Talk:History of the Internet/Archive 2

From Wikipedia, the free encyclopedia
Jump to: navigation, search
← Archive 1 Archive 2 Archive 3 →

International contributions to the creation of the Internet

The recent anti-US edits are not helpful. If people bother to study the actual history, y'all will discover that the project had a strong international component from the very start. Many of the most important ideas distinguishing it from the earlier ARPANET work came from the CYCLADES work of Louis Pouzin in France. The research group at Stanford which did the initial concept work included Gerard LeLann from IRIA in Frnce, Dag Belsnes from the University of Oslo, and Kuninobu Tanno from Japan. The actual Internet project included a number of non-US personnel and groups, including Peter Kirstein and his group at UCL, NDRE in Norway (with Paal Spilling and Yngvar Lundh), as well as a number of individual non-American participants (e.g. Danny Cohen (Israel), me (from Bermuda), etc). If you want to add an international flavour, try adding that, instead of adding irrelevant stuff, e.g. about X.25 (which was, frankly, basically completely irrelevant), and about Donald Davies (who played a role in the ARPANET work, duly noted in that article, but had nothing to do with the Internet). But the fact remains that the project was principally driven and staffed from the US, and its simply rewriting history to try and pretend otherwise. Noel (talk) 02:57, 30 August 2005 (UTC)

PS: For instance, now that I check, X.25 wasn't done until 1976 - but the early Internet work was done in 1973, several years before X.25! So X.25 is in no way a significant part of the history of the Internet. Noel (talk) 05:56, 30 August 2005 (UTC)

Erroneous edits

Here's a partial list of the errors in the material recently added to History of the Internet.

  • Prior to the Internet, communication networks would operate based on their physical transmission method - Ah, technically, no, Even the ARPAnet needed nothing more than a bit-string and clock, and it used a number of different physical transmission technologies (including satellite links) to get that. (You may have meant something slightly different, but I can only go by what's there.)
  • as the research project at the UK's National Physical Laboratory under the direction of Donald Davies - I am unaware that Davies played any role at all in internetworking (he certainly had no role after '77, which was when I joined the project); please provide a reference to his work on that topic. (His role in packet-switching is well known, but that's different.)
  • who developed the concepts of packet switched networks - The credit for packet switching is split several ways; see the ARPANET article for a more detailed look at this topic (Baran's work is the earliest).
  • Parallel to DARPA's research, packet switching networks were developed by the International Telecommunication Union in the form of X.25. - There was nothing "parallel" about it. Telenet was an ARPANet clone, done by Larry Roberts after he left DARPA, and went online in '75. X.25 was heavily based on Telenet. In any event, the X.25 spec was only done in '76, contemporaneously with the early Internet work.
  • X.25 would see large take up in business, particularly banking and public access networks. - So? It was a non-related branch that died off (like a whole bunch of other failed networking technologies, such as SNA, DECnet, etc, etc, etc), and irrelevant to the history of the Internet.
  • The Internet's technical roots lie within the ARPANET, - You added the "technical", but this is incorrect. The "network culture" of the early Internet was a direct transplant of the culture which had grown up on the ARPANet (with mailing lists, etc, etc), so it's more than just "technical".
  • which was initially the core network in the collection of networks in the American Internet Backbone - Again, incorrect. See the 1982 and 1985 maps. In the early stages, the ARPANET was the backbone, there was nothing else.
  • Early TCP/IP work - In the earliest Internet work, there were no separate TCP and IP protocols, so this title is inaccurate.
  • The early Internet, based around the ARPANET, was government-funded - The material you struck out was factual and accurate. See the circa-'85 Internet map I added to the article.

Noel (talk) 15:17, 30 August 2005 (UTC)


Prior to the Internet, communication networks would operate based on their physical transmission method - Ah, technically, no, Even the ARPAnet needed nothing more than a bit-string and clock, and it used a number of different physical transmission technologies (including satellite links) to get that. (You may have meant something slightly different, but I can only go by what's there.)

This what a typo. I'd intended to write internetworking. Which by definition is the unification of diferent physical networks.

as the research project at the UK's National Physical Laboratory under the direction of Donald Davies - I am unaware that Davies played any role at all in internetworking (he certainly had no role after '77, which was when I joined the project); please provide a reference to his work on that topic. (His role in packet-switching is well known, but that's different.)

