a big difference...

is there a big difference between this article and this one:

Yes, or at least there should be. "Peer-to-peer" is just a way of building applications. You hear it most often in the term "P2P file sharing" but there are many other kinds of P2P systems as well, e.g. multicast, or BGP. --Nethgirb 21:07, 17 March 2007 (UTC)

Where are the P2P communications protocols?

I'm a newbie looking for info about SIP, XMPP/Jingle, and other such nifty things. I expected these to be referenced from this page. There was some general info about P2P protocols at the beginning, but it quickly digressed (as I saw it!) into file sharing protocols to the near exclusion of all others. Thanks!

  • You're right. The current state of the Networks, protocols and applications section is rather horrific. While many file sharing applications are peer-to-peer systems, we already have a different article for that. The whole sectional probably needs a complete rewrite from scratch. When it comes to SIP and XMPP/Jingle, those are client-server protocols. There is a P2PSIP working group going on at IETF developing a peer-to-peer version of SIP.--Teemuk 06:05, 18 May 2007 (UTC)

Peer-to-peer or File Sharing??

This article seems to talk almost exclusively of file sharing networks, while it should cover much more; for example P2P LANs (e.g. workgroups) and such. --Arny 16:52, 20 May 2007 (UTC)

-- I'll agree with that. Peer-to-Peer is a network architecture/topology, not Internet based file sharing. Discussions on File sharing have no place in this article. Internet based P2P networking refers to the fact that files are downloaded directly from one machine to another, its a loosely used term, as most services make use of a central tracking server.

Authoritative Reference

I would recommend reading "A survey of content distribution technologies", by Androutsellis-Theotokis and Spinellis ( to get informed about peer-to-peer architectures.

They propose the following definition:

Peer-to-peer systems are distributed systems consisting of interconnected nodes able to self-organize into network topologies with the purpose of sharing resources such as content, CPU cycles, storage and bandwidth, capable of adapting to failures and accommodating transient populations of nodes while mantaining acceptable connectivity and performance, without requiring the intermediation or support of a global centralized server or authority.

Besides, I think that this article should talk less on the filesharing topic, despite its popularity, it is only a particular application of p2p technologies. As described in Ian J. Taylor "From P2P to Web Services and Grids" examples of P2P applications could be:

File sharing/storage programs: Gnutella, Napster, Limewire and KaZaA.
CPU resource-sharing systems: SETI@home, United Devices, Entropia and XtremWeb.
Instant Messaging: ICQ, Jabber.
Conferencing applications: NetMeeting, Skype

Alberto Cuesta

( 09:03, 13 July 2007 (UTC))

University networks; block many applications

I am usually connected to the internet via a university network. This is ordinary cable, but unfortunately via the servers which are being operated by the university. They have succeeded in blocing all/most filesharing applications. First they blocked applications like kazaa(lite), later they also blocked bittorrent and Ares.

Many univesity networks block file sharing applications. Wouldn't it be interesting to explain in this article how they block these applications, explain possible solutions to cicumvent this and list applications which might still work? Especially students don't have a lot of money, so especially for them this blocking is nasty. 20:46, 21 September 2007 (UTC)

Its not really a Wikipedia kinda thing, i mean explaing how they block it maybe but possible ways to get around it im not sure that belongs here. (although id be interested to know and support that kinda knowledge been shared. I support our rights to share) Maybe that's the kinda of thing that belongs in one of the many how-to sites. I don't see any harm in a bit about how they are attempting 2 block P2p programs and networks not only unis but governments and ISP's

Wonx2150 (talk) 01:42, 7 September 2008 (UTC)

Article move to peer-to-peer (computing)

I'm proposing to move this article to peer-to-peer (computing).

I believe that most people that type in 'peer to peer' expect to get to a file sharing article, but will reach this article, which is a fine article on peer-to-peer, but is about a computer networking technology that most people probably won't understand or care that much about (regrettably).

I think we need peer-to-peer to redirect to file sharing and include disamb links back to peer-to-peer (computing).

I'm not completely happy with it, but it seems the right thing to do.

File sharing contains billions of links to peer-to-peer anyway, so we're breaking it to them gently. WolfKeeper 12:47, 9 October 2007 (UTC)

I don't share your belief that "most people that type in 'peer to peer' expect to get to a file sharing article." That isn't what "peer-to-peer" means. It's likely that most people familiar with the term associate it with file sharing, but that doesn't mean that they regard the two concepts as one and the same. —David Levy 12:55, 9 October 2007 (UTC)
Unfortunately in English, if people use the term in a particular way then that's what it means.WolfKeeper 13:04, 9 October 2007 (UTC)
As stated above, I dispute your assertion that most people use the term "peer-to-peer" to mean "file sharing." It commonly refers to a technology used for file sharing. —David Levy 13:35, 9 October 2007 (UTC)
The technology documented in the article is what is used for file sharing. It isn't as though we're sending people to an article about some other definition of the term "peer-to-peer." We're presenting an article that provides useful information about the relevant technology. —David Levy 12:55, 9 October 2007 (UTC)
Not really, not in the way that they can usually understand.WolfKeeper 13:04, 9 October 2007 (UTC)
My point is that this article documents precisely the technology that is used for file sharing (not some unrelated concept). I don't know how many people understand the technical aspects, but reading an encyclopedia article is a good way to learn. —David Levy 13:35, 9 October 2007 (UTC)
The file sharing article is linked at the top of the page. If that's what the reader seeks, it's one click away. —David Levy 12:55, 9 October 2007 (UTC)
I don't think they always know what they seek, and when they don't know, they probably want the file sharing article. The article needs to take them to the page that is likely to be most informative, which in this case is probably file sharing. Let's face it, that's what 99.99% of the time people use term to mean and use the technology for right now.WolfKeeper 13:04, 9 October 2007 (UTC)
Obviously, I disagree. Can you cite a source for this "99.99%" statistic? —David Levy 13:35, 9 October 2007 (UTC)
The title Peer-to-peer (computing) doesn't make sense, as it implies that the parenthetical term serves as disambiguation from one or more non-computing-related definitions of the term "peer-top-peer." This is misleading and unhelpful. —David Levy 12:55, 9 October 2007 (UTC)
Unfortunately for you, there already is a peer-to-peer (disambiguation) page, so that's untrue. The term already has multiple meanings.WolfKeeper 13:04, 9 October 2007 (UTC)
If the disambiguation page were located at the title Peer-to-peer, this would be a valid setup. But you redirected Peer-to-peer to an article pertaining to the very same computing-related definition of the term. It doesn't make sense to use the word "computing" to differentiate an article from one about computing. —David Levy 13:35, 9 October 2007 (UTC)
In any case I'm moving it to Peer-to-peer (computing) but leaving the redirect in place right now. The title helps explain the area of discourse at the very least and is more accurate.WolfKeeper 13:15, 9 October 2007 (UTC)
1. This is a contested move. Why did you bother proposing it above if you were just going to reinstitute it before receiving any outside feedback (let alone establishing consensus)?
2. We use parenthetical disambiguation only as necessary. We don't redirect Foo to Foo (disambiguating term), as this is entirely pointless. And how is Peer-to-peer (computing) "more accurate"? —David Levy 13:35, 9 October 2007 (UTC)
With all due respect David, I don't really think you're looking at it from the point of view of the users of the encyclopedia; you're looking at it from a purely technical point of view. To give an example, try doing a google on 'peer to peer'. You should find a number of pages came up, and most of them will be using the term 'peer to peer' to mean file sharing, although a very few might be about Skype. In particular the news organisations such as the BBC use peer to peer to mean file sharing. The file sharing article talks about peer to peer implementations almost exclusively anyway. For people unfamiliar with the term they probably want the file sharing article. For people mostly familiar with the term, they probably want file sharing. For computer experts and people who are more technically oriented (like me), they want Peer-to-peer (computing) which would be a single click away. I regret the news organisations error.WolfKeeper 14:03, 9 October 2007 (UTC)
I disagree with your interpretation of the available data. File sharing is the most common application of peer-to-peer technologies, so it makes sense that the term "peer-to-peer" is most often used in a context pertaining to file sharing. This does not mean that it's being used to mean "file sharing." I frequently see the phrases "peer-to-peer file sharing" and "P2P file sharing," which clearly convey an understanding that the two concepts are closely related but not one and the same. —David Levy 14:19, 9 October 2007 (UTC)
The News organizations could have used 'file sharing' very easily, but have chosen not to. You might like to consider why that is. I think by making this change we would be following current usage, while emphasising the fact that we don't consider it entirely accurate. This would hopefully have a positive impact on the terminology used in future, but I wouldn't like to bet on it, and that's not why I would be doing it. In any case I think that the current article positions are not very helpful.WolfKeeper 15:13, 9 October 2007 (UTC)
1. I provided links to Google searches for "peer-to-peer file sharing" (about 849,000 hits as of 17:10, 9 October 2007 (UTC)) and "P2P file sharing" (about 1,270,000 hits as of 17:10, 9 October 2007 (UTC)). Why is it so common to use the terms together? Are people saying "file sharing file sharing"?
2. Here is a link to a Google News search for "file sharing." That there are 3,837 hits (as of 17:10, 9 October 2007 (UTC)) contradicts your assertion that the news organizations "have chosen not to" use this term. —David Levy 17:10, 9 October 2007 (UTC)
Why the heck did you just perform a copy-and-paste move?! —David Levy 13:39, 9 October 2007 (UTC)
Well? —David Levy 14:19, 9 October 2007 (UTC)
I still await your explanation. —David Levy 17:10, 9 October 2007 (UTC)

Asynchronous file transfer with p2p?

