Talk:History of the Internet/archive3

From Wikipedia, the free encyclopedia

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents 1 Article dispute 2 Protected 3 Structure 4 Pictures 5 Poll 6 What should this article cover 7 International connections 8 NSFNET contributions to the Internet 9 I give up 10 Unprotected 11 Mediation 12 What's the dispute about? 12.1 Consensus making 12.2 Primary Sources 13 Mediation result

Article dispute

You are the one who showed up at an article that had been basically static for months (if not years), and started making extremely aggressive edits to have it follow your original research POV.

I changed the article to have separate "technical" and "usage" threads because I thought (see comments above from User:Sdedeo 03:13, 6 September 2005 (UTC), and you 12:36, 6 September 2005, at #Peer review results) that that was something you thought would be an improvement; it also seemed to me (see my comments above) that it would make it easier to straighten this issue of "what influenced what" out if we separate the technical and usage threads.

However, I see that you have seen fit to completely discard the hours of work I put into that (keeping much of your text, moving other text to articles where it seemed more appropriate, etc), and simply reverted to your preferred version. (In this light, the irony of your comments about "Unanamous editing is not a good thing" is fairly stunning.)

If you do wish to bring up the amount of time spent on an article... You spent a few hours undoing the work I'd spent three days on. --John R. Barberio ^{talk, contribs} 10:35, 16 September 2005 (UTC)

I didn't "undo" it. A lot of it is still there, some I moved to other pages (e.g. ARPANET, where I put the story of the 3 terminals, which is an ARPANet story), etc. Yes, I did delete some: it's called "editing". You are the one who didn't keep one single atom of my work. Noel (talk) 18:48, 16 September 2005 (UTC)

I never consented to your massive rewrite (I don't have time to monitor every page on a daily basis, I've been busy with material elsewhere, as well as duties such as maintaining RfD). You and one other editor (Sdedeo) is not "consensus" - I don't concur, and I'll bet CoolCaesar (when he has time to drop in) won't agree either.

Since you don't like my version, and I don't like yours, we should simply revert to the last version before this dispute started (edit by User:Nixdorf on 21:33, 25 August 2005), and discuss the matter before making any changes at all to the prior content. Noel (talk) 17:15, 15 September 2005 (UTC)

At the top of this page is a yellow box. It tells you to discuss on the talk page before making substantial changes. You did not discuss on the page. CollCaesar is no longer involved in this dispute, and you are the sole disputer over this issue. If you check the history of the temp page you will find some other people went over the edits too. (No, the IPs are not mine. All my edits will come from 217.155.134.152/29)

I checked with an Admin User:Evil_Monkey and on #wikipedia before making the revert. And I had already listed this page as an RFC. --John R. Barberio ^{talk, contribs} 19:56, 15 September 2005 (UTC)

Making a massive change to the article, and then sticking a "disputed" box at the head of the page once it's done, does not mean that you get to use that as an excuse to protect your massive change from further editing.

And I did say something (twice, once here, once on your talk: page) about editing the page (to switch to a scheme which separate the technical and user threads). There's also a certain irony in you berating me on my talk page for "no longer contributing to the article", and then totally deleting my work when I do make a contribution. Noel (talk) 00:44, 16 September 2005 (UTC)

The contraversy tag has been present since 30th of August. I'm going to assume you just haven't been paying attention, rather than take your accusation as a personal attack. But you are comming close to the line here.

And 'I am going to Edit the Page' is not discussion of the edits you intend to make. --John R. Barberio ^{talk, contribs} 10:04, 16 September 2005 (UTC)

Let me point out to you that the notice box at the head of this page (the same one that was there when I made the edit)says only Please read this talk page discussion before making substantial changes. So your claim that it said I needed to discuss on the talk page before making substantial changes is incorrect. As should be completely obvious from that fact that I posted a lengthy note at 02:00, 15 September 2005, shortly before I did the edit to the article, I had read the talk page. Noel (talk) 18:48, 16 September 2005 (UTC)

Protected

Alright, I've reverted to Squeakbox's version of September 4 and protected. Why doesn't everybody calm down and work this out? I know nothing of the subject, so I shall retreat. Notify me if you've worked it out and I'll unprotect. john k 18:27, 15 September 2005 (UTC)

That apparently didn't stop some anon from screwing around with the text formatting. It says protected, yet an anon was able to edit. Everyking 06:04, 16 September 2005 (UTC)

Well, that's very odd. Let me note here that the revert to Squeakbox was an error, and I've reverted to an earlier version from August. john k 06:33, 16 September 2005 (UTC)

Structure

OK, I figure the first thing is to work out what structure the article should have. It seems to me that there are several possibilities:

1. Strict chronological, mixing technical and growth/user
2. Complete separation, with two completely separate technical and growth/user sections
   • An obvious variant on this is to move either (let's call this A), or both (B for this one), to separate articles
3. An intermediate between the first two, with "early technical", "later technical" interlaced with "early growth/users", "later growth/users", etc

Did I miss any reasonable ones? I don't have any particular preference among these, any would be fine with me, but I think readers might find #1 a bit confusing. Noel (talk) 18:56, 15 September 2005 (UTC)

This is getting the issue confused.

Atempting to split the article into Social and Technical issues would not only be confusing to read, but impractical to impliment. How do we define 'technical history' vs 'cultural history'. Is UUCP a culture or a technology? Is the spread of TCP/IP in europe cultural or technological?

This also will confuse the matter while we are already trying to settle a major dispute over the significance of other networks. --John R. Barberio ^{talk, contribs} 19:18, 15 September 2005 (UTC)

OK, so if I understand you, you don't want to separate out the technical development from the rest of the history in any form (either to separate articles, or to separate sections). Does this mean you want to do strict chronology?

If not, how do you want to organize the article? What overall scheme do you think will best serve to organize it (bearing in mind the reviewer's comment about "the real problem is that the article seems rather unclear as to what it is describing .. These various questions are answered only partly, in a confused order"? Noel (talk) 19:42, 15 September 2005 (UTC)

Please go and read the comments I made over the peer review responce. I rewrote the article as componant sub-histories, and the editor who provided the initial peer review felt that addressed his problems perfectly.

Since the person who pointed out the problems, feels the problems were fixed by componant sub-histories, then I suggest there is no issue here. --John R. Barberio ^{talk, contribs} 19:57, 15 September 2005 (UTC)

First, I want to make sure I understand what you mean by "component sub-histories". Do you mean a capsule history of each of the networks that were subsumed into the Internet?

Second, I was perhaps being potentially confusing in simply describing things as "technical". By technical, I mean "changes to the protocols and engineering, both application and lower-level" - no less, no more. So, to me, the growth of the Internet in Europe (to pick an example from above) was not a technical change (rather, it was one of expansion of the user community), whereas both the introduction of subnets (to pick a change that was visible only internally) and the creation of the WWW (to pick one that the users could see) were technical changes.

With that clarification in hand, do you feel a need to describe the history of the technology in this article? E.g. The insides of the Internet have changed considerably over the period 1980-2005 (and I don't mean the applications, such as the Web, etc). Where would such things fit, in your scheme, if we have them here (since you said you didn't want to move the technical history elsewhere - or would this kind of technology be something you would rather see elsewhere)? Noel (talk) 20:31, 15 September 2005 (UTC)

By "component sub-histories" I mean the format demonstrated in my rewrite. Capsual histories of each grouped subject matter. Not limited to the networks.

On the matter of seperating a technical history. The issue is that this article is not a technical article, and discusions of technologies and their histories are best suited to pages on those specific technologies. (As with packet switching) Some brief discusion of the technology used it warrented, but not any indeapth discussion. Specific user visiable applications such as WWW and E-Mail are includable, because they are directly related to the Use and Culture of the internet, but the focus should be on describing their impact on how the internet is used and what for, not the technologies involved. This article will be read by people who are not looking for technical histories, and people looking for technical histories can read the specific articles on the technologies they are looking for. --John R. Barberio ^{talk, contribs} 20:49, 15 September 2005 (UTC)

Please bear with me as I ask some more questions, because I still haven't clearly understood some things.

First, what would qualify as a "grouped subject matter"? That particular term doesn't give me a lot of guidance as to what might, and might not, qualify as a suitable topic for a "grouped subject matter". I.e. what's the dividing line between things that would, and would not, in your mind, qualify as being worthy of treatment in such a component sub-history?

Also, how would you order these component sub-history blocks in the article? I.e. how do you decide that X should come before Y?