Um... Packet switching is inherently an internetwork technology, not sure I can imagine any other way it would be used?

who developed the concepts of packet switched networks - The credit for packet switching is split several ways; see the ARPANET article for a more detailed look at this topic (Baran's work is the earliest).

Well, the article on Packet Switching gives Davies credit?

Parallel to DARPA's research, packet switching networks were developed by the International Telecommunication Union in the form of X.25. - There was nothing "parallel" about it. Telenet was an ARPANet clone, done by Larry Roberts after he left DARPA, and went online in '75. X.25 was heavily based on Telenet. In any event, the X.25 spec was only done in '76, contemporaneously with the early Internet work.

Then edit this to be correct. Say that X.25 is a branch of the early ARPANet development. I do not accept it's a clone of ARPANet since Telenet

X.25 would see large take up in business, particularly banking and public access networks. - So? It was a non-related branch that died off (like a whole bunch of other failed networking technologies, such as SNA, DECnet, etc, etc, etc), and irrelevant to the history of the Internet.

X.25 was the home for a wide wide range of comercial public access networks, including AOL and Compuserve. Leaving them out makes it appear that comercial public access networks sprung up after NSFnet.

The Internet's technical roots lie within the ARPANET, - You added the "technical", but this is incorrect. The "network culture" of the early Internet was a direct transplant of the culture which had grown up on the ARPANet (with mailing lists, etc, etc), so it's more than just "technical".

I dispute this. USENET, and BBSs in general gave the Internet it's "geek" culture. Remember, ARPANET was a military development, the D in DARPA. The mailing list culture came over from USENET and the colleges where research took place.

which was initially the core network in the collection of networks in the American Internet Backbone - Again, incorrect. See the 1982 and 1985 maps. In the early stages, the ARPANET was the backbone, there was nothing else.

Fair enough, I won't dispute this. However, between 1985 the X.25 college and institute networks in europe were being converted over to a European Internet so it is important to note that ARPANET was the US Internet backbone. (see http://www.mkaz.com/ebeab/history/)

Early TCP/IP work - In the earliest Internet work, there were no separate TCP and IP protocols, so this title is inaccurate.

Never the less, the section in question was solely about the development of TCP/IP.

The early Internet, based around the ARPANET, was government-funded - The material you struck out was factual and accurate. See the circa-'85 Internet map I added to the article.

Which lacks any of the European Internet based around CERN.

I'll re-edit to adress your concerns, and restore information on the non-ARPANET contrabutions to The Internet.--John R. Barberio talk, contribs 10:45, 30 August 2005 (UTC)