Are there any P2P protocols suited to file transfer between two users like ftp only asynchronously? For example, user A wants to send a file to user B but B is offline. A sends anyway, B comes online later and receives the file. Pgr94 (talk) 16:48, 18 April 2008 (UTC)

Attacks on P2P networks section

I removed this section; it was a near-verbatim duplicate of a section in the File sharing article; since this information all appears to be specific to P2P file-sharing apps as opposed to P2P as a networking concept in general, it seems more appropriate to have it in that article. There were two sentences at the end of the section in this article that weren't in the file-sharing article; I've inserted those in the FS article for now, althought they're unsourced and kind of vague, so they maybe should just be deleted entirely.

evildeathmath 16:00, 23 May 2008 (UTC)


Added Source for "Advantages of p2p", found via google link. I think this is correct. Not sure though if there is permission to reproduce this, although it's only 1 paragraph long and i don't think i could put it better myself. silvarbullet1 (talk) 00:47, 26 May 2008 (UTC)

Other movements

P2P is a very good metaphore for peerialistic movements in general, you know like open source, file sharing (whether illegal or not), information sharing in general, networks of cooperation where investements are returned in thousandfold, etc. etc. I would like to see links to these kinds of phenomena. —Preceding unsigned comment added by RedMurcury1 (talkcontribs) 14:57, 21 July 2008 (UTC)

Should xfire be on the list as Windows is?

Dosnt xfire run a p2p client between its programs for downloading files from there server? so its a controlled p2p network. Im no expert but that's the impression i got as to how it works anyone know more?

Wonx2150 (talk) 01:33, 7 September 2008 (UTC)

Adding External link to website on P2P research

The external links section contains several links to websites related to published books on P2P technology. Most of these are out of date. Here is a new site which contains: 1) a comprehensive glossary of P2P terminology, 2) over 500 references to research literature, 3) links to leading conferences and journals publishing P2P research, 4) research tools including simulators, emulators, and implementations. I am SME working in this field for 5+ years and I believe this new link is a good addition, and is of the same character as other links already in the list, such as and, but with more information and resources. —Preceding unsigned comment added by Mobilon (talkcontribs) 16:03, 5 December 2008 (UTC)

P4P cometh

I'd like to add a section on P4P and have wiki searches for P4P direct to that section in the p2p article. Searching for P4P results in this page: How do I get a sub-section of p2p to show up on that result page? —Preceding unsigned comment added by Hyper formance (talkcontribs) 03:25, 3 January 2009 (UTC)


I would like to beg to differ on the FidoNet part of this document. Within the already established FidoNet description, technically it is a hierarchical structure, and was actually critically attacked by sysops for its hierarchical structure. Furthermore the term Hub was used a lot to define the network. The hub, acting as a distribution point. Defying the whole definition of a P2P network. WWIVnet was the second largest network of sysops in the world and actually did use a distributed and decentralized model. True P2P. So i am removing the Fidonet reference here and replacing it with the proper reference of WWIVnet. --Deathmolor (talk) 16:15, 14 February 2009 (UTC)


Included in the "Multi-Network Applications" table are DC++ and gtk-gnutella which each only support one P2P network. This is a direct contraction and should be somehow dealt with via the removal of the applications or correctionif they do indeed support multiple applications (which I believe not). Why should these be here when others (KTorrent, etc.) are not given there own row? —Preceding unsigned comment added by (talk) 22:57, 15 February 2009 (UTC)

Notablity of FilesWire and Zultrax

Both have a second network aditionnal to their standard network, but the so-called 'G3' just doesn't exist and the second network of Zultrax is also somehow not really important. Also, I really doubt the notability of the clients themselves. Zultrax is OK, somehow and seems to be recently updated etc., whereas FilesWire seems to be only proprietary crap... I'd suggest we should at least remove FilesWire from the list...

Greetings, Old Death (talk) 19:36, 19 May 2009 (UTC)

P2P is a type of ad hoc network

We should place P2P within the ad hoc distributed networking paradigm. I would contend that it is the equivalent, but others would not. Recently Wireless ad hoc networking is starting to capture the term "ad hoc" for itself, but this is incorrect. Wireless ad hoc networking is just a kind of P2P, with all the related problems of free-riding, security, etc.

Ooskapenaar (talk) 13:54, 4 June 2009 (UTC)


A large portion of this article is not in the right place, or needs cleanup, or is original research. Most of this confusion is due to not distinguishing between peer to peer networks as a whole, and file sharing. I'll be deleting parts of this article, or moving them to talk. Then I'm going to merge this and client-server into Network architecture to eliminate the redundancies. Because the vast, vast majority of what links here is strictly in the context of peer-to-peer file sharing, I will make this page redirect to file sharing.   M   06:11, 11 June 2009 (UTC)

Would a merge or a move be more appropriate?   M   20:30, 14 June 2009 (UTC)


You claimed that some of the article was copyvio, and deleted it. I found that that the date that the material was present in the wikipedia is prior to the date quoted on the 'original' i.e. the copying is the other way around; since the undo feature failed I reverted that edit and all subsequent ones. Please be more careful not to delete material from the wikipedia for false reasons.- (User) Wolfkeeper (Talk) 19:24, 14 June 2009 (UTC)

I'm sorry, but unless you can show that you actually did the work to show that any of the 'copyvio's actually were copyvio's then I will revert back to the earliest claim you've made along these lines. Your edits do not have any credibility right now.- (User) Wolfkeeper (Talk) 19:29, 14 June 2009 (UTC)

I see, and I think you mean this old version. Ok, I will re-insert that content, and clean it up.   M   19:35, 14 June 2009 (UTC)
I'm pretty sure that is also copied from here, not the other way around, but you should check it. There's wholesale copying from the wikipedia on this topic; this isn't illegal for them to do, although they are supposed to state where they got it from, which they practically never do.- (User) Wolfkeeper (Talk) 19:39, 14 June 2009 (UTC)
Yeah, I'll double-check that. If that's the case, then someone managed to use a 2 year old copy of this article as a citation for the current copy of this article. I probably won't re-add much. Most of the article is uncited original research, and repetition of 'not like client/server':
"A pure P2P network does not have the notion of clients or servers but only equal peer nodes that simultaneously function as both "clients" and "servers" to the other nodes on the network"
That, for example, won't be re-inserted. It's redundant, and simply confused and contradictory. A peer doesn't function as "both client and server", rather they both consume and supply resources. The entire article is halfway to being incoherent.   M   20:00, 14 June 2009 (UTC)
I'm sure a lot of things are contradictory and confusing if you don't understand them; and it's looking like you may not understand the topic.- (User) Wolfkeeper (Talk) 20:06, 14 June 2009 (UTC)
Do you have a content-related objection to what I said above, or to any of my edits?   M   20:09, 14 June 2009 (UTC)
The article as it stands currently simply doesn't cover the subject. The lead if anything is slightly better than before, but even that is too short now, and violates WP:LEAD in lots of ways. Everything else is not as good. The old article was somewhat sprawling and could have done with a prune, but was still better than as it is right now.- (User) Wolfkeeper (Talk) 20:13, 14 June 2009 (UTC)
<undent> The previous lead was:
"...uses diverse connectivity between participants in a network and the cumulative bandwidth of network participants rather than conventional centralized resources where a relatively low number of servers provide the core value to a service or application."
"diverse connectivity" and "core value" are both meaningless in this context; "relatively low number of servers" is presumably relative to the number of servers in a peer-to-peer network (which is something like 0, with perhaps some bootstrap or indexing servers). The article is full of language like this. It's often difficult to distinguish a giant block of this sort of text from something coherent, which is why I'd like to go over specific examples of where I might have removed meaningful content. Do you have some specific section of cleanup that you object to?   M   20:26, 14 June 2009 (UTC)


The concepts of 'P2P' as well as 'client-server' computing have nothing to do with 'network architecture', other than that both are Application Layer service delivery models in IP networks that can be transacted over any network architecture that IP supports. Before you make any such changes, please learn about the concepts and seek consensus on the relevant talk pages.

P2P is a distinct topic from file sharing. If you think this article doesn't cover other applications sufficiently, then please add what is lacking, not prevenany further improvements of this article by eliminating it and denaturing its intents and goals.

M, what you claim as cleanups here are more or less distortions of content with a highly biased view of what filesharing and related technologies are. Why don't you first clean up the filesharing article from the complete mess and confusion you created by the existing mergers into that article before you ruin this article as well. Sorry for this what could be viewed as a personal attack, but I don't know how else to state the situation. Sometimes, one has to address the problem head on. Kbrose (talk) 20:42, 14 June 2009 (UTC)

That's not a personal attack. I'm in the process of doing that cleanup, but since frequent edits are frowned upon by some editors, I'm giving a few days for the dust to settle. When the problem with the copyvio removal was brought to my attention, I re-inserted the material (but this re-insertion was reverted - apparently they didn't look at the diff). If we include 'overlay network architecture' into 'network architecture' (which we might not want to do), then it would be a good place to merge it. I agree that peer-to-peer and file sharing are entirely different - this is why I've tried to remove much of the social/legal junk from this technical article.   M   21:26, 14 June 2009 (UTC)

I have reverted the mutilation of the lede of this article and provided a correct statement of P2P definition, based on notable professional resources. Kbrose (talk) 21:05, 14 June 2009 (UTC)