Second, as to the technical content, let me try and understand you, because I'm trying to figure out how this article is not a technical article fits with your earlier desire not to split [apart] the Social and Technical issues. It seems there must be some technical content you want to keep, and some you don't want to see here, but I want to make sure I understand what your criteria are. Perhaps some examples of some boundary cases will make this clearer.

For instance, obviously the article should not go into depth on the graph theory behind routing (i.e. path selection, not forwarding packets inside a router), but at the same time, would you want to see if cover the big CIDR debate of the 90's (when it was introduced), and why the change was made so that people can't take their IP address with them when they switch ISP's?

Also, please don't mistake the fact that I'm simply asking questions at this point to mean that I agree with any of your positions; I'm just trying to make sure I completely understand them. Noel (talk) 00:35, 16 September 2005 (UTC)

I'd thought these were common english terms that should make sence to people with experience of writing?

When writing a document, you increase readability by grouping subjects into the same paragraphs and headings. The order you put these paragraphs should form a logical progresion. In cases of history, this means you should group similar subjects together, but avoid sudden jumps back and forth along the time line.

Here's an example of a bad thing to do. In your rewrite, we jump from NSFNet, to UUCP. NSFNet is put at the end of the 'Creating the Internet' section, with a date of 86. UUCP get's put in under the confusing 'Later Developments' heading, but described as 'in paralel', and then dated at 79. This simply jars and confuses the reader.

'Technical History' means 'A history that goes into technical details and methods'. This is not suitable for the article, as a Technical History would be over a managable length. The proper method is to link to articles on the specific technologies, and keep mention of technical details to as brief as they can be kept while the organisational, population and cultural issues will still make sence. Since readability is king, this is flexable to the needs of making a readable article. However, splitting technical details off to another seperate timeline in the article would defeat the point, and also make the article less readable. To get a whole picture, the reader would have to read a bit of the cultural history, then refer over to the technical history, then back to the culutural history...

Since it becomes obvious we are at deadlock, I suggest a simple poll between the two versions of the article. --John R. Barberio ^{talk, contribs} 18:39, 16 September 2005 (UTC)

I can figure out roughly what a "grouped subject matter" might be, but that was not my question. To repeat it, it was:

what would qualify as a "grouped subject matter"? That particular term doesn't give me a lot of guidance as to what might, and might not, qualify as a suitable topic for a "grouped subject matter". I.e. what's the dividing line between things that would, and would not, in your mind, qualify as being worthy of treatment in such a component sub-history?

To put it another way, what are the criteria you have for what topics belong in the article (and, therefore, those that do not)?

As to the business of temporal jumps in the text, any time one adopts anything other than a strict temporal organizational for presenting material you're going to get items that aren't in time order. Needless to say, if we did use strict temporal ordering, the reader would then be totally confused because we'd have to interleave events happening in a whole flock of different indepdendently-developing networks, all of which were happening at the same time. It would be completely impossible to get a clear view of the development of any one of them.

The "technical history" issue I will take to a separate section, as it's a separate major issue. Noel (talk) 21:20, 16 September 2005 (UTC)

Pictures

I have taken the liberty of moving this section to the bottom so that recent arrivals will see it. Noel (talk) 18:32, 16 September 2005 (UTC)

At the moment, we only have two pictures for the article. Both are escentialy showing early network maps, but are not really giving any tangiable content to the article. Nor do we have any information from cited sources as to what these maps are off. I'm still not sure of the interpretation of these as Internet maps. We've no texts to go with them to explain what they mean, or even what they are mapping. IntFeb82's date and content, showing what I assume are MilNet hosts prior to the MilNet split, does seem to imply this is, rather than a map of 'The Internet', a map of the TCP/IP testing network within ARPANET. Since IntCirca85 comes from a period where the Cern based Internet was growing, I feel that this would be better titled 'The ARPANet based Internet'. Noel, can you tell us where these maps came from, and if there are full descriptions of what they are actualy showing, or cites to go with them?

I don't think we need two network maps, since a second one adds nothing to the article. We need a few pictures on other topics to show. Network equipment used, early web pages and so on. --John R. Barberio ^{talk, contribs} 03:19, 8 September 2005 (UTC)

We have a photo of the first webserver, for what it's worth - Image:First Web Server.jpg. Some headshots of notable individuals at relevant points should be easy to find, as well, and would at least break up the text a bit. Shimgray 17:50, 9 September 2005 (UTC)

It does show the network at a very different stage of its growth. The first map, from '82, shows the Internet when it was still almost entirely experimental. The second, from late '85, shows the network after almost 3 years of frenetic growth after it became a production network at the start of '83. This is an article about the History of the Internet, so I would assume graphics of its growth (and these maps are the best images you are going to get of that) would be right on point. I placed the maps at the appropriate points in the page, contemporaneous with the events being described in the text at that point. I will expand the captions to more clearly explain what they show.

Some errors in the message above: there was no MILNET in '82 (the time of the first map - you do not seem to be aware of that). None of the maps show any hosts: just networks and routers. There was no Internet "within" the ARPANet; the ARPANet was a component of the early Internet. Not all the ARPANet hosts as of '82 spoke TCP/IP, of course, but all hosts attached to the other networks in that map did. The reason that CERN doesn't show up on the '85 map is that, as of '85, it was not attached to the Internet. A lot of sites (including CERN) were running internets using TCP/IP back then, but were not connected to The Internet; none of them are shown, and I don't think there's any way to find out what they all were.

As to where did they come from, I personally scanned them in from original hardcopies that are in my files. I don't know which meetings I was at that I got them at, I went to innumerable Internet-related meetings between '77 and around '90 or so (when I pretty much stopped going). Noel (talk) 22:17, 14 September 2005 (UTC)

Um. If you don't know what they are, or where you got them from, then it's not really apropriate for you to make any interpretation as to what they are beyond 'Early TCP/IP Network MAP' and 'Network Map showing ARPANET and MILNET'. Your opinion as to what they show is not encyclopedic. --John R. Barberio ^{talk, contribs} 21:06, 15 September 2005 (UTC)

At first, this statement by you irritated me (since it basically implies that I'm either a) incompetent, or b) mendacious), but on reflection I'm glad you made it, because it's a perfect example of the kind of egregious and ridiculous behaviour you engage in. It will therefore be a great education to everyone reading this, as a perfect example what we have to deal with from you.

First, I never said anything like [I] don't know what they are. (I know perfectly well what they show.) What I said was I don't know which meetings I was at that I got them at, a totally different thing. You seem (perenially) to mis-read or mis-interpret even the simplest declaratory English sentences.

Second, as to what they actually are (and show), if you would bother to do the slightest bit of investigation, you could verify that description I gave of them is correct. (For the '85 one, try reading, say, RFC 900, and read the list of assigned IP class A and B network numbers, which you can cross-check with the map.)

Third, since you seem unwilling to accept them (since I can't provide the exact date/place I got them - what's next, wanting my lunch menu and the name of the waitress for the days I received them?), let me point you at the Proceedings (note: 13MB pdf file) of the Second meeting of the Internet Engineering Task Force, held at the Ballistics Research Laboratory on 8-9 April, 1986, where you will find the very map I labelled "circa 1985" reproduced on page 118, part of a presentation entitled (humourously) "The Internet Through the Ages", by Bob Hinden. (Which also helps to verify my description of what it shows.)

Fourth, when you say it's not really apropriate for you to make any interpretation as to what they, that's equally untrue. See, in a legal proceeding, I would be characterized as an expert witness on this matter. If you look at the list of attendees given on page 9 of those Proceedings, you will find me listed. (This meeting may have been the place my copy came from: I still have my hardcopy of the Proceedings, but the attached presentations are no longer with it.) In other words, I was at the table when Bob gave that presentation.

A more perfect illustration of your willingness to use any excuse, no matter how flimsy or ridiculous, to accept any information that contradicts your erroneous POV would be hard to arrange. When I said I personally scanned them in from original hardcopies that are in my files (emphasis added), most people would find that good enough. But apparently not you.

Now that we've (hopefully) settled that the map is real, etc, can we get back to discussing what the structure of the article ought to be? Noel (talk) 18:32, 16 September 2005 (UTC)

Evidence you provide based on your own personal knowledge and conclusions is not suitable for a Wikipedia article, it is Original Research, and not allowed. I do not dispute that the maps are real. They are, they came from ARPANET. The first one obviously shows the early IP/TCP test networks, and yes, that's verifiable from RFC 900. But that does not provide a basis for the description 'The Internet in Preproduction'. I am happy to accept your credentials as someone who has lots of experience with ARPANET, but Wikipedia has no such thing as 'Expert Witnesses', and we can not publish your opinion as fact. (Otherwise, I'd be publishing the Opinions, such as from people who ran UUCP nodes, that seem to conflict with yours.)