Replies:
  • Packet switching is inherently an internetwork technology, not sure I can imagine any other way it would be used? - This statement is completely nonsensical. Packet switching was used in non-internetworking networks like the ARPANET, before the concept of internetworking was developed. LANs use packet switching ideas, i.e. time-multiplexed access to communication resources without a fixed allocation of resources. Etc, etc. So your statement about "not being able to imagine any other way it could be used" is, well, seriously confused.
  • the article on Packet Switching gives Davies credit [the concepts of packet switched networks]? - Haven't got around to fully straightening out that article yet. (Although I did defend it from people who wanted to give Kleinrock all the credit.)
  • Say that X.25 is a branch of the early ARPANet development. I do not accept it's a clone of ARPANet since Telenet - This is incomplete, so I'm not sure what your point was, but the point remains that X.25 (whose spec was done in 1976) was in no way any sort of predecessor or inspiration for internet ideas, which were done in 1973.
  • X.25 was the home for a wide wide range of comercial public access networks, including AOL and Compuserve. Leaving them out makes it appear that comercial public access networks sprung up after NSFnet. - You will note that I carefully left in your (unmodified) text about "Commercial public access X.25 networks had also taken root in many countries; such as Compuserve, Tymnet and Euronet.", which was a) correct, and b) in the appropriate place in the article. They had nothing to do with the early growth of the Internet (this article is about the Internet, let's not forget; it is not History of computer networking), so it's inappropriate to mention them in a section which discusses early work on the Internet.
  • USENET, and BBSs in general gave the Internet it's "geek" culture. Remember, ARPANET was a military development, the D in DARPA. The mailing list culture came over from USENET and the colleges where research took place. - You again appear to have a very confused, mistaken and completely incorrect notion of the early networking world. Try looking at the 1985 map. You will see a large blob labelled "ARPANET", which has mostly universities attached to it, and a much smaller blob labelled "MILNET", which is where all the military stuff that used to be on the ARPANET lived after they split the combined network into two separate ones (one for military, one for research/academic). This ratio of non-military to military prevailed for a long time before the split - dig up a copy of HOSTS.TXT from the early 80's, and you can check it.
  • However, between 1985 the X.25 college and institute networks in europe were being converted over to a European Internet so it is important to note that ARPANET was the US Internet backbone. - The situation in the US in the late 80's was very fluid, and the ARPANet rapidly faded as the US backbone. (I seem to recall that the first NSFNet procurement, the one that wound up using Fuzzballs, was early in '86.) It is certainly true that the ARPANet was never extended to Europe to form the backbone of the Internet over there, but nobody ever tried to say it was. The point remains that during the first couple of years of the growth of the Internet as an operational network, i.e. its most critical and formative stage (i.e. from '83-'86), the ARPANET was the principal backbone of the entire Internet, as shown by the 1985 map. So this "US Internet backbone" terminology is just, well, wrong.
  • Never the less, the section in question was solely about the development of TCP/IP - What came out the far end eventually was TCP/IP, but the start of it was about the development of internetworking as a general concept, so it's wholly inappropriate to try and (incorrectly) pigeonhole it as being only about TCP/IP.
  • Which lacks any of the European Internet based around CERN. - There's a reason CERN's not on the 1985 map. There was no Internet there (European, US, International, or any other kind) in 1985. CERN didn't connect to the outside world using TCP/IP until 1989. See A Short History of Internet Protocols at CERN - which also notes that the Internet only became widespread in Europe in 1988.
Please go away and study some histories so you have a better idea of what actually happened. Noel (talk) 07:01, 31 August 2005 (UTC)

UUCP

Expanding out the UUCP section, since Usenet was a very significant and widespread pre-cursor to the modern Internet. --John R. Barberio talk, contribs 16:57, 30 August 2005 (UTC)

You seem to be confusing this article with one entitled History of computer networking. Perhaps you need to go work on that one. UUCP has zero connection to the technical evolution of the Internet, and only a minor one to the development of networking culture. Detailed material on UUCP belongs in the UUCP article, which we can then reference. Noel (talk) 06:19, 31 August 2005 (UTC)

Misplaced material

The following material is not relevant here, most of it because it describes work that had no influence on the development of the Internet. It might be appropriate for an article on History of computer networking; perhaps someone would like to move it there. Noel (talk) 07:31, 31 August 2005 (UTC)

The UUCP networks had a direct and tangiable input to the development of the Internet. Many common internet useages were fostered here. It provided a hugely widespread email system, and Usenet was (and to some still is) the major discussion system prior to web based forums. A large portion of the internets cultural history lies in the volunteer run UUCP networks.
We also mention both X.25 and UUCP networks later in the article as being absorbed into the greater internet, but we have not provided any context as to what these were.
Restoring them both. --John R. Barberio talk, contribs 17:23, 5 September 2005 (UTC)

CERN

During the 1970s, CERN, the European Organization for Nuclear Research, had developed its own internetworking system, CERNET, independantly developed but sharing some aspects of ARPANET.

Usenet and the UUCP mesh

In 1979 students at Duke University, Tom Truscott and Jim Ellis, came up with the idea of using simple Bourne shell scripts to transfer news and messages on a serial line with nearby University of North Carolina. Following public release of the software, the mesh of UUCP hosts forwarding on the Usenet news rapidly expanded. UUCPnet would also create gateways and links between Fidonet and dial-up BBS hosts.

UUCPnet hosts would link in a large mesh network of hosts that allowed broadcasting of data from one site to the rest of UUCPnet. However, unlike true packet switched networks, UUCPnet lacked a method to simply address other hosts on the network. To send e-mail to a user would require you to know to path to that user's host from your own, which was known as a bang path. Users had to consult a UUCPnet Map to find the route. Some would publish routes from their own node to popular and well connected nodes to simplify the process.