I've reverted the article back to 2009-06-14T00:44:07, because the claimed copyvio did not exist, some of the deleted text was found in 2005, but the 'original' only claimed 2006 copyright. It was too complicated to redo all the intermediate edits; I apologise for that.- (User) Wolfkeeper (Talk) 21:12, 14 June 2009 (UTC)
This has nothing to do with the copyvio - incidentally, you deleted a rather good definition of peer-to-peer that Kbrose added just recently. Do you have any specific objections to my non-copyvio edits?   M   21:26, 14 June 2009 (UTC)
I've also reverted a good number of edits you made to other articles, including changing the size of one article up to 85k. The guideline is that articles should be about 30k, and the original article was only slightly above that.- (User) Wolfkeeper (Talk) 21:36, 14 June 2009 (UTC)
I think in general you need to get agreement before making major changes to the structure of the wikipedia's article; and removing material that probably has been contributed in good faith and then claiming that it is copyvio, without checking carefully, and particularly when it actually isn't, violates the norms of the wikipedia.- (User) Wolfkeeper (Talk) 21:36, 14 June 2009 (UTC)
See WP:BOLD. It would be great if I had more time to very carefully check every one of my edits, but I don't. I reinserted the copyvio edits as soon as you informed me of them, and told you that I would reinsert the other ones shortly. Unfortunately, you reverted pretty much all the cleanup. Even if half of what I removed should have remained (I seriously doubt this), you should still have given me time to fix what you had a problem with, instead of reverting my edits to this article - and also most of my edits in the past few days.   M   21:42, 14 June 2009 (UTC)
Your edits to multiple articles have been extensive and so far as I can tell, overall, entirely non consensus changes.- (User) Wolfkeeper (Talk) 22:41, 14 June 2009 (UTC)
The article version you produced at [1] is the least encyclopedic excuse for an article I have ever seen produced by any editor on any major topic in my entire time in the wikipedia.- (User) Wolfkeeper (Talk) 22:41, 14 June 2009 (UTC)
Yes, that's because I removed a lot of OR, nonsense, and so on. I've asked you a number of times now to point to something substantial that I removed, and should not have. If I removed no meaningful content, then what you're basically telling me here is that you think that "least encyclopedic excuse for an article ever" somehow becomes better when you add meaningless and irrelevant content, and OR. Point to something specific that I should not have removed.   M   22:56, 14 June 2009 (UTC)
No, it's because you removed information about their history and how they work. Encyclopedia articles aren't simply extended definitions of something, they're really supposed (within length limits and so forth) to talk about all aspects of them.- (User) Wolfkeeper (Talk) 23:02, 14 June 2009 (UTC)
Do you have a specific example of where I removed 'history and how they work'? I'm looking for a sentence or something. Once you give me this, I can know what you're talking about, and will be able to ask you why you object to certain changes that I made. You'll reply, and so on, and pretty soon we'll figure out a good solution to our problems. Do you have a specific example?   M   23:08, 14 June 2009 (UTC)
The version you created[2] doesn't contain any information on how they work. It lists some applications that use P2P technology, and then lists a few things in the history... and that's it. The current article explains how P2P networks are constructed and managed, how they deal with nodes entering and leaving the network and it mentions current major P2P networks and applications. These are the kinds of things that an encyclopedic article are expected to contain. If anything the current article doesn't cover enough ground. Encyclopedia articles are supposed to be 'encyclopedic'; that has connotations of generality and completeness.- (User) Wolfkeeper (Talk) 23:31, 14 June 2009 (UTC)
Do you not have a specific example? It should be very easy to find one. Look in the version before my edits for a sentence or two that is missing from the current version. Exclude the copyvio blankings, since these are not the issue. The sentence should contain something that obviously belongs in this article. Paste it in below. Then we'll have some common ground to work from.   M   23:40, 14 June 2009 (UTC)
I do not wish to build common ground with you, I simply can't be bothered with you or your moronic games. That is all I have to say.- (User) Wolfkeeper (Talk) 00:55, 15 June 2009 (UTC)
Ok, so you have absolutely no concrete example to support your case. If you don't care to explain yourself, then at least please don't screw up the work of other editors.   M   18:02, 15 June 2009 (UTC)

My observation has been that User:M makes sweeping modifications that few if any other editors agree with and then engages in circular discussion requesting specific reasons for reverting with respect to specific edit details while never providing any substantive or correct reasoning himself for implementing changes in the first place. This leads to endless arguments as evidenced in this thread until his opponents simply go away in disgust of M's behavior or tactics. Kbrose (talk) 18:30, 15 June 2009 (UTC)

Unfortunately, this has not prevented you from getting involved with and escalating such disputes. Most of your contributions to establishing consensus involve jumping in on objections when any changes are made to these articles. The previous (and only) dispute was with User:Deathmolor, who made repeated bad-faith edits (inserting a program which he seems to have written) and was at one point blocked under WP:LEGAL. I explain myself both on this very talk page, and in my edit summaries. Would you like to start dispute resolution? Perhaps an rfc, or mediation?   M   18:41, 15 June 2009 (UTC)

I think we should remove that 'Windows P2P' thing in its current form from the P2P networks table. The text should either get its own paragraph in the article itself or be recuced to a minimum. Also I'd like to have examples of actual programs using it besides a Win7 Help functionality and some Remote Assistance usage.

Also, is there anywhere in the article written that WLM/MSN and ICQ also use P2P comparable techniques for their avatar transmission, custom smiley support (in the case of WLM) and file send/recieve/ resp. shared folders? I think that fact should be notable enough to be included. —Preceding unsigned comment added by Old Death (talkcontribs) 13:26, 22 June 2009 (UTC)

Greetings, OldDeath (talk) 13:27, 22 June 2009 (UTC)

Yeah, I was looking at it and thinking the same thing before checking this talk page. It should just be dumped out into its own article's talk, and I don't feel that bad about deleting it. Yeah, find a source for the icq thing and add it.   M   08:29, 26 June 2009 (UTC)

Multi-network applications (comparison table)

I don't know if clients having only eD2k and KAD support should be mentionned in this table, as KAD is +/- an addition to the traditional eD2k protocol. If we give these clients a place in this table, we have to add LimeWire also as multi-network application. Greetings, Old (talk) 22:31, 14 June 2009 (UTC)

I'm not sure why this article seems to focus on file sharing clients at all, and a giant list of them is entirely inappropriate (which is why I moved it to file sharing).   M   22:39, 14 June 2009 (UTC)
At least the first table does clearly not focus on filesharing, it only lists P2P protocols/networks and their resp. applications. Only, as most P2P networks are filesharing networks, this may generate the impression of focussing on the subject. Greetings, Old (talk) 23:32, 14 June 2009 (UTC)
Have a look at the contents of this book. This article lists every random file sharing network, and neglects the wide range of other p2p and related programs, such as jabber, folding@home, dns, mixmaster, tor, and so on. If we remove the clients (which belong elsewhere), and group the entries by the second column, then we'll be left with a very short list.   M   23:53, 14 June 2009 (UTC)
DNS and folding@home are not generally considered to be P2P networks, they are usually considered to be client server. Tor is P2P, I don't know about mixmaster or jabber.- (User) Wolfkeeper (Talk) 00:51, 15 June 2009 (UTC)
Both DNS and folding@home have significant p2p components, and should be mentioned as related programs.   M   18:05, 15 June 2009 (UTC)
No. They download the data to work with from their central server, they don't exchange data directly between the clients and everything the clients do is directly coordinated from one single entity. There is no P2P in these distributed computering projects (except for distribution, and that's a really poor character to identify P2P with).
EDIT: back2topic: I am still under the impression that applications featuring only KAD and eD2k should not be mentionned in a multi network comparison table, for the reasons I gave above. Greetings, Old (talk) 14:05, 17 June 2009 (UTC)
One of few prominent books on peer-to-peer technologies (and the current top result on amazon) dedicates a chapter to SETI@home. I read it a long time ago, so I don't remember exactly what it says, but distributed computing of the sort where regular people share their resources to do some serious computation was something important to the sort of peer-to-peer technologies we see now. It doesn't matter if it's peer-to-peer itself, it's a related technology. Anyway, the point is just to show that we're focusing on file sharing too much. As for the tables, I don't think that any clients should be mentioned, and that we can replace most of those tables with about 6 file sharing networks, in a short list. What do you think of that?   M   04:34, 18 June 2009 (UTC)
As I said, this is only related because of its distributed nature, not because of anything else. There is no problem mentionning it somewhere, but giving it more than two sentences in the entire article would already be too much.
The problem I see when we try to focuss less on filesharing by removing the networks & clients tables is that we would have to name only the most notable P2P networks... which means that there would be exclusively filesharing networks listed in the article, because that's the most important usage of P2P. Also, I think the tables are very handy, because they are relatively complete. Therefore, it should already be enough if we move the tables to the botton of the page ( = delete 'History' and move that 'Social issues' thing up).
Greetings, Old (talk) 10:57, 18 June 2009 (UTC)


I think this should be deleted in its current form (and only the timeline link kept) or be rephrased into somethign a bit more worth than only a list of dates that somebody thought might be important. Greetings, Old (talk) 14:39, 17 June 2009 (UTC)

It's usually detrimental to split an article into two parts, even if it is huge - subjects should be split, not articles. Sometimes, a brand new subject arises in the article, or some unimportant aspect grows enough to merit a chopping-back-to-size, and a new article. The history, though, is both central and important, and a history isn't the same as a timeline. One problem is that it's redundant with the chronologies in the copyright sections. As for the dates, fix them up. They're pretty much the important events from the timeline. Except for that whole paragraph specifically about Usenet that some editors want left in there.   M   04:44, 18 June 2009 (UTC)
I was talking about this:


I wasn't talking about what's mentionned under Generalities.
Greetings, Old (talk) 10:49, 18 June 2009 (UTC)
Hmm, I think I was confused. Yes, this should be removed.   M   23:54, 21 June 2009 (UTC)

Move to Peer-to-peer (network architecture)

Many of the what-links-here articles link here in the context of file sharing. This has led to a fair bit of confusion, and a lot of content that belongs in file-sharing frequently ends up being added here, since file sharing and peer-to-peer are often synonymous. This article, however, is about the technical concept. A redirect page at peer-to-peer will make this distinction, and link to both the move destination and file sharing.   M   01:12, 22 June 2009 (UTC)

Do you mean a disambiguation page? Two-article disambiguation pages seldom are beneficial, as they merely ensure that no one visiting the title arrives at the intended article. We link to the file sharing article via a hatnote, so it already is one click away. Adding an additional step for people seeking information about the technical concept wouldn't help anyone. —David Levy 02:19, 22 June 2009 (UTC)
The problem is that they are not seeking information about a technical concept. The client-server architecture is much more popular, and seems to be a non-niche term, yet this article gets many more hits (consistent with other file sharing articles). The problem is that people come here expecting to find file sharing information - when they don't, they add it, which is not appropriate. I think there are at least 4 or 5 entries that could be added to that disambig page too, though this and file sharing would be the two main ones. The last point is that a disambig page would encourage cleanup of a large number of links that confuse file sharing and peer to peer.   M   08:33, 26 June 2009 (UTC)

Ok, done. Now we can track what the readers are doing over the next couple of days to see how this is working out.   M   02:25, 1 July 2009 (UTC)

Requested move

The following is a closed discussion of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the proposal was no move. The arguments in favor of the need for p2p being a disambig page are not as strong as those in favor of p2p being the basis for all uses of p2p networks, for example file sharing. (talk) 00:29, 8 July 2009 (UTC)