Again, can I please ask you not to induldge in personal attacks. I fail to see why you can not accept perfectly neutral titles such as 'A map of the TCP/IP test network' for these images, that allow the user to make their own conclusion if they wish that they show an 'early Internet'. --John R. Barberio ^{talk, contribs} 18:51, 16 September 2005 (UTC)

I have really stumbled into quite a scuffle here, but I can't help but weigh in on this as a neutral observer. First, I figured that following a comment attempting to discredit Noel's personal knowledge of the subject matter would be a great place to say kudos to you Barberio (who, according to his own bio, "learnt the basics of UNIX, software engineering, and how the Internet works" at school in the late 90s), for having the balls to engage in a debate with someone who has been doing this for decades and who actually participated in the development of the Internet.

Second, despite your courage, Noel obviously has the upper hand in this discussion. I don't think it is fair to push someone to the point of absolute frustration by responding to fact-filled specific references and responses to your own comments, with mostly irrelevant and factually inaccurate comments of your own over and over again, and then berate that person for engaging in personal attacks when they get a litte frustrated with you. I would be frustrated too given your obstinance in what appears to be an attempt to merely play down the fact that what we know today as the Internet was largely developed from the ARPANET here in the U.S. Your motivations are quite obviously politically motivated, and thus highly suspect.

Third, I'm particularly baffled by the "technical history" bit. Isn't a history of the Internet by definition going to primarily be a technical history? When I read a history of the CD player I expect to read a history of the various technologies that go into CD players and how they were brought together and developed into what we know as a CD player.

Anyway, I'll continue trying to wade through this mess and provide more substantive comments as the discussion progresses. -Jefu 08:00, 17 September 2005 (UTC)

Ah. The back handed compliment. Nice way to start off with a personal attack that you can semi justify as "but I was complimenting you". Ah well.

Your Second Point. I am not claiming, or have ever claimed during this argument, that ARPANET is not a hugely significant part of the development of the Internet. I agree, and wrote in the article that ARPANET was the sole major contributor of technology.

This discussion is over a simple question. "Was ARPANET the sole significant contributor of culure and methods of use for the Internet as we know it today?" I say No, Noel says Yes. I belive, and have backed up with evidence, that ARPANET while a major part of growth, and a source of a culture of openess of standards, was not the sole significant source of culture and methods of use. And that the other networks and groups of people produced some important things. If you check the article history you will find that I've had to strugle to get UUCP, Compuserve et all, and CERN mentioned in more than a passing reference.

Your third point. Well, go read the discussion below. I won't repeat myself. In fact, I'm getting a bit tired of personal attacks, circular arguments, atempts to polarise the issue into ARPANET or Nothing, followed by personal attacks saying I'm trying to polarise the issue into ARPANET or Nothing, and why do I keep insisting on bringing up ARPANET anyway, and I should just shut up and let the big boys do the work... I think I'm going to just re-edit the poll, then sod off for a week or so to cool down. --John R. Barberio ^{talk, contribs} 09:37, 17 September 2005 (UTC)

First, they didn't c[o]me from ARPANET (sic). They are hard copies. Someone personally gave them to me, several decades ago. How many times do I have to repeat this? (This ARPANET fixation of yours is really irritating. Please cease with it. Or are you just a troll? I'm starting to wonder.) I could explain to you why your proposed caption was inaccurate, but I am certain it would be a complete waste of my time, because you'd just come up with some other bullshit reply.

So, instead, whose word (if any, since you seem determined not to accept anything that conflicts with your fixed POV) will you take that they are correctly captioned? Maybe the current chair of the Internet Engineering Task Force, Brian Carpenter? He's a European (and ex-CERN to boot). If he's not good enough, how about Vint Cerf himself? This is not sarcasm - I know them both, and would gladly ask them if it will stop you from yammering about this.

And why the devil do we need to allow the user to make their own conclusion if they wish that they show an 'early Internet', when one of them comes from a 1986 presentation whose label is The Internet Through the Ages - a fact which anyone can verify by looking at the offical repository of the Proceedings of the Internet's engineering body?Noel (talk) 21:04, 16 September 2005 (UTC)

You said yourself that the title was given as a joke, not a reflection of the content of the talk. If Carpenter or Cerf wish to contribute to the article, I'd be happy to see them do so. Because I'm confident they'd keep to reporting the historical facts, not their opinion on if ARPANET is the sole significant contributor to the internet. If you want to start a section titled 'dispute over the origin of the internet' with quotes from them to support your POV, then I'd be happy with that too.

Then I might quote the Internet Society history on the matter -

'The original ARPANET grew into the Internet. Internet was based on the idea that there would be multiple independent networks of rather arbitrary design, beginning with the ARPANET as the pioneering packet switching network, but soon to include packet satellite networks, ground-based packet radio networks and other networks. The Internet as we now know it embodies a key underlying technical idea, namely that of open architecture networking. In this approach, the choice of any individual network technology was not dictated by a particular network architecture but rather could be selected freely by a provider and made to interwork with the other networks through a meta-level "Internetworking Architecture".'

And,

'In an open-architecture network, the individual networks may be separately designed and developed and each may have its own unique interface which it may offer to users and/or other providers. including other Internet providers. Each network can be designed in accordance with the specific environment and user requirements of that network. There are generally no constraints on the types of network that can be included or on their geographic scope, although certain pragmatic considerations will dictate what makes sense to offer.'

and,

'At the same time that the Internet technology was being experimentally validated and widely used amongst a subset of computer science researchers, other networks and networking technologies were being pursued.'

which seem to indicate that the writers accept that while ARPANET gave a fundamental and significant role in technical development, it was only the catylist in the creation of a global network from groupings of orignaly seperate networks. --John R. Barberio ^{talk, contribs} 21:35, 16 September 2005 (UTC)

Yet again, you have misrepresented what I said earlier. (I am somewhat at a loss to explain this continual habit of yours; you manage to do it in almost every single reply you post (and not just to me): do you not bother to read things, are you just a troll, or what? Perhaps you can explain what the cause is.)

Anyway, what I actually said was that it was part of a presentation entitled (humourously) "The Internet Through the Ages". I did not say that the title was not a reflection of the content of the talk. The humour - since it seems to have escaped you (along with much else) - was in the use of the "Through the Ages" phrase.

In the (perhaps useless) interests of showing how mindless, petty, and idiotic your complaints are, though, if you look at the Proceedings from the 3rd IETF (7MB pdf file), you will find an almost identical map on page 139, from a presentation entitled "Internet Capacity Planning" - no humour at all in that title.

Tell you what, you can find other, almost identical, versions of this map, from slightly later IETF's, Looking through my harcopy originals, I see that the 9th and 10th IETF Proceeding both contain them. Why don't you look and tell us what the titles of the presentations that use them there are? Noel (talk) 22:23, 16 September 2005 (UTC)

Can I risk pointing out that what they called the Internet then, does not equate to what we call the Internet now? I actualy mentioned this in the article, that the phrase 'the Internet' when used by ARPANET users originaly refered to the TCP/IP portion of the ARPANET. The phrase now refers to the global network, regardless of what technology it uses. If we switched over to SCTP or some other protocol tomorow, people would still call it 'The Internet' afterwards. --John R. Barberio ^{talk, contribs} 22:52, 16 September 2005 (UTC)

You seem to want to redefine the term "the Internet" to mean whatever you want it to mean, at which point it's obviously simple to include (or not) whatever you want. To use "the Internet" as a synonym for 'all computer networks, over all time' may please you, but is neither accurate (and one thing an encyclopaedia should never do is propogate inaccuracies), nor one that will receive much support from people who are knowledgeable about computer networking.

You keep harping on about the "ARPANET" (you do it again above), and then refer to it meaning regardless of what technology it uses. Well, even in the (very early) 1982 map of the Internet, there were a lot of other technologies represented. Everything labelled "<something>-PR" is a packet radio network, "SATNET" is a satellite-based radio network, the "NDRE-RING" is some sort of ring network (not sure which flavour), I think the thing labeled PPSN is an X.25 network, and goodness knows what all the other things are. This is not to mention stuff that is hiding here and there: for instance, inside the blob labelled "LCS NET" there was at that date (if I recall correctly) a prototope 1 Mbit/sec ring, a 10 Mbit/second ring, an experimental 3 Mbit Ethernet, (I don't recall when we got our first 10 Mbit Ether, it was around then), and a number of 4 MBit/sec CHAOSNETs. No ARPANET, though.