More errors

Oh, here's yet another mistake: and the Internet had spanned the atlantic (in the CERN section). If you will look at the 1982 map, you will see sites labeled "NDRE" (the Norwegian Defence Research Establishment, in Norway, oddly enough), and "UCL" (University College London, in London, equally oddly). Since Norway and the UK are on the other side of the Atlantic, at least the last time I checked, this statement joins the long list of other errors you have inserted into this article.

If I seem like I'm getting really irritated, I am. I'm tired of fixing mistakes you keep introducing into the article because you know nothing of the history, and don't seem to be willing to take the effort to educate yourself. Please take more time to research things before you edit articles. Noel (talk) 07:39, 31 August 2005 (UTC)

Sigh, more errors that were just introduced. The new, 'improved' picture captions are wrong.
  • Map of ARPANET, in January, 1982 - NO! Did you even bother to look at this image, before relabelling it to fit into your prejudices? It is exactly what I had it labelled as - a map of the entire Internet in 1982 (complete with IP addresses - note that it only has class A network numbers; class B numbers had not been added yet). Again, notice the nodes labeled NDRE and UCL (see above). (PS: University College London is an academic institution, not a US defense contractor.)
  • Map of the US Internet Backbone prior to NSFNet, circa early 1985 - Again, incorrect; this is a map of the entire Internet at the date I specified. In addition to the previously attached UCL and NTA (which replaced NDRE), notice the nodes labelled e.g. "RSRE" (the Royal Signals and Radar Establishment, in the UK), "DFVLR" (I forget what the acronym stands for, but it's German), "Italy" (that one should be fairy obvious), etc. No doubt the British, Germans, Italians, etc will all be rather bemused to be told that you think they are part of the US.
I've really had it with fixing your errors. I've got better things to do than fix all your mistakes. Noel (talk) 08:33, 31 August 2005 (UTC)
To Noel or Jnc or whatever your username is: After thoroughly reviewing the page history, I agree with your exasperation, and I think that Barberio's edits are incompetent at best, and in bad faith at worst. The error you pointed out involving the image captions is particularly glaring; only a troll, an idiot, a mentally ill person or a child would make such an obvious error. Barberio himself confesses that he has CFS on his user page. I suspect that the disease may have affected his sanity as well — I know several people with CFS, and from what I've read, the loss of self-esteem caused by CFS can sometimes lead to depression, followed by semi-psychotic behavior.
Network topology is not that hard. The pictures cannot be maps of ARPANET because they do not show its internal topology; rather, they show the networks connecting to it (or internetworking with it) and are therefore maps of the Internet.
I would support a request for protection or a request for arbitration if necessary. Any admins out there monitoring this mess? --Coolcaesar 09:08, 31 August 2005 (UTC)
Plenty, but they're more concerned about the way this discussion is turning out. You do not make personal attacks, whether direct or veiled, and you do not refer to any medical conditions people may have to explain behaviour you don't like. The former is part of policy and the latter is just common decency. Please remember that when there are content disputes, contributors are asked to step back, and begin forming consensus over what should go in the article. No one person dictates what does or doesn't. Always assume good faith and try to be more accommodating to newcomers. Thanks. Rob Church Talk | Desk 19:24, 15 September 2005 (UTC)

Since this has devloved quickly into biggoted personal attacks, I'm going to step back from this article. However, I have to say that this article is not an article about the history of the internet. And has become a history of ARPANET. UUCPnet had a big cultural and technical impact on the early Internet, and I fail to understand the excising of this from history. --John R. Barberio talk, contribs 10:32, 31 August 2005 (UTC)