Peer-to-peerPeer-to-peer (network architecture) — "Peer-to-peer" is used both to refer to the network architecture and to peer-to-peer file sharing (and some others). Readers and editors often expect to see an article about file sharing here, which is why this article is more popular than its client-server counterpart, why most of the what-links-here link here in the context of file sharing, and why a lot of content that belongs in file-sharing frequently ends up being added here. A redirect page at peer-to-peer will make this distinction, and link to both the move destination and file sharing. It would also let us look at the stats tool over the course of one day to determine where readers are actually going.   M   19:33, 1 July 2009 (UTC)

Peer-to-peer file sharing is already discussed in File sharing, so I don't think another article is necessary. That page is also linked in a hatnote on this page. It would be somewhat inaccurate to call file sharing "peer-to-peer" as opposed to "peer-to-peer file sharing", and we don't generally create articles about adjectives per WP:NC. This seems to be the one thing that is accurately referred to as "peer-to-peer" in a noun form. There's also a note on my talk page to the effect that the contents of this page are not accurately referred to as "network architecture". Dekimasuよ! 00:06, 2 July 2009 (UTC)
Peer-to-peer file sharing is just file sharing using peer-to-peer network technology.- (User) Wolfkeeper (Talk) 01:32, 2 July 2009 (UTC)
That's kind of my point. File sharing is file sharing; peer-to-peer is peer-to-peer; they are discussed in their respective articles. The type of file sharing called "peer-to-peer file sharing" doesn't need to have a separate article because it is already discussed in its parent article. Even if it did have an article, it wouldn't be a valid dab target for "peer-to-peer" by itself, just like we don't need to make Water a dab page in order to distinguish it from "Water skiing". Dekimasuよ! 02:15, 2 July 2009 (UTC)
File sharing was popularized through a number of peer-to-peer programs/networks, so the names peer-to-peer and p2p (cf "p2p traffic", "p2p program") have stuck as a reference to file-sharing in general. Yes, it's not perfectly correct, but it's what people are looking for. And it's also a pain to clean out a bunch of file-sharing info from this article because editors think it's talking about file sharing. The two media citations above, by the way, appear to concern file sharing.   M   03:58, 2 July 2009 (UTC)
It's clearly not about file sharing. That might be a reason to put a hidden message on the page asking editors not to add file-sharing info to the page, or perhaps strengthen the hatnote, but it's not a reason to move it from the correct title. I don't see any problem here so far that wouldn't be solved by linking peer-to-peer and file sharing separately in running text: peer-to-peer file sharing is simple and accurate. Dekimasuよ! 04:18, 2 July 2009 (UTC)
The insertion is a symptom of general confusion about the terminology. The hidden message addresses the symptom that we are aware of, but there's still the problem of readers in general being confused about the terms.   M   04:45, 2 July 2009 (UTC)
I don't think many people could read the article and end up confused. As far as the "pain to remove" thing you were talking about, I've been looking through the page history, and it looks like only one editor has made that mistake over the last month and a half, so that doesn't appear to be too widespread a problem either. Dekimasuよ! 04:52, 2 July 2009 (UTC)
I suppose I should make it clear that I oppose this move. Dekimasuよ! 04:58, 2 July 2009 (UTC)
  • Opposed: The premise of the proposal is simply a misunderstanding of terms. It is wrong. The intent of this article is a discussion of the general principle of 'peer-to-peer'. P2P is a distributed application design, or architecture, similar to the concept of 'client-server' design. These are not network architectures, despite the usage of the term network as a group of nodes. It is often also termed 'P2P computing'. If this article contains too much file sharing, which it does, then it should be rectified by providing a better treatise, not by manipulating links. Naming articles according to popular searches is not providing proper terms. P2P is not file sharing, and file sharing is not P2P. The user should be educated by proper usage and explanation in the article, not by misleading redirects and disambiguation. Kbrose (talk) 00:47, 2 July 2009 (UTC)
P2P computing is computing that uses P2P networking to connect up the nodes so that they can share computing resources. It's an application running on top of a P2P network architecture; the P2P network is an overlay network.- (User) Wolfkeeper (Talk) 01:32, 2 July 2009 (UTC)
Exactly. This article is about the house and how it is built. The article about file sharing is about the activity within the house; the house party. Binksternet (talk) 04:06, 2 July 2009 (UTC)
Thus, they're both at appropriate titles now. Dekimasuよ! 04:26, 2 July 2009 (UTC)
This is wrong. The term 'P2P networking' is not a method of networking, but a method of dividing an application into tasks or sharing resources. P2P is the act of sharing computing resources, not the goal of P2P networking. P2P uses ordinary networking, just like client-server. In fact, at finer granularity, P2P can be seen as bi-lateral client-server computing. Only *some* P2P system are built on an additional networking layer, such as onion router, which would constitute an overlay network. But P2P can simply be implemented directly on plain TCP/IP, or even other networking, such as pure OSI. The definition of P2P is clear, it has nothing to do with any specific network architecture and it does not constitute a new way of networking, only a different way of distributing application loads. Kbrose (talk) 02:19, 2 July 2009 (UTC)
Careful here. P2P networks don't necessarily change the application loads; they're often used as an index to discover the network's application-level resources, and then the resources are consumed client-server. In those cases the application loads are unaffected.- (User) Wolfkeeper (Talk) 02:36, 2 July 2009 (UTC)
As far as the name goes, the problem is that some people do (in my view incorrectly, but I don't think they'll listen to me) use the term 'peer to peer' to mean file sharing. It's wrong, but they do it.- (User) Wolfkeeper (Talk) 02:36, 2 July 2009 (UTC)
And for these individuals, the File sharing article is prominently linked at the top of the page. How would the proposed setup further assist such readers? —David Levy 03:35, 2 July 2009 (UTC)
If that prominent link did its job, then this page wouldn't see the very frequent addition of court cases, info about Napster and Kazaa, file sharing timelines, tables of near-exclusively file-sharing programs, and so on. Individuals who would like to read about the technical compsci concept can simply click the link. For the other individuals, who may not even realize that they're reading about just a small part of peer-to-peer file sharing (that is, the peer-to-peer part), they'll be directed to the right article.   M   04:04, 2 July 2009 (UTC)
You're acting as though the two concepts are distinct and unrelated. People frequently add information to articles that would be better placed in related articles. The confusion arises from the similarity of the subjects, not of their names (which is why there isn't a rampant problem of people mistakenly inserting seismology-related information at Tom Hanks). We don't move articles from their correct names for the purpose of steering users away so that they won't edit them. —David Levy 04:46, 2 July 2009 (UTC)
  • Oppose, per my comments in the section above this one. Both of the other articles that you listed on your disambiguation page already are linked at the top of this article (so you merely added an extra step for users seeking this article, while in no way assisting those seeking one of the others).
    It was inappropriate to unilaterally make this change (reverted by Dekimasu) when you knew that it was contested. —David Levy 01:16/01:20, 2 July 2009 (UTC)
If Tag redirected to Dog tag, and the people at Tag (game) requested that it become a disambiguation page, the fact that all articles involved had hats on them would be irrelevant. The important point is what readers expect to see when they come here. What we have here are two entirely distinct concepts: peer-to-peer the architecture and counterpart to client-server, and file sharing aka "peer-to-peer file sharing" which is often called simply "peer-to-peer" (p2p programs, p2p traffic, etc.). Given the hit counts of client-server and the frequent injection of file-sharing info (copyright, legal, programs), it seems clear that readers want to know about peer-to-peer file sharing, not about the very technical compsci concept. And to settle any doubts, we can simply observe where readers end up going once the redirect is in place.   M   03:48, 2 July 2009 (UTC)
1. Your "Tag" example is invalid; "Peer-to-peer" is not a redirect, let alone to a term of greater specificity. ("Peer-to-peer" is the subject's complete designation in common usage, while this obviously isn't the case with "tag"/"dog tag.") And the term "tag" has numeous meanings, while "peer-to-peer" has only two common meanings (which are interrelated) and one less common meaning.
Taking your argument to its logical conclusion, we would never use hatnotes instead of disambiguation pages. (For the record, when, in your assessment, are hatnotes appropriate?)
2. Again, the File sharing article is prominently linked at the top of the page. It already is one click away. Under your proposed setup, it still would be one click away, so it would be no more accessible than it is now. The only difference is that the Peer-to-peer article also would require an additional click to access (thereby inconveniencing readers seeking it). Instead of some readers immediately reaching their desired article and others needing to follow a link, everyone would need to follow a link. How would this be beneficial? —David Levy 04:46, 2 July 2009 (UTC)
  • SupportOppose: P2P is a networking technology.- (User) Wolfkeeper (Talk) 01:32, 2 July 2009 (UTC)
If that's what it is, unambiguously, we don't need a disambiguator. We wouldn't move Water to Water (liquid). Dekimasuよ! 02:53, 2 July 2009 (UTC)
This article is a networking technology, but the term "peer to peer" refers to both a) precisely but very narrowly to a network architecture, and b) imprecisely but very popularly to peer-to-peer file sharing. (And then, there are a few other things called peer-to-peer.) Hope that clears the intent up :)   M   03:52, 2 July 2009 (UTC)
I did a quick check, and Internet, Kazaa, Warez, and BitTorrent (protocol) all link to this article correctly. They are either discussing the networking aspect where they link this article, or they have text that runs "peer-to-peer file sharing". Can you show us where there are examples of people using "peer-to-peer" specifically to mean the act of sharing itself? Dekimasuよ! 04:13, 2 July 2009 (UTC)
Yes, someone has gone through a large number of articles and done the disambiguation within them (by adding peer-to-peer file sharing as a pair). This is imperfect - the link to peer-to-peer in Warez, for example, is inappropriate. Four articles, by the way, is not at all statistically significant. Instead of arguing about what readers are looking for, I had wanted to leave the redirect open for a few days, and then check the stats tool to see where readers were actually going. Unfortunately, this was reverted before we could gather any real data. Hopefully we can get consensus here to put the disambiguation page back up, if only for a few days.   M   04:28, 2 July 2009 (UTC)
Would the tool be able to differentiate between users arriving at the articles via the disambiguation page and those reaching them via other means? —David Levy 04:46, 2 July 2009 (UTC)
No. He thinks the number of people clicking through to this page would be considerably lower than the number of people reaching the dab page (some of whom would go to File sharing) and that this would show the topic's ambiguity. I'm not really comfortable with experimental moves, though. There's not really a precedent for it, and such an idea might be better discussed at WP:DAB than at an individual page. Dekimasuよ! 05:03, 2 July 2009 (UTC)
Also, it's undisputed that many of the readers arriving at the Peer-to-peer page seek the File sharing article (and to a lesser extent, the Peer-to-peer (meme) article). What's disputed is the assertion that there would be any advantage in switching to a navigational setup requiring all users (instead of just those users) to follow an additional link. —David Levy 05:17, 2 July 2009 (UTC)
Yes, it would. Unambiguously too, if we wanted it.   M   22:55, 2 July 2009 (UTC)
1. Please elaborate.
2. Please explain how the resultant data could possibly lead to the revelation that it makes sense to create a less convenient navigational structure for many users while leaving it the same for the rest. —David Levy 23:50, 2 July 2009 (UTC)
What do you mean by less convenient? A majority of people use the word "peer-to-peer" (frequently as an adjective) to find information on file sharing, in the same way that others use the word "panda bear" to find information on the giant panda. I too would like to see ambiguous and ill-formed words excised from English (not really), but that would be OR on our part. I give you stats that the majority of people are using the word "peer-to-peer" to find file sharing, exactly as is stated in WP:DAB, and you seem to be telling me that all these people are wrong in their use of English.   M   20:26, 3 July 2009 (UTC)
Where did I write anything about people being "wrong in their use of English"?!
Your proposed setup would be less convenient for users seeking the content presently located at Peer-to-peer (who would need to follow a link instead of immediately arriving at the desired article). Meanwhile, the change would in no way benefit users seeking the File sharing article, which already is linked at the top of the page. I'll try one more time to explain this to you:
If you still don't understand this argument, I can't help you. —David Levy 21:29, 3 July 2009 (UTC)
see below   M   00:43, 4 July 2009 (UTC)
  • Over the 6 hours (peak time) that the disambiguation page was active, peer-to-peer (network architecture) received 265 views (how many went to file sharing we can't tell - I regret not making the disambig page link a seldom-visited redirect). Peer-to-peer received the expected 2.8k over the whole day. To believe that file sharing received less than architecture's 265 views, you have to assume that this 6-hour peak window accounted for only 19% of the hits - that is, only 530 out of 2800 visited during those peak 6 hours, and exactly half went to each. This is improbable, so we can very safely assume that more than half the people went to file sharing (or to that other link).