And of course by the time of the circa-1985 map of the Internet, there was no more TCP/IP portion of the ARPANET" - the ARPANet by then was no longer a distinctly separate network (e.g. with its own protocol suite) in any way, it was effectively just an long-haul backbone connecting together all the LANs (with their attached host computers - VAXes, etc) at the various sites, the whole conglomeration then (as today, albeit much larger) being "the Internet".

You keep trying to either i) complete conflate the ARPANET and the early Internet, or (ironically) go to the opposite extreme, and ii) deny the close connection! Which is it? The ISOC history which you (ironically) quoted to me elsewhere said it best:

The original ARPANET grew into the Internet.'

I couldn't have said it better myself. Noel (talk) 00:17, 17 September 2005 (UTC)

Poll

This has gone on a little too long, so let's try to settle this with a poll over the prefered article. (edited)

"Was ARPANET the sole significant contributor of culure and methods of use for the Internet as we know it today?"

Vote as '''yes''', or '''no''' Give full reasons for the choice. This is not to decide which version wins, it is to determine what the page should reflect and how. You may suggest changes or a rewrite. Please note, question is not "Was ARPANET the sole significant contributor of Technologies used for the Internet as we know it today.", which concensusus says is anwsered by Yes.

--John R. Barberio  talk, contribs  19:20, 16 September 2005 (UTC) edited 09:40, 17 September 2005 (UTC)

Let's not polarise things on "this version is better or that version is better" but examine the points of dispute individually. The first thing to hash out is structure, as everybody already seems to have started on above. I've read the discussion and can honestly say I'm still not clear on what the dispute is here. To me the obvious choice seems at least some sort of chronological ordering, that generally works best for histories. Since judging by the favoured editions everyone seems to agree on that how about as a next step we find a set of headings/topics, and work our way down from there? It's easier to come to some sort of agreement over small specific things than over the state of the entire article in one go. --fvw* 19:50, 16 September 2005 (UTC)

I take it that's a vote for rewrite? --John R. Barberio ^{talk, contribs} 20:08, 16 September 2005 (UTC)

No, that's a vote for "Let's not vote about this but discuss each individual choice and its merits". --fvw* 21:02, 16 September 2005 (UTC)

Once again, your contentions are factually incorrect. Your statement:

"Was ARPANET the sole significant contributor of Technologies used for the Internet as we know it today.", which concensusus says is anwsered by Yes

is incorrect, because as the article mentions (and as I keep pointing out to you, although you seem to keep ignoring it - at least, I cannot offer any other explanation as to why you keep making inaccurate statements about it), the early CYCLADES network provided some important ideas.

Well, you and Coolescar put up a huge rant in the talk page about *why* ARPANET was the sole significant contributor of Internet Technologies... If you've now changed your mind, I'll accept a shift to the shared development in technical matters too. (ps, to quote what you said - And of course the ARPANet's role as the principal technical precursor of the Internet (modulo CYCLADES, which was already mentioned in the article before you showed up) is a given. Noel (talk) 21:32, 14 September 2005 (UTC)) --John R. Barberio ^{talk, contribs} 15:52, 20 September 2005 (UTC)

Your principal statement:

Was ARPANET the sole significant contributor of culure and methods of use for the Internet as we know it today?

is equally misguided. Nobody is trying to say that it was the sole "significant contributor". However, it was the most important - as the ISOC quote you yourself used on me indicates:

"The original ARPANET grew into the Internet"

I couldn't have said it better myself. Noel (talk) 04:56, 19 September 2005 (UTC)

What should this article cover

Copied from above.

'Technical History' means 'A history that goes into technical details and methods'. This is not suitable for the article, as a Technical History would be over a managable length. The proper method is to link to articles on the specific technologies, and keep mention of technical details to as brief as they can be kept while the organisational, population and cultural issues will still make sence. Since readability is king, this is flexable to the needs of making a readable article. However, splitting technical details off to another seperate timeline in the article would defeat the point, and also make the article less readable. To get a whole picture, the reader would have to read a bit of the cultural history, then refer over to the technical history, then back to the culutural history. --John R. Barberio ^{talk, contribs} 18:39, 16 September 2005 (UTC)

The thing is that the Internet is a technological construct, and its development as a technology is going to be hard to present, and even harder for the average reader to to grasp, if that evolution is not given as a unified whole, but rather thrown in as little and dribs and drabs here and there in a narrative which is in large part about other things (such as population and cultural issues).

(For a good example of a reasonably-well-done "History of <technology X>", one which gives a coherent history of how that technology developed, check out History of the automobile.)

So I had that goal in mind when I sectioned the article into "Technical development of the Internet", "Initial growth of the Internet" "Later growth of the Internet", and "Further technical development". What I tried to do with that organization was to try and keep those two intertwined, but very different histories from becoming so entangled with one another that neither made sense as a coherent picture of how that aspect evolved. I further divided up the "technical development" line into two blocks so that I wouldn't wind up describing technical developments from the 2000's before we described organizational developments from the 1980's.

Yes, that resulted in a certain amount of temporal disordering, but that's almost inherent any time you try and tell a multi-faceted story in a coherent way. For example, look at Abbate's Inventing the Internet - Chapter 5 covers stuff up through the mid-80's, and then chapter 6 drops back to the early 70's to pick up another thread.

I used two sections for "growth of the Internet" for a number of reasons.

For one, it was so I could add the text

"Parallel to the Internet, other networks had grown up .."

"As the Internet took off, in time it absorbed these other networks. In addition to their user communities and something of their culture, in a number of cases the Internet garnered new applications from these networks as well."

as a header to the collection of subsections which came immediately thereafter ("UUCP and Usenet", "X.25 networks", "America Online", etc), by way of explaining how they fit into the big picture.

For another, the early growth of the Internet (prior to sometime in the late 80's) was mostly internally-driven; i.e. new components were added piecemeal throughout the existing network. However, sometime around the late 80's (not sure exactly when, will have to look) a different phenomenon appeared. Pre-exisiting network using other protocol suites started converting to use TCP/IP so that they could become an integral part of the Internet. This seemed to me a big enough difference to warrant a separate section.

I am not, however, tied to this particular structure. I do, however, insist that we have an article (somewhere) that tells of the development of the technology of the Internet as a coherent story - just the way we do for cars. Noel (talk) 22:00, 16 September 2005 (UTC)

'The Internet is a Technical Construct' is as true a sentence as 'Printing is a Technical Constructs'. Acurate, but misleading, since culture plays a significant part in its history as well.

What works for books does not work for Wikipedia. The 'space' between chapters allows for greater elasticity in ordering.

I have no problems with you authoring your own article on the technical history of TCP/IP networks. However, I suggest you use the name 'History of Internet Protocols' to properly identify a history of the technology used, with a redirect from 'Technical history of the Internet' if you want.

Also can you please stop moving your discussions down the page to try and get me to reply quicker. I know how to use a watchlist, and you could even send me a message if it's that important. --John R. Barberio ^{talk, contribs} 22:22, 16 September 2005 (UTC)

That is not why I'm putting things at the bottom of the page. As I said above The "technical history" issue I will take to a separate section, as it's a separate major issue. Noel (talk) 22:29, 16 September 2005 (UTC)

Culture also plays a role in cars (or vice, versa, actually); it some countries (the US, Canada, Australia, etc) it has had a big impact on the physical organization of the country. But the development of the automobile as a technology still rates (as I think it should) a clear, coherent telling. So I don't see the point of your culture plays a significant part in its history as well comment. Noel (talk) 16:22, 17 September 2005 (UTC)

As I've been thinking about this, I've had an idea.

There were literally hundreds of computer networks around the world that the Internet "ate" as it grew. A lot of people know of the ones in their home country, but most industrial countries had them, as did many specialist fields. Many (most?) are now largely forgotten. I know of some of them, because I ran across them in my decades as a professional in the field, but I expect that even I know of only a fraction.