If you will go examine the ARPANET article, you will see that it contains a long chunk of history, none of which is replicated here. So your contention that this article has become a history of [the] ARPANET is not accurate.
In any event, after 1983, when NCP was turned off in the ARPANET, the ARPANET was only a component of the Internet (i.e. it could only be use with TCP/IP), so at that point the history of the two entities does merge.
I agree that the Internet absorbed some cultural influences from the UUCP world later on (not sure exactly over which period, see below), and I have no problem with pointing out (in fact, I would like to see that pointed out), but we don't need a detailed history of the UUCP network to make that point. I'm not sure of exactly when this influence appeared (my memory's a bit fuzzy on that), but it would obviously post-date the growth of the UUCP network, which trailed the growth of the ARPANET/Internet by some years (the UUCPnet was comprised, in large part, of sites which couldn't get connected to the ARPANet).
I'll add some appropriate words to the article, if this influence is not already adequately covered. (And of course the technical influence of the UUCP system was minimal - except for NNTP, which is an application, there was no other technical transfer.)
Note that the Internet also absorbed sizeable user communities, and their social traditions, from other networks, such as BITNET and HEPNET, but their contributions are even less well understood/documented. I also don't fully understand the cultural influence of various more informal things, such as the large network of BBS sites. Noel (talk) 15:53, 31 August 2005 (UTC)

ARPANET

There is a *lot* of information here that is about ARPANET specificaly. There is already an article on ARPANET, and it actualy lacks some of the information here. So unless anyone has any major dispute, I will be trimming down the content on ARPANET, moving some of it to the ARPANET article, and giving a 'main article' pointer to the ARPANET article. --John R. Barberio talk, contribs 17:13, 5 September 2005 (UTC)

Wow, I can't believe you took out the section about the ARPANET/Internet link. I usually refrain from ad hominem attacks, but your ignorance is appalling.
If you bother to skim a few dozen of the earliest Request for Comments memos (just search Google), it's immediately apparent (to any intelligent observer) that the ARPANET's culture led directly into the Internet culture. The RFC process pioneered by the original ARPANET personnel is still used by the Internet Engineering Task Force today, and the more than 3,000 extant RFCs form a continuous record of ARPANET and Internet history right up to the present; in fact, TCP/IP was originally proposed as RFC 793 in 1981 (see the official list of Internet standards [1]).
Plus, none of the big Internet historians (Hafner, Abbate, Waldrop, etc.) have ever disputed that the ARPANET community was not the origins of the Internet community. --Coolcaesar 20:35, 5 September 2005 (UTC)
The sentence presented the Internet as the sole origin of Internet culture. I've atempted to rephrase this a few times, but none have been deemed acceptable, and were reverted. The article was representing a POV that internet culture originated in ARPANET. And as I've pointed out, a lot of the cultural roots of the current internet lie in Usenet. (SPAM, Netiquette, FAQs, lots of our jargon...) Since atempts to mention that ARPANET was primarly the source of technical aspects (And the RFCs are a technical process, not a culture!) and cultural aspects were absorbed from other networks.
Frankly I'm annoyed at the personal attacks I've recieved on this. I admit I've made factual mistakes, but the insults and personal attacks I've recieved over this is not acceptable. If I make factual mistakes, correct them. But a lot of this discusion has made little resort to cite, instead using personal knowledge or opinion.
Marking this article with NPOV templates. Nominating for peer review. --John R. Barberio talk, contribs 21:34, 5 September 2005 (UTC)
Actually, the RFCs are both a technical process and a culture. If you actually bother to read a statistically significant sample, you will realize that they contain many interpersonal, non-technical discussions which are normally considered inappropriate in technical standards documents (at least those released by nearly every other standards organization). Furthermore, another indication of the cultural significance of RFCs is the prevalence of joke RFCs (such as the notorious one about IP packet transmission over avian carriers).
Also, I am not basing my assertions solely on personal knowledge, although I will acknowledge that other editors here have. I have cited to public documents available at numerous sites all over the Internet (the RFCs), and authors whose books are available in every well-funded English-language public library in the world (science writer Mitchell Waldrop, science historian Janet Abbate, and journalist Katie Hafner). On the other hand, it is you who keeps inserting edits based solely on personal knowledge or intuition without adding references to generally available, reliable, and easily authenticated sources. --Coolcaesar 04:43, 6 September 2005 (UTC)
Please give me a direct cite to support the conclusion that ARPANET is the sole major cultural predecesor of the Internet. Yes, there were occasional RFCs with jokes in them. This does not reflect the entire early internet culture, and should not be represented as such in the article.
If the article does not contain information on the other sources of Internet Culture, the predominant one being UUCPNet, then we should make no claim that "The Internet's Culture" originated in ARPANET.
You are also continuing your personal attacks. Please stop. If you had bothered to check, you would note that my information came either with a cite, as with Cern, or a link to a wikipedia article on the subject that provided the same information. --John R. Barberio talk, contribs 07:55, 6 September 2005 (UTC)
Well, Wikipedia is not a reliable source in and of itself, especially where the subject at issue is disputed. The point I'm getting at is that you need to learn to cite to reliable printed sources. Anyway, unlike some editors of this page, I have a full-time job, and I have other articles which I am interested in contributing high-quality research to, so I am dropping this page from my watch list and withdrawing from this discussion. Feel free to distort this article all you like!
Although I am quite passionate about Internet history, I am refraining from drawing upon my own personal library of books and articles to contribute to Wikipedia on this subject because I am slowly drafting a book on the history of a certain technology (not the Internet but something related to it) which I hope to publish someday. I do not want to inadvertently license choice paragraphs under the GFDL and then struggle to rephrase them for my own book. --Coolcaesar 03:42, 8 September 2005 (UTC)