    So, while we are arguing semantics about what peer-to-peer is (If there is extended discussion about which article truly is the primary topic, that may be a sign that there is in fact no primary topic, and that the disambiguation page should be located at the plain title with no "(disambiguation)".), the stats quite clearly show that when people type in "peer-to-peer", they more often mean file sharing, not network architecture. Is there still any valid objection to setting up that disambiguation page?   M   22:55, 2 July 2009 (UTC)
Why are you still attacking that straw man? Again, it's undisputed that many of the readers arriving at the Peer-to-peer page seek the File sharing article (and to a lesser extent, the Peer-to-peer (meme) article). What's disputed is the assertion that there would be any advantage in switching to a navigational structure requiring all users (instead of just those users) to follow an additional link. Assuming that your data analysis is 100% correct, it still has absolutely no bearing on that argument. When you ask whether there is "still any valid objection to setting up that disambiguation page," I can only conclude that you either don't comprehend this objection or are pretending not to. I'll assume the former. —David Levy 23:50, 2 July 2009 (UTC)
"Peer-to-peer" refers to two things: file sharing (usually as an adjective) and the network architecture. It's not that people are seeing "peer-to-peer", thinking "that reminds me, I'd like to look up file sharing", and clicking. Ask the average person what "peer-to-peer" means, and they'll tell you about file sharing. Under what conditions should "peer-to-peer" be a disambiguation page, or point to file sharing, for you?   M   20:16, 3 July 2009 (UTC)
A redirect to File sharing would be appropriate if the term "peer-to-peer" overwhelmingly referred to file sharing (to the degree that all other usage was negligible). This clearly is not the case.
Placing a disambiguation page at Peer-to-peer would be appropriate if there were too many primary meanings for a hatnote to reasonably accommodate. This clearly is not the case. —David Levy 21:29, 3 July 2009 (UTC)
In cases where the primary topic (the one that most readers want to see when clicking "go") is reasonably contested, the DAB page is placed at that page. If you disagree with this, I would like to know, but you would have to take your analysis (above) over to WT:D. Here's a bit more information for you: Wikipedia talk:Disambiguation/Archive 20#Primary Topic A Wikipedia talk:Disambiguation/Archive 20#Movement to get rid of the primary topic guideline B. There are some good reasons why a reader click count is not used to resolve this.
"When there is a well-known primary topic for an ambiguous term, name or phrase, much more used than any other topic covered in Wikipedia to which the same word(s) may also refer..."
The guideline is very clear on this, as are the discussions. My task here is not to argue with you about hatnotes, it's to establish that when the majority of readers use the words "peer-to-peer" in search, they want to see file sharing, not this article about a network architecture. If I can establish this, then will you agree that the DAB page should be placed here?   M   00:43, 4 July 2009 (UTC)
No, and your continued attack of this straw man leads me to believe that you still don't understand my argument (which I'm done trying to explain to you).
Guidelines are not sacrosanct laws to be blindly followed when doing so is unhelpful. When someone points out that there would be no advantage (and even a significant disadvantage) to doing something, responding that we should do it anyway "because the rules say so" (scare quotes) is a weak argument.
This particular situation differs from most because the two primary subjects are closely related and the one not occupying this title is correctly referred to by a different term. —David Levy 01:07, 4 July 2009 (UTC)
But guidelines are also not to be cast aside without careful consideration. What does the relation of the subjects have to do with it? If they are more related, there is a greater danger of readers not even noticing that they are at the wrong article. You're busy counting clicks - meanwhile, a large number of readers who are clearly looking for file sharing are winding up at a deceptively similar article.   M   01:26, 4 July 2009 (UTC)
1. When someone uses the term "peer-to-peer" in reference to file-sharing, he/she is referring specifically to file sharing performed over peer-to-peer networks (hence the commonness of the phrase "peer-to-peer file sharing," which obviously doesn't mean "file sharing file sharing"). Therefore, this article is directly relevant.
2. Are you seriously suggesting that a hatnote reading This article is about a method of networking and its implementation, not peer-to-peer file sharing. For file sharing performed over peer-to-peer networks, see File sharing. is insufficient (and somehow less helpful than the text "Peer-to-peer file sharing" on a disambiguation page)? —David Levy 01:53, 4 July 2009 (UTC)
Absolutely. A disambiguation page gives a clear warning that there are alternative meanings. A hatnote, on the other hand, is ignored unless something clearly strange is going on (especially a long one like that).   M   02:10, 4 July 2009 (UTC)
Okay, so you oppose hatnotes in general. The community disagrees. —David Levy 02:27, 4 July 2009 (UTC)
I've changed my mind. I did some google searches, and although peer-to-peer gets about 5x more hits, all the top hits in google seem to be using it in the correct sense, so I see little point in renaming this and introducing a redirect or disamb.- (User) Wolfkeeper (Talk) 21:45, 3 July 2009 (UTC)
This is why I was relying on DAB stats, not on google (your sample size is tiny, and representative of SEO value, not actual usage by readers). Here are some results from google news, though:
Did you try searching P2P, which redirects to this article? That term is probably a good candidate for having file sharing as its primary topic - nearly every single google, google news, etc. hit is about file sharing.   M   00:43, 4 July 2009 (UTC)
P2P doesn't redirect to this article. Dekimasuよ! 03:36, 4 July 2009 (UTC)
Disambiguates, though that doesn't seem to affect my point.   M   07:38, 4 July 2009 (UTC)
Since it doesn't redirect here, how it's used isn't really applicable to what to do with this article. Dekimasuよ! 11:29, 4 July 2009 (UTC)
  • Oppose the move - if "people" think something means something else because something else is currently more popular than something, that's no justification for changing the actual meaning of terms as used by reliable sources. LjL (talk) 23:24, 2 July 2009 (UTC)n
"The names of Wikipedia articles should be optimized for readers over editors, and for a general audience over specialists."[3] It's not currently more popular, it always has been. File sharing popularized peer-to-peer to the point that for the average "person", peer-to-peer simply means file sharing: p2p traffic, p2p groups, etc. Incidentally, is a political party a reliable source? ("Greens support an active vision of the Web as a platform for the exchange of information, with peer-to-peer groups in which each user can upload or download content and applications of choice.")[4]   M   20:16, 3 July 2009 (UTC)
That policy means nothing when it comes to a name conflict between articles, in which case, one reverts to using The Right Names. The most "easily recognized" name for the "peer-to-peer" typology is "peer-to-peer". Whether that term also happens to have been "hijacked" for another use (namely, file sharing) is irrelevant. LjL (talk) 01:19, 4 July 2009 (UTC)
The words "peer-to-peer" (and especially "P2P", which redirects to peer-to-peer) in the majority of cases refer to peer-to-peer file sharing, not peer-to-peer (the network architecture). What makes this not a name conflict?   M   01:31, 4 July 2009 (UTC)
Well, for a start the fact that when I Google for "peer-to-peer" or "p2p", the top hits I see are about the networking technology, not file sharing (or if they are file sharing, they qualify it as "peer-to-peer file sharing"). So, perhaps it's not the common name in anyone's books to begin with? I see most the votes here are "oppose", as well. LjL (talk) 01:42, 4 July 2009 (UTC)
When I google 'P2P', I get LimeWire, FrostWire, live sports, isoHunt,, Kazaa, and OpenOffice P2P. Of these, live sports is irrelevant, OpenP2P is about networks, but the other 5 all treat P2P in the context of sharing files. These are not very good statistics, though. Why are you relying on google when we have some concrete and significant numbers for what people actually want to see when they arrive at this article?   M   02:10, 4 July 2009 (UTC)
P2P remains unrelated to this discussion. At any rate, numbers aren't king, nor are the numbers you present conclusive. And we also take into account things like recentism, systemic bias, and, as has been noted elsewhere, what will cause the least surprise. Dekimasuよ! 11:35, 4 July 2009 (UTC)
File sharing has its own common name. This isn't a situation in which both entities have only one common name between them.
Additionally, "peer-to-peer" doesn't refer to file sharing in general; when it's used in reference to file sharing, it refers specifically to file sharing performed over peer-to-peer networks. That connotation, therefore, relies on the usage currently occupying the Peer-to-peer title. —David Levy 01:53, 4 July 2009 (UTC)
It doesn't matter if a topic has its own name already, if another name is used more often to refer to that topic, it should point to it. "Peer to peer file sharing" refers to a type of file sharing, not a type of peer-to-peer architecture. In most cases, the use of peer-to-peer is dubious - something like saying "Firefox is a client-server web browser".   M   02:10, 4 July 2009 (UTC)
1. It does matter that file sharing has its own name; this means that we needn't append parenthetical disambiguation to either (or both) of the primary meanings of "peer-to-peer." We've simply assigned that term title to the primary subject commonly known only by that name (and not to a subject with its own common name encompassing the entire topic instead of merely a portion thereof).
2. Yes, "Peer-to-peer file sharing" refers to a type of file sharing...performed via peer-to-peer architecture. Both articles are relevant. —David Levy 02:27, 4 July 2009 (UTC)
I disagree that we should "assign" the term to the one of two primary subjects without its own name. Reader expectations assign the term for us, and when the term is assigned to two primary topics, we end up with a DAB.   M   07:38, 4 July 2009 (UTC)
Why are you still acting as though these are two entirely unrelated topics? Again, when someone uses the term "peer-to-peer" in reference to file sharing, he/she is referring to file sharing performed via a peer-to-peer network. This is not an irrelevant article, even if it isn't the one sought. The hatnote is every bit as helpful (if not more so) to such an individual as a disambiguation page link would be, and he/she might even be interested in reading about the underlying technology behind peer-to-peer file sharing. (If not, there's no harm done.) Meanwhile, users seeking this article needn't follow a link; it's right here under the technology's name. It's a win-win situation that you evidently seek to eliminate because you believe that "the rules say so" (scare quotes), because of your general opposition to hatnotes, and because of an utterly uncorroborated theory that the current setup somehow causes confusion. —David Levy 08:01, 4 July 2009 (UTC)
When the term "peer-to-peer" is used in the context of file sharing, it refers to file sharing performed via a peer-to-peer network. It's not the rules. I don't know how familiar you are with the file sharing and peer-to-peer articles, but they are (were?) a bloody mess in no small part due to nobody having straight the difference between the two concepts. And if readers really are finding their way to the right article, then why is peer-to-peer seeing abnormally many hits in contrast to file sharing, very much unlike client-server? Why would the majority seem to be going to file sharing, when that DAB page was set up? I don't actually know - but I do know how we could find out. Unfortunately, you seem dead-set against using a DAB page as a tool for determining what readers are looking for, because it would result in a few extra clicks.   M   08:23, 4 July 2009 (UTC)
1. Yes, when the term "peer-to-peer" is used in the context of file sharing, it refers to file sharing performed via a peer-to-peer network. Both concepts are directly relevant.
2. Again, there is a great deal of overlap between the two subjects. In such a situation, it's common for information to be placed in the less appropriate of the two articles.
3. Peer-to-peer receives a disproportionate number of hits because many of the readers arriving there seek the File sharing article (which is clearly, conspicuously and conveniently linked at the top of the page). Why do you insist on conducting an experiment for the purpose of proving something that is uncontested? —David Levy 08:53, 4 July 2009 (UTC)
  • Support - I agree with JHunterJ that there is no primary topic here and Peer-to-peer should be a disambiguation page, so make way for it by moving this article. I am surprised no one has yet mentioned (that I noticed) peer-to-peer counseling. --Una Smith (talk) 21:20, 5 July 2009 (UTC)
Where did JHunterJ state "that there is no primary topic here"? JHunterJ noted below that he/she hasn't read the discussion, but after examining the tally, "it looks like there may be consensus to retain the current primary topic at the base name." I haven't seen JHunterJ express any personal viewpoint on the matter. —David Levy 21:31, 5 July 2009 (UTC)
  • Oppose - This is the primary topic. Even when the term in question is used in reference to file sharing, it's file sharing over a peer-to-peer network—this article's subject. I find the practical arguments utterly unconvincing and am confident that readers would only be hindered by the proposed changes. Anyone who comes to this article purely looking for information about file sharing itself can click the highly-visible link right atop the page. Why should we make it so the people who want the article about peer-to-peer technology have to click an additional link? -CapitalQ (talk) 05:09, 6 July 2009 (UTC)