Here are some of the ones I recall: VNET and BITNET (both worldwide network using IBM protocols); SPAN (an international network for astronomers, used DECnet protocols IIRC); MERIT, in Michigan (a large network which used a home-grown protocol suite; its personnel later went on to man the T1 phase II NSFNet - phase I having been the Fuzzball-based 56K network); HEPnet (the high energy network, using DECnet protocols, IIRC; CERN was on that, I think); JANET (a large UK X.25 based network); ESnet (a network run by reseach organizations working for the US DoE; I think it ran DECnet, IIRC); JUNET (a Japanese network, don't know much about it - WIDE was the IP portion of it); MFEnet (a network in the US used by magnetic fusion people, started in '76, again DECnet IIRC); EARN (European academic/research network, don't remember much about it); SURFNET (in the Netherlands - they were a leading light in the formation RIPE, IIRC - another thing that ought to be covered); NSN (a NASA network, using ACTS satellite links);

There are some international efforts I recall, but don't know their names: I know New Zealand had a large DECnet network among universitites/research institions, and I assume Australia had some sort of networks too. I've heard of academic networks in Poland and the USSR, both of which are rumoured to have played roles in the political troubles there (the Polish one in Jaruzelski's crackdown, the Russian one in the KGB/Communist attempted coup), but I don't know any details.

And looking through some of my huge stash of archaic documents, here are some more: BCnet, ARnet, Onet, RISQ, NSTN (a group of 5 Canadian regional research/educational networks; not sure what protocols they used); WIN (an X.25 network of educational institutions within Germany); NORDUnet (Scandanavia, ran X.25); EASInet (some sort of European network, forget what they did); DFN (Germany, ditto); SWITCH (Switzerland); GARR (Italy); EUnet (some sort of European educational and/or research network); DEnet (Denmark), FUNET (Finland) SURIS (Iceland), UNINETT (Norway), SUNET (Sweden) (all components of NORDUnet); TISN (a Japanese research network that ran DECnet); NACSIS (a Japanese educational X.25 network)

And I also notice, looking through this list, that I see primarily research/academic networks - commercial networks are basically not represented at all. And of course there are North American and Western Europe with plenty of names, but what of the rest of the world? And did Europe, etc have BBS's, etc and that sort of culture?

So, here's the point: there was an entire world of computer networks other there, AKAIK all of them later folded into the Internet. If we really want to cover the growth of computer networking as a social/etc phenonenon, it's hard to see how to cover all this work in the same article as the history of the Internet itself.

It would seem to me that there's a place here for an article on the growth of computer networking as a social/etc phenonenon, with a title of something like History of computer networking or History of computer networks or something. It would seem to me that the obvious way to organize it would be in several sections:

• International generic networks (VNET, BITNET)
• Networks for a particular field (i.e. SPAN, HEPnet, MFEnet, etc)
• Regional networks (NORDUnet, EARN)
• National/local networks (JUNET, MERIT, JANET, etc)

with listings for each network, and a few words about them (plus links to articles, when they exist), so that someone who wanted to read about the scope of computer networks before the Internet had one place to go to get a well-organized overview of 'the rest of computer networking'. (Such an article would, of course, be linked to from HotI.)

I think it would only make sense to mention in the "History of the Internet" article only those networks whose inclusion had an impact on the Internet; e.g. Usenet (netnews), AOL (chat rooms and IM; although AOL wasn't the inventor of either, I think it was AOL who really got them to take off), etc. Most of the networks I wrote of above were folded into the network with hardly a ripple on the Internet as a whole - who can point to an effect from MERIT, or JANET, or JUNET?

And the internationalization of the Internet should be covered in a more organized way - there's some attention paid to Europe at the momen, but I don't think anything's said about the Pacific Rim, Asia, etc. Noel (talk) 16:22, 17 September 2005 (UTC)

I'm sorry, but now you're being silly. I do not want to, and never intended, to mention every little network or BBS that went to make up the internet.

However, mentioning the major networks, and mentioning the types of other networks, that would make up the internet is important. We do not need to mention every X.25 network, or dial in BBS, to mention that such networks existed, and significantly contributed to the internet we know today.

The major issue I have is the way the article was presenting the creation of the internet as a Race, that was won by ARPANET. (To quote the now excised summary - For a variety of reasons, some good, some accidental, the network we know as the Internet is the one that won the race to become that network.) Such a presentation was totaly at odds with the culture and development of the internet.

What's more, ARPANET belongs in the *before* the internet section of any article. Because, as I discovered when rereading my copy of Tanenbaum's Computer Networks, the implimentation of the first true 'Internet' came from NSF's fuzzball based network merging with an ARPANET converted over to TCP/IP in the mid 80s. The developers may have used ARPANET to comunicate about this development, but development of NSFNet did not *come* from ARPANET. The first 'Internet' came from the merging of both NSFNet and ARPANET, not from any one network along. And it grew by merging in other networks. (I need to fix this in my version as well, as I too read the other histories and asumed from their tendancy to focus on ARPANET that NSFNet came from it.)

It seems to me only fitting in a discusion of the Internet to mention these networks, and give them importance based on their population, applications and cultures.

Again, if you want to write an article about the technical development of TCP/IP, you are free to create a suitable article to do so in.

Since it appears the page was not protected after all, since it never went through the correct protection process, I'm considering reverting to the Peer Review Rewrite. Because that version was worked on and accepted by a total of seven people, three during the rewrite, and four before your rewrite. This indicates that most people were happy with that version.

To put this bluntly, your version did not add much new to the article. It added inacuracies such as the claim that AOL invented chatrooms. It confused the reader over the order of things, occurances before 1984 should not be labled 'Later growth of the Internet'. The article was hard to read, and did inapropriate things such as quoting Licklider in the article summary, when article summaries must be kept short and terse. (AOL deserves a mention, but Compuserve had much greater cultural impact. I initialy did not mention it because I could not identify if AOL used X.25, or used a proprietary internal network.)

If anyone else other than Noel wants to raise issue with the rewrite article, then I'll rewrite again in the same way I did before on a temporary page based on your points. Noel, sorry, your opinion is valid, but while you have filled this talk page with acusations and argument, your version and mine had few differences and despite your decrying their addition kept the expanded mentions of UUCP X.25 and such. You have been over inflating issues keeping us deadlocked. Till you can discuss the issues without draging it into a "I'm right, your wrong" argument, I'm going to have to put a low weight on your comments. --John R. Barberio ^{talk, contribs} 19:17, 18 September 2005 (UTC)

I was explicitly not suggesting we cover in this article all those networks I listed; to be exact, I said:

I think it would only make sense to mention in the "History of the Internet" article only those networks whose inclusion had an impact on the Internet

What I was saying is that in the history of "cyberspace", i.e. the development of the current networked world, all those networks do have a place, and that since they don't belong in this article (a position with which I see you agree), they need to be covered elsewhere.

As to your comment:

the way the article was presenting the creation of the internet as a Race, that was won by ARPANET. (To quote the now excised summary - For a variety of reasons, some good, some accidental, the network we know as the Internet is the one that won the race to become that network.)

First, you are once again conflating the ARPANet and the Internet! The original quotation you yourself clipped says Internet! Your incredible hard-on about the ARPANET is really amazing. Please try and control it.

Second, it is absolutely true that in the late 1980's there were several competitors, to the Internet and its TCP/IP technology, to become the world's 'global, ubiquitous network'. The principal one was the OSI stack from the ISO, but the CCITT's X.25 was also a major competitor, in addition to a variety of proprietary commerical protocols, such as IBM's SNA. At one point it was even planned to phase out TCP/IP, and convert to the OSI protocols!

I don't have the time and energy to give you a complete reading list to educate you, but here are a few that are easy for me to find: RFC 939 ("The committee recommends that the DOD move .. toward exclusive use of TP-4") and subsequent documents (e.g. RFC 945); RFC 1039 ("It is intended to adopt the OSI protocols as a full co-standard .. when GOSIP is formally approved .. Two years thereafter, the OSI protocols would become the sole mandatory interoperable protocol suite"). You should also look at all the IETF Proceedings from the late 80's to see all the networks in Europe and elsewhere that ran other protocols (e.g. DECNet, X.25, etc). It's all covered very well in chapter 5 of Abbate's book. ("Having survived the standards war, the Internet emerged with an even stronger basis for worldwide expansion." pp. 179.)

As to your contention:

the implimentation of the first true 'Internet' came from NSF's fuzzball based network merging with an ARPANET converted over to TCP/IP in the mid 80s. The developers may have used ARPANET to comunicate about this development, but development of NSFNet did not *come* from ARPANET

This is completely incorrect. Since I am fully aware that it's pointless for me to explain to you why it is incorrect (you will just blow it off, no matter what), I'm simply going to send your statement to the following people:

• The implementor of the Fuzzball (David Mills), used in the phase 0 (56KB) NSFnet
• The engineer who was the primary technical person at IBM on the router used in the phase 1 (T1) NSFNet (Yakov Rekhter)

to see if they agree with it.I'm very sorry I have to bother these people with this nonsense, but nothing less than a 2x4 between the eyes seems to get you to desist from your ridiculous assertions

Your version of the article was filled with so many similar errors that it would take me way too long to list them all. Noel (talk) 04:00, 19 September 2005 (UTC)

My version of the article had almost exactly the same factual content as your version.