Peer review results

Coppied over from the peer review page. --John R. Barberio talk, contribs 12:36, 6 September 2005 (UTC)

Informally, just looking it over, it is severely lacking in stuff on the European side, such as JANET. In general, the real problem is that the article seems rather unclear as to what it is describing: the origin of the physical structures? The emergence of internet "culture" (in which case it would have to cover things like BBSes, from which big chunks of internet culture today emerged? The origin of the software and protocols? (e.g., the article seems to discuss things like e-mail, aka, the SMTP protocol.) These various questions are answered only partly, in a confused order. Sdedeo 03:13, 6 September 2005 (UTC)

I think this seriously cuts to the problem we have. The whole article has no real unified structure, or any clear definition on what history we are telling. I suggest that we basicaly need a total re-write, starting with the basics.

  1. What history are we telling?
    • This is a history of 'The Internet', but that's only a vauge term. Is this article a sole discussion of the Technical history of the internet, or does culture and politics need to be covered. To be comprehensive, the answer should be yes. But this raises the issue of what did contribute to the internet. Again, to be comprehensive, I feel we should have a wide scope on the contributions to the internet.
  2. How do we structure the article?
    • The article right now has a semi-time line structure mixed into a categorised structure. I'm not sure if this is working. We need to chose one, tell the history as a time lined series of events, or tell sub-histories of each componant of the internet. Right now I lean towards componant sub-histories.

--John R. Barberio talk, contribs 12:36, 6 September 2005 (UTC)

Going through the article for a re-write. Discovered already that we'd missed out on some important stuff. We wern't mentioning Robert Taylor at all, and had been giving the impresion that Licklider had initiated the ARPANET project when Taylor had. The article gives far too much focus on Licklider over others. --John R. Barberio talk, contribs 17:20, 7 September 2005 (UTC)

The reason the article didn't mention Bob Taylor is that he had almost nothing to do with the Internet. He certainly had no role at all in the development of the Internet (i.e. from '73 on). His only influence was an inspirational one, through the ARPANET - and when it comes to inspiration, his influence was less than that of Licklider - whom the article no longer mentions. Noel (talk) 22:37, 14 September 2005 (UTC)
Um... Bob Taylor got the funding and authorisation for the ARPANET project. How is that 'nothing to do with the Internet'. Also you simply haven't even read the current article before critisising it, Licklider gets a major mention. He even appears photographed. Can you please back off from this article for a while to cool down. --John R. Barberio talk, contribs 23:37, 14 September 2005 (UTC)
I was looking at another article (I have yet to make any changes to this one) when I made the Lick comment, and got confused.
As for Bob Taylor, I already said (look it up, it's only a few lines above) that his influence was "inspirational .. through the ARPANET" (the importance of which to the Internet, ironically, you seem determined to denigrate). He had nothing to do directly with the Internet, because Bob left DARPA for PARC in 1970 (another fact easily ascertainable from histories, had you bothered to consult one), well before work on TCP/IP started in '73. (He did have an indirect role through PARC's PUP work, but that's as tenuous as his ARPANet connection.) Noel (talk) 00:57, 15 September 2005 (UTC)
This article is not a history of TCP/IP. Otherwise we could excise Licklider too, and we seem to want to keep him. --John R. Barberio talk, contribs 10:53, 15 September 2005 (UTC)