Input from dab project editor

I'm posting this here too, after initially posting it in response to a question at Wikipedia talk:WikiProject Disambiguation. I haven't read the long Requested move section above, only looked at the !votes. I think you have the tools needed (stats, Google, incoming links). If the editors at Talk:Peer-to-peer cannot come to a consensus on what the primary topic is, then that is a good indication that there is no primary topic. If there is no primary topic, then the page at Peer-to-peer becomes a disambiguation page, regardless of what the current pages and hatnotes are. But looking at the discussion on the requested move, it looks like there may be consensus to retain the current primary topic at the base name. If that is done, then hatnotes can be used to direct other readers to the file-sharing article. -- JHunterJ (talk) 19:45, 5 July 2009 (UTC)

There's no consensus to do that either.- (User) Wolfkeeper (Talk) 19:57, 5 July 2009 (UTC)
Consensus ≠ unanimity —David Levy 20:21, 5 July 2009 (UTC)
Right. WP:NOTUNANIMITY -- JHunterJ (talk) 20:26, 5 July 2009 (UTC)
So your argument, if I've understood this correctly, to create a disambiguation page, you want to do it, and you've proven that there's no consensus to keep the existing structure because you want to do it, and according to the policy that's all you need? ;-) Borat says: 'Nice'. Wolfkeeper says: if you're cynically invoking policy in this way I invoke WP:IAR.- (User) Wolfkeeper (Talk) 00:18, 6 July 2009 (UTC)
All kidding aside, I'm convinced a careful examination of the reliable sources does not support the idea that 'peer to peer' is a correct term for 'file sharing'. I do not subscribe to the wikiality view that you simply need consensus in the wikipedia; you're supposed to be generating consensus about what the neutral point of view is not just arguing about what you believe in your heart to be true or what you heard somewhere.- (User) Wolfkeeper (Talk) 00:18, 6 July 2009 (UTC)
JHunterJ stated that "looking at the discussion on the requested move, it looks like there may be consensus to retain the current primary topic at the base name" with "hatnotes ... used to direct other readers to the file-sharing article." This is the setup that you prefer, yes? —David Levy 00:29, 6 July 2009 (UTC)
JHunterJ also said "If the editors at Talk:Peer-to-peer cannot come to a consensus on what the primary topic is, then that is a good indication that there is no primary topic. If there is no primary topic, then the page at Peer-to-peer becomes a disambiguation page, regardless of what the current pages and hatnotes are." I don't see consensus here. --Una Smith (talk) 01:42, 6 July 2009 (UTC)
The discussion is not complete, but please elaborate. —David Levy 02:21, 6 July 2009 (UTC)
I count 3 support, 5 oppose. Of those 5 oppositions, 2 of those argue that a hatlink is sufficient (it isn't). Of those two, David Levy has made it extremely clear that he believes that a significant portion of readers arriving at peer-to-peer are actually looking for file sharing. A few oppositions are based on keeping the 'peer-to-peer' label correctly referring to the architecture, which is simply an invalid reason, as far as DAB goes. There are also the results of the stats tool to consider, which showed that very few people were actually choosing to go to this architecture page, in those 6 hours in which the page was available. And also the matter of the self-selection of voters - Elen of the Roads said "Oh, and if it's any help, I thought p2p and filesharing were the same thing... You can mark me down as one of the ignorati :)" over at WQA, but you don't see this being taken into account. Consensus also isn't voting, so I disagree that there's anything of the sort forming for keeping this page here.   M   06:39, 6 July 2009 (UTC)
1. There are six users opposing (though as you correctly point out, this isn't a majority vote).
2. It's your opinion that a hatnote is insufficient. You're arguing that we're wrong because we're wrong.
3. Yes, I have made it extremely clear that I believe that a significant portion of readers arriving at Peer-to-peer are actually looking for File sharing. I've also explained why I believe that the current setup is the best method of getting them there.
Your opposition is to hatnotes in general, which always are intended to lead to articles sought by a significant portion of readers arriving at the page. (Otherwise, what purpose would they serve?)
4. You're correct that it isn't our place to dictate common usage, but both common usages directly relate to the "correct" subject documented in this article. Therefore, that's a valid rationale.
5. I disagree with your description of the statistics and your assessment of their ramifications (even if your interpretation of the data is 100% accurate).
6. Elen of the Roads quickly realized that she was mistaken, which means that the current setup did not result in confusion. However, I won't attempt to count this as a de facto vote for the status quo. (Elen clearly hasn't expressed a preference either way.) —David Levy 07:29, 6 July 2009 (UTC)
It is not merely my opinion that a hatnote is insufficient, it is long-established consensus. If you would like to know why even a two-topic dab is more appropriate than a hat note, please ask at WT:D. Consensus among a small group of editors cannot override the community consensus found in WP:D. You have very clearly stated that you think there are 'two primary topics', your task now is to show why this article should be an exception (you've mentioned click counts, similarity of articles, neither of which have anything close to wide-community consensus support). The stats tool showed that only 1/10th of peer-to-peer's usual daily 2800 chose 'architecture' during the 6 hours that the disambiguation page was active. Make your objections to the stats tool known, or accept the data. It matters what readers want, and when 'asked' what topic they were interested in when looking for 'peer-to-peer', the architecture topic was clearly in the minority.   M   09:09, 6 July 2009 (UTC)
1. It's long-established consensus that hatnotes are insufficient? Then why haven't they been abolished?
Note that I'm addressing your assertion that a hatnote fails to supply the pertinent information.
2. No one seeks to override consensus. Once again, you're portraying this as a binary situation in which the guideline must either followed to the letter or completely ignored. This is not so, and we needn't modify a rule simply to abide by its spirit.
3. Yes, there are two primary topics in the sense that there are two likely search targets for the term "peer-to-peer." This, however, does not take into account the fact that both connotations directly relate to this article. This isn't a situation in which readers are being dropped into an irrelevant article.
4. How are you arriving at the "1/10th" figure? Peer-to-peer's 2,800 daily views divided by the 265 views received by Peer-to-peer (network architecture) during a much shorter period?
5. Again, that data is irrelevant to whether a hatnote or disambiguation page would better serve the users arriving at Peer-to-peer in search of information about file sharing. I've explained why I regard a hatnote as more helpful to everyone. —David Levy 17:13, 6 July 2009 (UTC)
When is the last time you all looked at Peer-to-peer (disambiguation)? It looks to me like a lot to disambiguate. Let's accept for the moment that all peer-to-peer computing is in some way related; the fact remains that there are multiple articles about peer-to-peer computing and they all need to be disambiguated so we don't have what now exists: "peer-to-peer network" and "peer-to-peer application" and "peer-to-peer file sharing", etc. See the problem? Here is an actual example: Flooding algorithm(version) refers to "peer-to-peer file sharing systems"; it appears this should be changed to "peer-to-peer file sharing systems" (and file sharing should be expanded to include a section on peer-to-peer file sharing). Usages like this involve links that need to be disambiguated, changed to link to one of the articles listed on Peer-to-peer (disambiguation). Has anyone examined the incoming links to Peer-to-peer? —Preceding unsigned comment added by Una Smith (talkcontribs) 20:31, 6 July 2009 (UTC)
Note that the disambiguation page was radically expanded less than two hours ago (before which it listed only this article and the two others linked in the hatnote). Accordingly, I was going to add Peer-to-peer (disambiguation) to the hatnote, but Kbrose beat me to it. (Such a link is suitable, as all of the other connotations are secondary to peer-to-peer networking and peer-to-peer file sharing.)
You're quite correct in noting that there must be incoming links in need of repair. —David Levy 21:02, 6 July 2009 (UTC)
It is consensus that when there is no one primary topic, a dab page is used. There are reasons a hatnote is not used. Have you asked what they are at the appropriate venue, as I requested? If there is no agreement here as to the primary topic, then a dab page is placed at peer-to-peer, and not a hatnote. If you would like to argue this, please take it to WT:D, and attempt to gather a widespread community consensus for making an exception here.   M   21:59, 6 July 2009 (UTC)
Is that how you believe our guidelines work? That we have to go to their talk pages to seek consensus for each and every exception? Wow. —David Levy 22:18, 6 July 2009 (UTC)
You can do it here if you'd like, but you might have a harder time attracting enough editors to overturn the community consensus, and it is often best to take such discussions to a neutral forum to avoid systemic bias. Have you asked about the reasons for not using hatnotes at WT:D yet? The information you gain there might avoid a dispute here.   M   22:30, 6 July 2009 (UTC)
Again, no one seeks to "overturn the community consensus." I'm familiar with the reasons in question, and I support the guideline (including the tag explaining that it "it is best treated with common sense and the occasional exception"). —David Levy 22:41, 6 July 2009 (UTC)
What you seek to do is clearly not 'common sense', and occasional exceptions require serious justification, especially when other editors oppose them. What are the reasons, then, and why does this article merit an exception to them? If you've explained this sufficiently well before, consider linking a diff or copy-pasting.   M   22:51, 6 July 2009 (UTC)
I'm not going to bother diff-linking or copying and pasting explanations that I've written so many times. Re-read the move discussion if you need to. —David Levy 23:04, 6 July 2009 (UTC)