As to NSFNet, Quote from p51 of Tanenbaum's Computer Networks, "The Software technology was different however: the fuzzballs spoke TCP/IP right from the start, making it th first TCP/IP WAN.", then on p52 "Sometime in the mid-1980s, people began viewing the collection of networks as an internet, and later as the Internet, although there was no official dedication with some politician breaking a bottle of champaign over a fuzzball." Now, it might be that he is mistaken in that claim, but it does seem prudent to assume he is correct. If people disagree over this, then it becomes a point of view issue, and should be balenced and report both.

Competing technologies does not equate to competing Networks. The networks, and the people who ran them co-operated to create the Internet. This was not like the space race.

Now, I would like to insist that you do list those errors you found, so that they may be corrected. It goes against the wiki spirit to say 'there are errors!' but not identify them so they can be fixed. Nor is it a reason to throw out a version of an article if it contains easy to fix errors but is otherwise fundamentaly sound.

If you can not provide a suitable list of reasons why the rewrite version [1] is an unsound article, then we must conclude the dispute. If you can provide the problems, then we can start fixing them.

'We should write a seperate technical history' is not a problem with the article itself, as it can be simply solved by creating a second article containing a detailed technical history of TCP/IP. I consider that discusion closed.

To sum up, there has been a demonstrated concensus in support of the article. The article was reviewed and found to be readable, and properly structured. If there are factual problems with the article, then they should be raised here, and fixed. --John R. Barberio ^{talk, contribs} 10:39, 19 September 2005 (UTC)

Here's the reply from David Mills, who wrote the Fuzzballs used in the phase I (I'm not sure that's official terminology, I used "phase 0" in a descriptive sense, not implying that was its formal name) NSFNET. His length replies (our conversation went on for a number of messages) included a lot of repetitive restating of historical material, which I have left out. So that others can verify I'm not changing the sense, I have attempted to include very lengthy quotations:

Date: Mon, 19 Sep 2005 15:33:02 +0000

From: "David L. Mills" <mills@udel.edu>

the implimentation of the first true 'Internet' came from NSF's

fuzzball based network merging with an ARPANET converted over

to TCP/IP in the mid 80s. The developers may have used ARPANET to

comunicate about this development, but development of NSFNet did not

*come* from ARPANET. The first 'Internet' came from the merging of

both NSFNet and ARPANET, not from any one network along.

> Can you take a few seconds to comment on this? My position (which will

> no doubt not surprise you) is that the Internet was a going concern

> *before* the NSFNet was joined on, and that the addition of the NSFNet

> was not the step that created "the first Internet',

The Internet existed in public name from the 1979 coming out party at

the National Computer Conference. ...

It is fair to say that from 1979 the technology had stabilized. ...

The oveseas partners in SATne, including UCL London, RSRE Malvern (UK

military), DFVLR Munich (German NASA), CNUCE Pisa and NTARE Oslo,

were funded by joint support from the respective national governments

and DARPA. They were in that sense DARPA contractors. ...

The NSFnet Phase I (not 0) backbone did indeed use Fuzzballs at the

six supercomputer centers. It started in 1986 and was overwhelmed by

IBM/MCI/Merit in 1988. So far as I can rmember all the sites were

connected to university campus nets and those nets were in turn

connected to ARPAnet IMPs. ...

However, the Internet in name and function was established well

before then with ARPAnet transport at many university sites. Thus the

backbone was more a political statement than a wideband transport.

So I think this should put to rest the notion that there was a major distinction between the Internet with the ARPANET as its primary long-haul core, and the Internet with the NSFNET. I draw attention particularly to Dave's opinion that the Internet in name and function was established well before [the phase I NSFNET]. Of course, the NSF adopting TCP/IP was an important step, and the NSFNet was for a period the key backbone, but those are stages in a long evolution, and no more important than many other steps along that long path. Noel (talk) 02:49, 22 September 2005 (UTC)

International connections

In the course of searching for material on SATNET, I ran across a very well done proto-paper on the international connections to the development of the idea of internetworking (which I have mentioned here, well before the recent imbroglio, see e.g. /archive1#Creation Myths - Europeans did have a very important impact on the early days of the Internet (i.e. prior to 1980)). The paper is:

• Ronda Hauben, The Internet: On its International Origins and Collaborative Vision (A Work In Progress)

available here, among other locations.

This proto-paper provides a large amount of detail about the very early days, in the early 70's. It does rely heavily on recent communication with the people involved, which is a source which can have its problems (especially this long after the fact). The use of multiple contact points (as in this paper) diminishes the chance of problems, of course, but original contemporary documention is still the gold standard. (E.g. I recently contacted one of the people cited in this paper on another matter, and their recollections contained a number of errors which original documentation pointed out.)

Although I would differ with the author on a few minor points of emphasis and interpretation (and there are also a few very minor errors, which I have notified them about), I am broadly in agreement with the thrust of this paper, and recommend it as a resource. Noel (talk) 19:36, 19 September 2005 (UTC)

It all seems to confirm what I've seen elsewhere. Particularly the sections confirming the importance of X.25 in the development of the Internet in the UK. Only mentions UUCP in passing, and CERN isnt mentioned at all, but it appears to be concentrating on the UK and Norway. There's some extra info in here that could go in the European Internet section. (Lets asume for now we'll be using the rewrite version)

It's a little dry and hard to read at 1am tho, so I'll go over it again tomorow to try and glean things we can use or quote. We do need to start looking for expansion details and figures on other regions following europe. But I fear the details are going to be sparse, and not 'joined up', making it very hard to research comprehensivly.

Noel, I hope this means the content dispute is over now? --John R. Barberio ^{talk, contribs} 00:12, 20 September 2005 (UTC)

Absolutely not; I'm not sure what should be on this page (more on this in a bit, still thinking about it), but your proposed draft is totally unacceptable.

But it is instructive to see what lessons you drew from reading this paper. For instance, in a paper of about 1100 lines (i.e. about 25 pages), in the body of the text there are a grand total of three references to X.25:

* "the main research initiatives were in pursuit of the X.25 protocol suite and its upper levels. There was almost no European activity on the Internet Protocols outside Oslo and UCL" (quotation from Kirstein)

* "From 1976, there was increasing pressure for using the emerging X.25 infrastructure (International Packet Switched Service – IPSS) as an alternative to SATNET." (text)

* "Finally, when the ARPANET had moved to Internet Protocols, we could abandon our relays in BBN and also leave SATNET; all the traffic could use IP/X.25 over IPSS." (quotation from Kirstein; this refers, of course, to events after 1983, in any event)

But instead of concluding from this that X.25 did not play a significant role in the formation of internet ideas (this paper, after all, is focussed on the subject of the early development of the ideas of internetting), you don't seem to have changed your position one whit. Noel (talk) 14:25, 20 September 2005 (UTC)

Um... I'm confused. So you're agreeing that the article outright states that UK internet roll out was dependant on using the IPSS X.25 infrastructure, and that X.25 was prefered over the older network technologies. But X.25 is still unimportant and the rewrite unaceptable because of it? I think now you're flat refusing to accept that X.25 was used as a foundation technology in spreading the early internet.

Noel, can you also please stop editing my comments. If I want to indent, I will, but I was not replying solely to you. --John R. Barberio ^{talk, contribs} 15:17, 20 September 2005 (UTC)

I was the only poster in that section. Who else were you replying to? Indenting your text is hardly "editing" it; I was merely trying to follow Wikipedia:Talk page guidelines#Layout, which says "Use indenting to keep the conversation straight: .. the next person starts with one colon". But since you object even to me keeping the pages formatted to follow the Wiki-standard (this is really getting ridiculous), I'll just change the indenting on my comments, then. Noel (talk) 17:12, 20 September 2005 (UTC)

Found some information about internet growth into Africa circa 1991. Yet again, this seems predicated on the X.25 IPSS Network for the internal infrastructure. And also significant use of UUCP for intermitantly connected machines, and direct dialing into overseas networks over the international phonelines. Global Networks for Africa --John R. Barberio ^{talk, contribs} 16:03, 20 September 2005 (UTC)

More on IPSS. Now that I know what to look for, it seems that the IPSS X.25 servide acted as the initial worldwide infrastructure that turned into the Internet.

