Wikipedia talk:Requests for mediation/Jnc and Barberio

Mediation

Since this is a content dispute, I assume both of you are happy to keep this discussion here on the wiki. If either would prefer, however, we can move this to email. I'd also like to ask both of you to avoid editing contentious sections of the article for the time being, and to keep the discussions on this page. Please note also that I am not here to agree or disagree with any specific fact related to the article — the purpose of this is to see if you two can agree with each other on any of the points. The best way to prove or disprove any specific fact will be through appropriate citations.

There are a number of debates that seem to have sprung up in the edits of History of the Internet, the talk page and related archives. In the interests of fire fighting I'd like to first collect the specific disagreements here in one place. These should be disputes related to the current version (approximate) of the article: this mediation is not about past disagreements. This should also help satisfy John's desire to see Noel's objections written out formally.

In the following section, please list your disagreements with current version as signed, individual, concise points. The other party can then respond to any point with their own, concise rebuttal. Please do not turn this section into a discussion, however: we can leave that for later.

— Asbestos | Talk (RFC) 19:49, 3 October 2005 (UTC)[reply]

Next week I'll be heading to Ireland for my grandmothers 80th birthday, so will be away from Internet connections. I hope we can have this wraped up, or at least progress a little before then. --John R. Barberio ^{talk, contribs} 00:33, 6 October 2005 (UTC)[reply]

Please note that I will be refactoring this page as appropriate to try to separate out the different points. However, this is your discussion and if you think the refactorings make anything confusing, feel free to revert me. — Asbestos | Talk (RFC) 23:59, 7 October 2005 (UTC)[reply]

Disputes (still outstanding)

Only told the American Internet's history. No mention of how the Internet was spread internationaly. Same cites referencing X.25/IPSS, since X.25/IPSS was the underlying technology used to spread international internetworking. --John R. Barberio ^{talk, contribs} 23:52, 3 October 2005 (UTC)[reply]
"The Internet" as a phrase refering to a single global network did not arise till the mid 80s, so it seems inapropriate to refer to things prior to mid-expansion NSFNet as 'early versions of The Internet' or similar. Phrases 'an internet', or refering to example networks as 'the internet' in texts confuses this, but these seem to be refering to individual examples of a type of network, not one specific network. It's dificult to find specific published etymology for this, but Tanenbaum places origin of capital I Internet being used as a specific idntification of the network with NSFNet development in the mid 80s. (Andrew S. Tanenbaum (1996) Computer Networks) --John R. Barberio ^{talk, contribs} 23:52, 3 October 2005 (UTC)[reply]
Example of phrase useage cites on the etymology issue, requested by Nixdorf.
Examining TCP-IP Digest, Vol 1 #10, handly archived by Google [1], Specificaly, their quote of Computer World, who were quoting earlier information from the digests, "Considered the world's first packet network, Arpanet is expected to become an internet -- a network of networks -- .. said an informed source, who revealed the cutover date." which indicates that common useage circa late 1982 was to still refer to the Arpanet TCP/IP network as 'an internet', not 'The Internet'. (A handy two in one cite, showing the useage both by the specialists in study, and the trade press) Xerox's naming of a subprotocol in their network suite as 'IDP - Internet Datagram Protocol' also indicates that the early TCP/IP developers had no monopoly on the word, but TCP/IP would be the most common item people would refer to as 'The Internet Protocols', which would later result in the TCP/IP network being refered to as The Internet. (See Xerox Network Services). Combined with my cite from Tanenbaum, I feel this is at least sufficient to put some doubt on ARPANET being the prime source of the phrase. --John R. Barberio ^{talk, contribs} 19:32, 4 October 2005 (UTC)[reply]
The article was not very readable. The article was confusing to non-technical readers, mixing a skeleton of a technical history in with a limited discusion social aspects and a confused sence of who did what. This gave it all a muddled feel. --John R. Barberio ^{talk, contribs} 23:52, 3 October 2005 (UTC)[reply]