It seems to me that the big problem here is the content of Peer-to-peer. Although the lead begins "Peer-to-peer (P2P) networking" (version), in fact the content is more about peer-to-peer computing more generally. That seems to be at the heart of the argument between David Levy and M: one editor sees the article as covering it all, and the other sees it as covering only networks as distinct from other computing stuff (ie, as a kind of network architecture). David Levy is correct in that all of these are subclasses of the class "peer-to-peer computing". However, M is correct that our goal here is not classification, it is disambiguation. David Levy asserts that the article Peer-to-peer is the primary topic of "peer-to-peer", but has provided no evidence of this. I agree with M that there is no primary topic and the disambguation page should occupy the ambiguous base name Peer-to-peer. I also think the article now named Peer-to-peer should be moved to Peer-to-peer computing, and rewritten to cover its history and link in context most of the articles on the dab page. (By the way, the first peer-to-peer file sharing systems were not constructed on peer-to-peer network architectures.) --Una Smith (talk) 03:32, 7 July 2009 (UTC)

To clarify, I don't assert that this is the primary topic (as defined at WP:D). I regard it as a primary topic (linguistically in the lexicographic sense of the term "primary topic") and one to which the other most common use of the term (peer-to-peer file sharing) is directly related. I therefore believe that it can be considered a de facto primary topic for navigational purposes, but I fully acknowledge that this is an exception to the guideline's usual application. (It's one that I view as sensible, but it's perfectly reasonable for others to disagree.) —David Levy 03:52, 7 July 2009 (UTC) [ambiguous text struck/replaced 06:20, 7 July 2009 (UTC)]
I think your suggestions are good, though the dispute seems to revolve around whether a hatnote or a dab page is appropriate - though it's hard to tell, since David Levy's position is confusing. I'll summarize, but a warning- it's long and not that pretty.
He states that this is not the dab-policy primary topic, but "linguistically" (lexicographically) it is the "primary topic" (it is prescriptively correct). So while he agrees that policy states otherwise, he thinks an exception should be made on the basis of this prescriptive evaluation. (One editor agrees, stating that we 'shouldn't change the meaning of terms as given by reliable sources'.) This is, however, simply not a legitimate reason to place a topic at some name, as David himself has stated just a few messages above ("isn't our place to dictate common usage"). He explains this contradiction by saying that, all things being equal (wp:d says they are not), this does tip the scales towards making architecture the default target of peer-to-peer. I think the other idea is that due to click counts, a hatnote is a better idea (but then, this applies to any article, so isn't really a case for an exception, but rather an overturning of the guideline). And, that article similarity is a good thing, because it reduces surprise, and because readers are probably interested in reading related topics. Further, similarity does not reduce confusion, what reduces confusion is the hatnote, which is the reason that hatnotes are less confusing than dab pages. As far as I can tell, this is actually a rather generous summary of his position.
My position is that there are a number of sound reasons to believe that a significant number of readers who come to peer-to-peer seek file sharing and therefore there is no primary topic (David agrees here), and that the guideline makes our next step clear: make this a DAB page. The reasons include common sense (most people don't know or care about architectures), the stats tool (see above; few readers sought the architecture), google news (see above; it seems that the meme is a popular buzzword), and google (p2p gets the vast majority of hits).
And that's the summary of the above few pages.   M   04:45, 7 July 2009 (UTC)
I'm sorry that you find my position confusing, and I sincerely appreciate that you've taken the time to attempt to understand my points. You've correctly summarized some and incorrectly summarized others, so I'll try again to clarify.
I do not regard this as the only "primary topic" (in the lexicographic sense of the term "primary topic"), but I do regard it as one of two (the other being peer-to-peer file sharing). My argument is not based on the term's prescriptive correctness. (My reference to linguistic meaning referred to the term "primary topic," not to the term "peer-to-peer," though I now realize that this was unclear.) In my assessment, part of what makes this a sensible location for the article is that the usage documented therein is highly relevant to both usages (whereas the topic of file sharing is relevant only to one). In other words, if someone seeks information about peer-to-peer file sharing and arrives at Peer-to-peer, there is a significant likelihood that this article also will be of interest to him/her (as it provides directly relevant information about the technology through which peer-to-peer file sharing is performed).
However, if the current navigational structure somehow impeded these users' ability to access the File sharing article, that would more than negate this benefit. But because the hatnote clearly and conspicuously directs these users there (regardless of whether the Peer-to-peer article also interests them), there is no navigational/disambiguational disadvantage. (To be clear, I'm not asserting that a disambiguation page would be more confusing than the hatnote is; I'm saying that both methods work equally well in this respect.)
Meanwhile, there is a navigational advantage for readers seeking the Peer-to-peer article, who needn't follow any links to arrive there. This would not outweigh any added confusion/inconvenience experienced by readers seeking the File sharing article, but it's my opinion that no such added confusion/inconvenience exists (as explained above).
Therefore, in my assessment, among readers seeking either Peer-to-peer or File sharing (which I believe we all agree are the two most common usages by far), every effect of using a hatnote instead of placing the disambiguation page here is either positive or neutral. This means that there's a net benefit.
I acknowledge that this does not strictly comply with the disambiguation guideline, but I absolutely do not contest that guideline's validity. I regard this situation as a rare exception warranting a minor deviation from the usual practice. In the five years since I discovered Wikipedia, I don't recall encountering another situation quite like this one, so a special revision to the guideline certainly isn't called for. Guidelines are general rules of thumb based on what typically makes sense most of the time, and they are not intended to cover every possible scenario that might arise. This is why we are formally advised that a guideline "should be treated with common sense and the occasional exception." Wikipedia policy is to follow a rule's spirit (not its letter), and the spirit of the disambiguation guideline is that we configure our navigational structure in the manner that best aids readers in accessing content. It's my position that the current setup accomplishes that.
I'm not claiming that this is self-evidently correct. I'm expressing my viewpoint in an attempt to establish consensus for such an exception (and a lack of consensus for the alternative). —David Levy 06:20, 7 July 2009 (UTC)
I see your point about article similarity. So when Joe becomes interested in file sharing, but erroneously searches for 'peer-to-peer', he'll wind up at an article with at least a subset of the information he was looking for right there. My problem with this is that Joe was looking for file sharing, and we have such an article, but our first "aha, I think I've got just the thing you're looking for!" is... a highly technical article that talks about clients, UUCP, processing power. We know full well that the term is ambiguous for technologically-ignorant louts like Joe, so "well, do you want the Napster/Kazaa/torrent thing, or the network thing?" should be our first response. This is a matter of convention and consistency.
Because it's consistent, readers often take a lack of disambiguation page to mean "yes, when people say 'topic', they pretty much always mean this - you're at the right place". When looking things up, do you read hatnotes? I ignore them entirely, unless I start reading the article and notice something funny. Clicking a link in a dab page is much faster than reading the first sentence of this article, skimming halfway down, (in our case, possibly) figuring out that the article is missing the stuff you're looking for, scrolling down to 'see also', possibly remembering to look for hatnotes, reading the hatnote, and clicking. Reading and clicking the dab link is much faster than just reading the hatnote! These, and many others, are the reasons we place dab pages, not hatnotes, at pages without a primary topic. Note that all of these reasons are generic.
An exception must be based on something exceptional (different), and the only difference you mention is atypical similarity. So "the two articles share content, therefore readers will have found part of what they are looking for, ???, therefore the hatnote will be more effective than it usually is". The ??? is the place where I don't follow your argument, though I'm glad to see that your premise is something specific to this case (otherwise, as with eg click counts, you would be arguing against the guideline as a whole, since those apply to every such article). Hopefully the reasons above do a better job of fleshing out why I don't like hatnotes in this context.   M   16:43, 7 July 2009 (UTC)
Likewise, I see your point, but I strongly disagree that this is how readers typically behave. We don't want users seeking secondary topics to be confused/inconvenienced either, so if hatnotes usually were ignored, they would never be an effective means of disambiguation. If that were the case, we always would use a disambiguation page (even when a clear primary topic existed). My position is that a properly implemented hatnote (which I regard this one to be) is an equally effective disambiguation tool.