"1977 ... The first international switched data network service was offered. Western Union International and Tymenet collaborated with the British Post Office in a trial service. In 1978 the service went live. Links to Canada, Spain, Hong Kong and Australia were added by 1981."

"1981... The first full public packet-switched service opened in the UK, based on GTE Telenet. The average delay of a message on the network was 150 milliseconds to 200 milliseconds. This national network was connected to the International PSS which had been established since 1977."

Events in British Telecomms History, abridged from now lost OFTEL documentation [2]

Seems that X.25 turns out to be very important indeed. --John R. Barberio ^{talk, contribs} 16:49, 20 September 2005 (UTC)

NSFNET contributions to the Internet

In the course of what turned out to be an extended email discussion with Yakov Rekhter, about the exact importance of the phase-II NSFNET (i.e. the T1 version) in the evolution of the Internet, he made useful point about the evolution of the routing (i.e. path selection, not packet forwarding), the importance of which hadn't really struck me before.

He pointed out that work associated with the T1 NSFNET produced a routing system for the Internet as a whole which was much more suitable to a network in which the overall connectivity model was mesh-oriented, as opposed to the hub-and-spoke model used by the Internet when the ARPANET was the core long-haul backbone in it.

Here's the conclusion of the discussion:

Date: Mon, 19 Sep 2005 10:50:26 -0700

From: Yakov Rekhter

...

>>> what were your significant sources of technical influence, in doing

>>> the T1 NSFNET? I would assume prior Internet work was the biggest?

>>> (I.e. technically, the NSFNET was mostly an evolution of the the

>>> pre-existing Internet?) Or am I wrong there?

>> From the routing point of view the T1 NSFNET was not exactly "an

>> evolution of the pre-existing Internet" as routing in the T1 NSFNET

>> was not exactly an evolution of the routing in the 56 Kb Internet

> Right; I was speaking in a broader context, though, of the network as a whole

> (i.e. applications, the whole concept, etc), not specifically of the routing

> (which of course you all had to re-do completely; the overall Internet

> routing used at the time of the 56KB NSFNET was simply not workable).

> Your points about the routing are most valuable; I hadn't thought about that

> angle much before. Would you say that that was the most significant technical

> contribution of the NSFNET work to the development of the Internet?

Yes.

The BGP, developed by Yakov and others to replace the previous EGP-2 has much greater capabilities, and really was a key step in the development (as big as Van Jacobsen's TCP congestion control stuff). Noel (talk) 17:12, 20 September 2005 (UTC)

This appears to be supporting the way the "ARPANET to NSFNet" section in the 19 September version was? Or should it be a bit more specific in saying "In 1984 this resulted in the first Wide Area Network designed specificaly to use the full capabilities of TCP/IP." rather than "In 1984 this resulted in the first Wide Area Network designed specificaly to use TCP/IP." Can we now move on so we can get the page unprotected and start putting this new information in? I think BGP is probably an important enough technology to demand a brief mention within the section and a link to [Border Gateway Protocol] for more detail. --John R. Barberio ^{talk, contribs} 19:26, 20 September 2005 (UTC)

After a little research, mention of BGP is more suited to the comercialisation and deconstruction of NSFNET section. As it is both a contempary event, and the enabling technology that allowed decenteralisation. --John R. Barberio ^{talk, contribs} 19:32, 20 September 2005 (UTC)

I give up

I'm fed up with the petty bickering, deadlock and personal attacks. Noel, you win, edit the page as you wish. --John R. Barberio ^{talk, contribs} 01:18, 22 September 2005 (UTC)

Well, I had come here to post this (initially composed offline earlier, and which I just finished tweaking):

I have reviewed what seems to be, from your posting above, your most recent preferred version of the article. (If there's another more recent one somewhere else, please let me know where it is).

I find that it still contains numerous errors (I have a list offline), but my most significant problems with it remain your novel personal historical interpretations, which are not in line with the major histories on this topic (e.g. Abbate, Inventing the Internet, MIT Press).

For instance, you persist in making a big point of X.25 at the head of the article, even though it's only mentioned in passing in both Abbate, and an article specifically about early international contributions to the idea of internetworking. There are a number of other similar structural problems.

This is exactly why I no longer want to bother with this. Instead of providing cites to disprove that X.25 is a significant contributor to the early internet, you cast aspursions, provide a personal opinion based on page space in your prefered text, and wrap it all up in a personal atack.

I'm sorry Noel, I can no longer work with you, since you make it imposable to do so. --John R. Barberio ^{talk, contribs} 11:54, 22 September 2005 (UTC)

I would therefore like to propose a different approach. We should start with the version of 21:33, 25 August 2005, by User:Nixdorf, and you should propose changes to that. Please don't rewrite the entire article, that will just put us back into trouble; chose one issue, let's work out a workable revision, which we can then install in the article, and you can then move on to the next point you'd like to adjust.

(I'd like to note here that of the 100 edits prior to that version, precisely 2 are by me - and this one consists of changing "[[University of Minnesota]]" to "[[University of Minnesota system|University of Minnesota]]" . So it's not like we're starting from 'my' version of the article. The old state represents the rough consensus of many Wikipedians who worked on it.)

If that doesn't work, I have another idea, but I think this one is more likely to be productive, which is why I mention it first. Noel (talk) 03:20, 22 September 2005 (UTC)

but I guess it's Overtaken By Events (as the saying from the 1st IETF goes...) I'm not interested in 'winning'; my main interest is in having an accurate history of the Internet. Noel (talk) 03:20, 22 September 2005 (UTC)

Unprotected

Is there still a dispute? I suppose the easiest way to find out is to unprotect and then sit back and watch. --Tony Sidaway^Talk 19:20, 25 September 2005 (UTC)

Oddly the page is still stuck on a version we know to have poor readability, be inacurate and omit significant sections of the history. Getting very tempted to return to editing the page. --John R. Barberio ^{talk, contribs} 08:04, 29 September 2005 (UTC)

Okay. The page seems abandoned on this old version. I'm withdrawing my retreat from editing, since we really need to update to remove the errors and readability problems on this page. (We definatly don't need that lengthy OSI rant back in the page!) Noel, I'll be tagging this with the 'two versions'. Before making any revert or rewrite, list a *full* list of your complaints with the page, how you wish to fix them, and full cites to support on this talk page, and give it at least a full day for a responce.

Incidently, I've also uncovered that the early TCP/IP development is substantuialy predicated on the Xerox Parc work on Xerox Network Services and PARC Universal Packet (Soon to be merged since they are escentialy the same article). Most specificaly in the use of the Routing Information Protocol. This really is a melting pot development, not a 'space race' style compertition. That said, not sure if its significant enough to warrent a mention. --John R. Barberio ^{talk, contribs} 17:12, 30 September 2005 (UTC)

I see it starts again. Well, I'm tired of fixing your errors - I will simply revert to the last edit before you started introducing all your errors and Original Research POV. Noel (talk) 00:41, 1 October 2005 (UTC)

Again, I ask for a full list of the errors you find. Also, since you bring this up again, a full list of any OR you found not supported by cites brought up in talk and referenced. Reverted back, since you don't appear to want to follow the big boxed instructions not to revert, and in your own words "Can't be bothered to weed out the Original Research POV".

Before making another revert, list your full objections in detail. 'There are errors' does not cut it, you must be more specific so we can actualy work to correct those errors, or reverify them if they turn out to be correct.

--John R. Barberio ^{talk, contribs} 09:25, 1 October 2005 (UTC)

If you must work on this page, I suggest you use the procedure I outlined in the section above (which see for more details): suggest a particular change you want to make, and we'll discuss it, and then insert it. Noel (talk) 11:58, 1 October 2005 (UTC)

The changes have actualy been discussed. You've had ample time to bring up specific complaints, and I've addressed the ones you did bring up. If you wish to make further changes, you may of course state the problems with the current article, cite to back up any factual changes, then we can procede. If you feel this is an unjust way of proceding, then I would be happy to enter into mediation with a third party over this. --John R. Barberio ^{talk, contribs} 13:13, 1 October 2005 (UTC)

I've restored protection. Evidently the dispute is still ongoing. --Tony Sidaway^Talk 17:17, 1 October 2005 (UTC)

Mediation

At User:Essjay's suggestion, I have filed a RfM. See WP:RfM#Pending. Noel (talk) 13:53, 1 October 2005 (UTC)

What's the dispute about?

Sorry for being a newcomer here, I have seen the edits popping all over this article but never paid close attention to it for a long time.