I didn't have any major problems with the article as it existed; it wasn't great, but it was vaguely adequate, and - most importantly - included "Further readings" which were i) easy to obtain, and ii) contained good histories in iii) more detail than we could put in the article anyway. So, the article was not a focus for me: I was more interested in putting material in the encyclopaedia on other topics, ones which are much harder to locate material on (e.g. Pilot ACE). I will review Mr. Barberio's comments and reply once I have studied them. Noel (talk) 13:59, 7 October 2005 (UTC)[reply]

Disputes (under discussion)

Licklider and the creation of ARPANET

My major objection with the 'old version' is that it lacked substantial information, and gave incorect impresions about the origin of the Internet. One specific and glaring error was the implication that Licklider initiated the creation of ARPANET, which was infact initiated by Robert W. Taylor. ([2] NYTimes Article on Taylor) --John R. Barberio ^{talk, contribs} 23:52, 3 October 2005 (UTC)[reply]

Let me start with what I will label 1b - "the implication that Licklider initiated the creation of ARPANET, which was infact initiated by Robert W. Taylor." It all depends on what you mean by "initiated".

The standard academic history of DARPA, Norberg/O'Neill Transforming Computer Technology starts out chapter 4 (The Development of Packet-Switching Computer Networks) by discussing Lick. (I should further point out that this volume is in itself a reworking of their earlier A History of the [IPTO] at the [DARPA], DARPA's own history of its operations, so it's pretty much the official history of DARPA.)

Abbate's book Inventing the Internet (the current standard academic history of the Internet) likewise starts its chapter on the ARPANET (2, Building the ARPANET) with Licklider: "the inspiration can be traced back to J[] C. R. Licklider".

In any event, a detailed discussion of the division of credit for the ARPANET belongs in the article on that, not here; Licklider is mentioned in this article because of his early papers, in particular Man-Computer Symbiosis, and his Intergalactic Network memo of April, 1963, which are now credited as the root of what eventually became the Internet, e.g. in Waldrop, Dream Machine, pp. 5 : "such visions would inspire [Licklider]'s hand-picked successors to implement his Intergalactic Network ... into the network of networks known today as the Internet."

I fail to see why Wikipedia should contradict these canonical works on the subject. Noel (talk) 14:51, 7 October 2005 (UTC)[reply]

PS: All the works on the subject use the definite article - the ARPANET. I am at a loss to understand why Mr. Barberio stubbornly insists on leaving it out, no matter how many times this is pointed out to him. Noel (talk) 14:51, 7 October 2005 (UTC)[reply]

Noel, please read the article as it is now. It does not give Licklider credit for his papers, and sugestions. It currently gives him credit for initiating ARPANET. Please do not muddle things by responding to what you preceive the complaint was.

If you re-read my rewrite, you will find I gave Licklider full credit for inspiring Taylor, adressing your objection that he should be mentioned as an inspiration towards the Internet. (Albiet, I might argue that Vannevar Bush sugested some of the things Licklider suggested too.) However, noting that Taylor had the direct influence behind the creation of the internet is not contridicting Waldrop's assertion that Licklider inspired Taylor.

Incidently, there is no such thing as a 'standard academic history'. History texts are not like technical standards, there is no canonical 'history' of anything. So we must go by a concensus based on various secondary sources, primary sources where appropriate, and note where they conflict. 'My Sources are Better than Your Sources', is not going to help us. --John R. Barberio ^{talk, contribs} 17:20, 7 October 2005 (UTC)[reply]

My comments were simply a response to what you said in your posting (above) (which I quoted at length - "the implication that Licklider initiated the creation of ARPANET, which was infact initiated by Robert W. Taylor").

There is, as far as I know, no other academic history (i.e. written by someone with a training in history, and properly footnoted) history of DARPA. So Transforming Computer Technology is very much "the standard academic history of DARPA". If there is another one, I would be very interested to hear about it.

Finally, I'm kind of curious why you persist with refusing to use the definite article with "ARPAnet" (e.g. "for initiating ARPANET"), even though i) nobody else does this, and ii) I have pointed this out to you on numerous occasions. E.g. Tannenbaum, Computer Networks (2nd Edition, 1989) uses the article (e.g. "connected to the ARPANET", pp. 35). I am rather at a loss to understand your persistence in this incorrect terminology - or is this just one more place where you are determined to go your own way, no matter what? Noel (talk) 03:09, 10 October 2005 (UTC)[reply]

A short search of Amazon dug up - Strategic Computing: DARPA and the Quest for Machine Intelligence 1983-1993 Alex Roland, Philip Shiman - from the MIT press so definatly an academic text. (Priced as an Academic text too) I'm a little concerned that you belive only qualified historians can write histories, and appear to wish to apply that standard to sources. A secondary source text should be primarly judged on its merit, it's basis on primary sources and the quality of research, and peer review. Not on the providence of its writer.