My point isn't that this article is close enough to what Joe seeks (with some overlapping information). It's that it provides additional information of direct relevance to the topic of peer-to-peer file sharing (file sharing via a peer-to-peer network). Here are the scenarios that I foresee:
  • Joe seeks the article about peer-to-peer networking (and might or might not be interested in peer-to-peer file sharing). Joe types "peer-to-peer" and immediately reaches the desired article.
  • Joe is interested in peer-to-peer file sharing and wishes to read about the overall concept (encompassing the file sharing itself and the type of network through which it occurs). Joe types "peer-to-peer" and arrives at one of the two relevant articles (with a clear, conspicuous link to the other).
  • Joe specifically seeks information about file sharing itself (not the type of network through which it occurs) and types "peer-to-peer." Upon arriving at this article, he is immediately informed that it is not about peer-to-peer file sharing and linked to File sharing. Joe might then be interested to read this article first (especially if he's just realized that there's more to the overall subject than he was aware of) or might proceed directly to File sharing (just as he would upon reaching a disambiguation page). —David Levy 20:56, 7 July 2009 (UTC)
When you make your case, it must be based on some exceptional/special circumstance with this article. If it weren't, then it could be used to knock down WP:D for any article, and you certainly don't claim that your argument is strong enough to do that. (I could say "Joe is searching for usefulness, and arrives; or for the generic idea of good, and finds part of the information; or for altruism, but then it's just one click away! Therefore, usefulness should be placed at good, with a hatnote to altruism and the dab.") Aside from similarity (which applies to many though not all dab pages), you mention that what makes this article exceptional is the very clear and detailed hatnote - but this too can't possibly be your position. So, if it's neither behavior (common to all dab), similarity (common to most), nor a detailed hatnote (insufficient), then what makes this particular article an exception to WP:DAB?   M   21:57, 7 July 2009 (UTC)
In my assessment, the exceptional circumstance is the fact that both articles are directly relevant to peer-to-peer file sharing (not merely similar).
I'm not citing the very clear and detailed hatnote as an exception; I'm merely noting that it isn't a substandard disambiguation tool. —David Levy 22:23, 7 July 2009 (UTC)
I don't understand your argument. Are you saying that (all?) readers are seeking the term "peer-to-peer file sharing", and both articles are relevant, therefore the hatnote belongs on only one of them? This seems to be an argument for putting a DAB page at peer-to-peer file sharing. If [peer-to-peer file sharing] really were a topic, and your estimate of reader interests is correct, then it too should be added to the DAB page. If they are both related, then this seems a case for a dab page where both are listed - what exceptional circumstance provokes us to choose one over the other? Your premise (an exceptional circumstance with our situation) seems disconnected from your conclusion (that a hatnote is more appropriate) - unless your arguments above are a general objection to dab pages for related articles...   M   23:23, 7 July 2009 (UTC)
I'm sorry that you don't understand my argument. The above does not accurately describe my views, which I've attempted to make as clear as possible. I honestly don't know how to better express them. I'm truly sorry. —David Levy 23:30, 7 July 2009 (UTC)
Excuse me, but how is not changing a term's meaning as given by WP:reliable sources "simply not a legitimate reason to place a topic at some name"?
WP:NAME says that "Wikipedia determines the recognizability of a name by seeing what verifiable reliable sources in English call the subject".
I don't believe the WP:Common name is necessarily what an ill-informed reader might look for, even if most readers are ill-informed somehow, and it looks, to me, like guidelines agree. Sure, "the names of Wikipedia articles should be optimized [...] for a general audience over specialists", but optimized doesn't mean "changed without regard to anything else" - it means, in this case, "made as usable to a general audience as possible while making sure that Wikipedia remains an encyclopedia".
--LjL (talk) 13:57, 7 July 2009 (UTC)
When a word refers to two topics, you can't place both at the same location, so we use disambiguating brackets. The name is what comes before disambiguating brackets. So [peer-to-peer], [peer-to-peer (meme)], and [peer-to-peer (racehorse)] all have the same, correct, name. If the suggestion was to move the article to [peer-to-peer system computing architecture], then you would be absolutely correct. I am almost entirely neutral as to the destination name, though.   M   16:06, 7 July 2009 (UTC)
With all due respect, although peer to peer is part of computing-as is networking in its very widest sense but the term peer to peer computing does not usually refer to all types of peer to peer, but only to the subset of peer to peer networks where the networks are created for the purposes of creating large parallel processing networks as opposed to other systems that require little processing like file sharing. So the idea that this article should be renamed to that is simply erroneous.- (User) Wolfkeeper (Talk) 15:31, 7 July 2009 (UTC)
Article naming is mostly to do what things are commonly called. The article as it currently stands is at the article name that it is mostly commonly referred to already.- (User) Wolfkeeper (Talk) 15:31, 7 July 2009 (UTC)

This move request has been posted for a week and it appears that the consensus tally clearly indicates that the majority of editors oppose the move request. It is therefor time to close it and leave the article title as it stands. Kbrose (talk) 16:36, 7 July 2009 (UTC)

There's no such thing as a 'consensus tally'. Discussion is ongoing. Incidentally, if we ended it now, without arriving at a consensus that this article is the primary topic of 'peer-to-peer', the most appropriate action would be to set up the dab page. You would no doubt revert this, so the discussion would have to resume.   M   16:51, 7 July 2009 (UTC)
There need not be consensus that this article's subject is the primary topic (as defined at WP:D). If there is consensus that the proposed move would not be beneficial, it is not carried out. We don't follow a rule when doing so is unhelpful. —David Levy 20:56, 7 July 2009 (UTC)
Sure, but in that case, you need consensus that an exception should be made. (Not to say that there's consensus just looking at the numbers.)   M   21:57, 7 July 2009 (UTC)
Agreed. I believe that a thorough analysis of the discussion demonstrates such consensus, but obviously, neither of us is impartial. —David Levy 22:23/22:26, 7 July 2009 (UTC)
As M pointed out, one cannot gauge consensus via a numerical tally. I agree that there is consensus against the move, but both you and I are far from impartial. Therefore, I've reverted your closure. Please allow an uninvolved editor to evaluate and close the discussion when deemed appropriate. —David Levy 20:56, 7 July 2009 (UTC)
Oddly enough, I was seeing a lack of consensus, and therefore grounds to default to placing a DAB page here.   M  
Indeed, but neither of us is impartial (obviously). —David Levy 23:30, 7 July 2009 (UTC)

Requested move 2

Peer-to-peerPeer-to-peer (computing) — Also Peer-to-peer (disambiguation)Peer-to-peer. The outcome of a prior move request, now closed, was that (a) this article's scope is limited to the computing domain and does not include Peer-to-peer (meme) etc., and (b) this article's incoming links need to be disambiguated. --Una Smith (talk) 20:40, 7 July 2009 (UTC)

This new request appears substantially identical to the previous one. Regardless, I've reopened that discussion (which should not have been closed by an involved editor). —David Levy 20:56, 7 July 2009 (UTC)
It is not substantially identical to the previous one. However, I did not realize until after the fact that the "closure" was by an involved editor. I'll ask WP:RM to look in. --Una Smith (talk) 21:46, 7 July 2009 (UTC)
How does being a contributor to the article and voter on the request not constitute involvement? Kbrose (talk) 21:49, 7 July 2009 (UTC)
It does, which is precisely the problem.   M   22:06, 7 July 2009 (UTC)
Exactly. —David Levy 22:23, 7 July 2009 (UTC)

Discussion has [not] been closed

Consensus was achieved, and the result was OPPOSED to changing the articles title. It properly reflects the major meaning of the term. Article should be expanded if lacking, not renamed. Any further discussion can be conducted on people's talk pages. It doesn't seem to change the outcome here. A week is long enough. Kbrose (talk) 21:44, 7 July 2009 (UTC)

Kbrose's "closure" is invalid. --Una Smith (talk) 22:00, 7 July 2009 (UTC)
Indeed. —David Levy 22:23, 7 July 2009 (UTC)
Consensus has not been achieved, stop interfering. If you continue to close and "evaluate" discussions in a biased matter, you'll end up blocked.   M   22:03, 7 July 2009 (UTC)
It only takes a simple count to determine consensus. A count is not biased and should be taken after a 'few days', which certainly have elapsed and your prolonging of this endless debate does not change the outcome. You threats are ridiculous. Kbrose (talk) 22:15, 7 July 2009 (UTC)
Your assertion that "it only takes a simple count to determine consensus" is completely incorrect. Please see Wikipedia:Consensus.
And keep in mind that I'm on the same side as you in the debate, so I'm obviously not objecting to your closure because of any bias on my part. —David Levy 22:23, 7 July 2009 (UTC)
The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.