I just can't get a grip about what the dispute here is all about. Could you line up the controversial issues, one by one, so we could at least get some consensus on a few of them? Nixdorf 18:35, 1 October 2005 (UTC)

Well, right now, your guess is as good as mine. I'm not entirely sure what Noel still finds objectionable which I havent adressed. His complaints above were, to sum up,

• Mention of the alternative networks, specificaly X.25

  Addressed by pointing out how much cultre, use and infrastructure impact these other networks had. The X.25 IPSS provided the first international packet switched network, and this network was used to provide the international infrastructure of the early Internet. UUCPNet gathered a huge userbase, and spread the concept of e-mail and the usefullness of networking. Both of these networks brought in comercial use prior to NSFNet's opening to comercial use.

• Giving NSFNet priority over ARPANET as being the initial (captial I) Internet.

  Addressed by cites showing NSFNet was specificaly designed with internet protocols in mind, and co-incided with the start of 'The Internet' being used to refer to a single network.

  David Mills seems to indicate he thinks 'The Internet' was around before then, but this contradicts other authorities on the matter. (Specificaly, Tannenbaum) I suspect that Mills is refering the the lowercase i internet, which was also used by several other network developers of the time. (Such as Xerox with XNS) I'm flexable on the etymology issue since the texts are vauge. But the way internet is used in documents from the time to refer to methods or examples of a possible network rather than a specific network seem to support Tannenbaum's assertion.

I am not sure where Noel's claim of 'original research' is, since these were all based on referenced research. --John R. Barberio ^{talk, contribs} 19:43, 1 October 2005 (UTC)

Nixdorf: It's fairly simple, actually. Barberio arrived here with a novel historical interpretation (one totally out of tune with major academic works on the subject, such as Abbate's book), and also knowing little about the actual history (as shown by his numerous factual errors). After that, he argued extensively and pointlessly about correcting even the smallest details in his limited/erroneous knowledge (a perfect example being that of the map captions - see the final outcome at Image talk:InetCirca85.jpg). Through his lack of knowledge of the field, he continues to make numerous factual errors.

However, the biggest problem is that his basic novel historical interpration, which is just as erroneous as all the details he got wrong, is something he still clings to tenaciously (viz his comments immediately above). (His much-ballyhooed "referenced research" is akin to citing reports of Leif Ericson's settlement in Newfoundland as "referenced research" which proves that the Scandanavians were the European settlers of the New World. A few correctly cited minor facts does not a correct overall narrative make.)

A perfect example of his erroneous novel historical interpretation is his attempt (above) to add this concept: "Giving NSFNet priority over ARPANET as being the initial (captial I) Internet." To scotch this one, I had actually gone to David Mills, the person who did the Fuzzballs used in the Phase 1 NSFnet. (I shouldn't have had to go to this extreme, but with Barberio, nothing less works.) David's reply (quoted above, at the very end of the section) was "The NSFnet Phase I .. backbone .. started in 1986 ... and was overwhelmed by IBM/MCI/Merit in 1988. ... However, the Internet in name and function was established well before then". In other words, here we have the person responsible for operating the Phase I NSFnet saying "the Internet ... was established well before then".

However, Barberio is still trying to insert into the article "Giving NSFNet priority .. as being the initial .. Internet".

A more perfect example of his extra-ordinary resistance to abandoning his (incorrect) novel historical interpretation would be hard to arrange. Noel (talk) 11:51, 4 October 2005 (UTC)

Noel, these complaints should be placed on the arbitration discussion. I am not going to respond here. (Mostly because this would just be me restating what I have before.) --John R. Barberio ^{talk, contribs} 14:01, 4 October 2005 (UTC)

I was replying to Nixdorf's request. Noel (talk) 15:10, 7 October 2005 (UTC)

Consensus making

OK so I take this to mean that the core of this dispute is what to give credit to as being the true origin of the Internet: the ARPANET or NSFNet.

• Barberio wants to stress NSFNet, UUCPNet and X.25.
• Noel wants to stress the ARPANET and claim this is what most historians do.

The way we would resolve this if we tried to work as historians would be to start with citing your sources with complete references to source texts, then, indicate which are primary sources, i.e. documents that were produced at the time of the actual events or very close in time and which are secondary sources, which includes any books, papers or whatever written post-1970 or so.

The common critical method in history is not to rely on people because of their merit (i.e. "this person was involved in this thing in that year and s/he claims...") but first and foremost on primary sources, i.e. that person should provide some document from the time that states it is really true. A persons memory easily fills in the blanks with anecdotes and interpolated information.

Agreed? If so, then bring in the sources. Nixdorf 18:00, 4 October 2005 (UTC)

According to WP:NOR, we're not supposed to be relying principally on primary sources: there is a tendency by some Wikipedians to produce novel narratives and historical interpretations with citation to primary sources to back up their interpretation of events. Even if their citations are accurate, Wikipedia's poorly equipped to judge whether their particular synthesis of the available information is a reasonable one.

Were we to go down that road, I have an entire office full of primary source material: paperwork from the '77-'90 period (papers, drafts, meeting agendas, presentations, proceedings, minutes, etc, etc), and there is also a great deal online. We'd be here forever, in addition to it being a bad idea.

I maintain we should stick to policy and rely on secondary sources, by professional historians who have looked through some of these mountains of material and done the synthesis. Noel (talk) 15:10, 7 October 2005 (UTC)

However, you will note, I have supplied both primary and secondary sources on the etymology issue. --John R. Barberio ^{talk, contribs} 16:43, 7 October 2005 (UTC)

Noel. Please do not misrepresent official wikipedia policy in order to support your argument. WP:NOR actualy says the following. "Original research that creates primary sources is not allowed. However, research that consists of collecting and organizing information from existing primary and/or secondary sources is strongly encouraged." --John R. Barberio ^{talk, contribs} 17:15, 7 October 2005 (UTC)

In intrests of keeping this co-ordinated, please see my responces on Wikipedia_talk:Requests_for_mediation/Jnc_and_Barberio#Disputes --John R. Barberio ^{talk, contribs} 19:33, 4 October 2005 (UTC)

Primary Sources

List here all primary sources you can find relating to the history of the Internet, arranged in date order.

• Development of the actual Internet:
  • January 1, 1981: TCP-IP Digest, Vol 1 #10
  • November 1981: NCP/TCP TRANSITION PLAN, RFC 801 by Jon Postel
• The history of ideas perspective of the "Internet idea":
  • July 1945: As We May Think by Vannevar Bush
  • March 1960: Man-Computer Symbiosis by J.C.R. Licklider
  • October 1962: AUGMENTING HUMAN INTELLECT: A Conceptual Framework by Douglas Engelbart

It's an intresting, but irelevent, side issue that the 'predictors' of a world wide network and data store seemed to think that Voice Recognition would not only be more important, but easier to provide and use. --John R. Barberio ^{talk, contribs} 10:10, 5 October 2005 (UTC)

Mediation result

Unfortunatly, Mediation stalled. The mediator has suggested that we return to editing the page, and ensure to use full cites.

Unless anyone has objections, I'm going to edit the page a lot to get it to a standard where we can put it on the Featured Article path. I'll try to keep any new information added since mediation started. And we'll be restoring some info that was reverted during the mediation process.

Things that I would like to happen, and urge people to research :

1. Merge in Australian Internet history, and get some cites for it!
2. Add in Aditional information on the spread of the internet to other continents. We're missing Asia and Africa and South America.
3. History of Internet Games. This could really be a whole article in its own right, so if anyone wants to run with that, and give us a summary version for this article?

I'd love to see this article be a Main Page featured article some day. --Barberio 12:39, 23 November 2005 (UTC)

For my own reference, and to use as a cite, The Evolution of Packet Switching, paper from 1978 detailing the networks of the time. Appears to authoritivly dismiss 'X.25 developed from TELENET'. --Barberio 22:17, 23 November 2005 (UTC)

---

As the mediator, I'd like to just say that the easiest way out of such disputes is to use citations as much as possible, and would therefore recommend using the Wikipedia:Footnotes structure within this article as much as possible (which is also the only way that you'll be able to make this article featured, if you'd like it to be). It's also particularly important to actually read the citations that another person gives you if you're arguing with them, otherwise you won't get anywhere.

For the record, the mediation can be found here. I link this only because many good points were raised and a lot of citations were given, and so will hopefully cut down a little on redundancy in these discussions.

— Asbestos | Talk (RFC) 21:05, 23 November 2005 (UTC)