On, 'The ARPANET' and 'ARPANET' discussion. I will defer to your judgement when editing on the article on that, but I don't feel it matters so much in casual discussion on talk pages. (Incidently, generaly, it is appropriate to drop the definite article when it causes readabity issues. For example 'Tumptutumptu technology was based on the ARPANET based research' vs 'Tumptutumptu technology was based on ARPANET based research'.) --John R. Barberio ^{talk, contribs} 09:37, 10 October 2005 (UTC)[reply]

The origin of the Internet

Copied from above for the first part of this point.

My major objection with the 'old version' is that it lacked substantial information, and gave incorect impresions about the origin of the Internet. ... --John R. Barberio ^{talk, contribs} 23:52, 3 October 2005 (UTC)[reply]

I can't respond to this one without knowing more of what you think the correct "origin of the Internet is". Can you please provide more specifics? Noel (talk) 03:12, 10 October 2005 (UTC)[reply]

Those have been listed in the other objectins above. --John R. Barberio ^{talk, contribs} 09:22, 10 October 2005 (UTC)[reply]

Impact of X.25/IPSS, UUCP and the Fidonet linked BBS's

Following on from that, is the condensing of the entire impact of X.25/IPSS, UUCP and the Fidonet linked BBS's into a single line. This misses out a huge part of the history of what we now call the Internet. Some versions of the article also implied a 'space race' battle of development rather than a melting-pot of ideas. (X.25's Cites. To sum up, European participation in ARPANET was limited, X.25/IPSS being used instead. X.25/IPSS would later be used as a transport for IP protocols. [3], [4], [5]. Various from Usenet and UUCP, the Usenet article is pretty explicit on its significance.) --John R. Barberio ^{talk, contribs} 23:52, 3 October 2005 (UTC)[reply]

This point covers too much ground for me to respond in detail (to establish the influence and importance, or lack thereof, of X.25, UUCP and Fidonet would take a modest amount of investigation for each).

(1) However, let me point out (as I already have, at Talk:History of the Internet that your first citation (the paper by Ronda Hauben) says nothing related to these points: as I pointed out in my comments there, X.25 is only mentioned tangentially (a total of three passing references in the entire paper).

The amount of text taken to say something does not negate it being said. The article clearly states that X.25 had primacy in europe, and that the IPSS network was later used as a fundamental infrastructure for international IP rollout. --John R. Barberio ^{talk, contribs} 09:55, 10 October 2005 (UTC)[reply]

(2) The second citation is a pastiche of material gathered from the Web ("The information gathered in these articles were obtained from the above sites") and in any case doesn't say much relevant to these points.

(3) The third citation seems to be mostly about British telephone stuff (and British Telecom), and contains no mention of either "Internet". "X.25", or "UUCP", so I don't the relevance of that one at all.

The article provides dates and details about the creation of IPSS, the X.25 network, which was the first international packet switched network. And as already mentioned, IPSS went on to be a big part of the international infrastructure. It is unwise to tell the history of the spread of the international internet without mentioning IPSS, since IPSS made an international internet possible. --John R. Barberio ^{talk, contribs} 09:55, 10 October 2005 (UTC)[reply]

(4) I spent a day some weeks ago reading about the networks involved in the spread of the Internet across Europe and the Pacific Rim (this material is all in the various IETF Proceedings, available online here; since it's the earlier ones we are talking about, from the late 80s and early 90s, the scans of the hardcopy Proceedings will of course have the same information as the original hardcopy I looked at); I don't want to try and characterize that mass of material from memory, but I will run down it again and list which networks did what; I seem to recall that most (outside the UK) either had multi-protocol backbones, or used leased lines for carrying IP traffic. Noel (talk) 03:46, 10 October 2005 (UTC)[reply]

The multi-protocol backbones would have included many TCP/IP over IPSS links. Certanly JANet, certanly the converted european UUCPNet, certanly CERN. As stated in the cited texts, these all used the IPSS X.25 service. --John R. Barberio ^{talk, contribs} 09:55, 10 October 2005 (UTC)[reply]

(5) There definitely was a race between TCP/IP and the ISO stack (CLNP/TP4/etc); you will find it well documented in Abbate. Noel (talk) 03:46, 10 October 2005 (UTC)[reply]

Such information may be of relevence to an article solely on the history of TCP/IP, or a history of networking technologies. The ISO stack has had little to no relevence on the history of the Internet. --John R. Barberio ^{talk, contribs} 09:55, 10 October 2005 (UTC)[reply]

Structure of the article

I think another important difference between the two versions of the article is the that of the article's top-level structure. Here are the different structures that were originally up:

From [6]

    * 1 Before the Internet
    * 2 Motivation for the Internet
    * 3 Early Internet work
    * 4 Early growth
    * 5 Commercialization and privatization
    * 6 Early applications
    * 7 Standards and control
    * 8 World Wide Web
    * 9 See also
    * 10 Further reading
    * 11 External links

From [7]

    * 1 Before the Internet
    * 2 The networks that would become the Internet
    * 3 Merging the Networks, and creating the Internet
    * 4 Use and Culture
    * 5 External Links
    * 6 References

I believe that within these structures there is a fair bit of room for compromise. For instance, John, do you think that there is no need for a "Motivation" section? And if the references are complete enough, is there need for a "Further reading" section?

I think it might be worth a lot to try to come up with a top-level structure that keeps the strengths of both. I'm going to put a rough possibility below based on the information that was originally in each section: can an adequate structure be reached by editing it? — Asbestos | Talk (RFC) 19:49, 10 October 2005 (UTC)[reply]

Please edit:

    * 1 Before the Internet
    * 2 Motivation for the Internet
    * 3 The networks that would become the Internet
    * 4 Merging the Networks, and creating the Internet
    * 5 Standards and control
    * 7 Use and Culture
    * 7.a Early use and applications
    * 7.b Commercialization
    * 7.c E-Mail and Usenet
    * 7.d World Wide Web
    * 7.e Search Engines
    * 7.f ...
    * 9 See also
    * 10 References
    * 11 External links

It's tricky to keep a linear readable history of this, since it's actualy several histories combined. I'd prefer to keep it a seperate sub-history of the Networks joining up, the Standards organisations, and the way the Internet is used. I think Comercialization of the internet might belong under a subheading of Use and Culture, seperating out the change in use from the history of the networks. --John R. Barberio ^{talk, contribs} 21:10, 11 October 2005 (UTC)[reply]

Further note. The History of the Internet is actualy three different histories.

The history of Internet technologies.
The history of the Internet's infrastructure.
The history of the Internet's useage and culture.

A lot of the debate here seems to me to stem from the confusion of the first two. The technology used in an infrastructure is not the same thing as the infrastructure itself, and both have diferent stories that need to be told. eg. The debate over X.25 is not, from my side, about the technology used, but the infrastrcture that was in place.

It may be that the article would be better suited to being seperated into three articles, with the main article summarising each one. (ps, back from Ireland now) --John R. Barberio ^{talk, contribs} 22:24, 17 October 2005 (UTC)[reply]

Ending Mediation

This mediation has clearly stalled. It looks like John, you're still interested in continuing, but Noel, you've gone on an extended wiki-break with no intention of coming back mediation to this if/when you come back to the project.

Obviously you're both free to go back to editing the History of the Internet article. However, although Noel is not here, I want to remind John that there were others who supported Noel's point of view, and so would like to make sure revert wars don't spark up again. Please make sure that reasonable citations are made for as much information as possible, preferably using the Wikipedia:Footnotes structure within the article. It's usually much easier to compare the relative authority of different footnotes than it is to compare the authorities of different editors.

I'm sorry that we didn't get to make as much progress here as we would have liked,
— Asbestos | Talk (RFC) 20:59, 23 November 2005 (UTC)[reply]