Initial work on the rewrite is done see Talk:History_of_the_Internet/rewrite. Things yet to be done,

  • TCP/IP and the early internet. (including TCP/IP spread over UUCPnet, and europe)
  • Email, the 'killer app'.
  • Initial culture of the Internet. Usenet News. Email lists.
  • History of Standards and RFCs. IANA to ICANN, short history of the change.
  • Gopher to the World Wide Web, how the 'Global Library' idea finaly realized.
  • On line chat. Talk to IRC to Instant Messages.

Anyone willing, chip in on the page. Anything major changes on the current page, and I'll try to merge them in. --John R. Barberio talk, contribs 18:17, 7 September 2005 (UTC)

One good way to spot things we should include, is the page's What Links Here list. There will be articles here, where there is a referal or 'see more' link to History of the Internet. Such as on Gopher Protocol, Search Engine and BITNET. --John R. Barberio talk, contribs 19:49, 7 September 2005 (UTC)

Skeleton article layout, with headings ready to be filled. --John R. Barberio talk, contribs 23:48, 7 September 2005 (UTC)

The rewrite is prety much done. It's been reformed into a more readable format, with information grouped by subject. Technical detail has been removed in preference to summary and direction to main articles. I've moved some ARPANET specific information to the ARPANET page, if there is any other info that needs moving to other articles, now is the time to do so.

Once we've checked over the re-write, we can paste it over the current article. Unless someone has a major objection, or there are any corrections to be made to the page, it'll probably happen some time next week. --John R. Barberio talk, contribs 17:39, 9 September 2005 (UTC)


Hi John -- I've looked over the rewrite [2] and it looks fantastic, really very good. The main thrust of my remarks I think you've answered very well, by breaking things into subsections and making clear what's being discussed. Again, I think you've done a wonderful job here, congratulations. This seems like a great skeleton to build off of (although IMO it can also be considered "finished".) Sdedeo 18:11, 9 September 2005 (UTC)

Rewrite merged in

The rewrite has been merged in. Since no one had big issues with it, the NPOV tag has been removed. I think we could start working on this to get it to featured article standard. --John R. Barberio talk, contribs 21:23, 12 September 2005 (UTC)

Cites for ARPANet influence

Since you asked (above) for cites about the ARPANet influence, here they are.

No, I asked for cites to show that the ARPANet was sole cultural influence. --John R. Barberio talk, contribs 23:31, 14 September 2005 (UTC)
Which is something ("the ARPANET was sole cultural influence" - emphasis mine) I never said - see quotes from above, which I repeated just below. Please stop creating rhetorical straw-men which you can then knock down. Noel (talk) 00:05, 15 September 2005 (UTC)
This is not an issue about you, it is an issue about the article. The article was repesenting ARPANET as sole originator of internet culture. And that was what I required cites to support. --John R. Barberio talk, contribs 00:50, 15 September 2005 (UTC)
Balderdash. The last version of the article I touched contained such text as:
Parallel to the Internet, other networks were growing. .. like BITNET and CSNET. .. like the UUCP network. Commercial public access X.25 networks had also taken root in many countries. ...
The early e-mail system was not limited to the Internet. Gateway machines connected Internet SMTP e-mail with UUCP mail, BITNET, the Fidonet BBS network, and other services. Commercial e-mail providers such as CompuServe and The Source could also be reached. ...
The second most popular application of the early Internet was Usenet, a system of distributed discussion groups .. . Usenet had existed .. as an application of Unix computers connected by telephone lines via UUCP.
Hardly trying to write all others out of the history. Noel (talk) 02:00, 15 September 2005 (UTC)

(I'm putting them here at the bottom since you seem not to read comments further up, such as my I agree that the Internet absorbed some cultural influences from the UUCP world later on and Note that the Internet also absorbed sizeable user communities, and their social traditions, from other networks, such as BITNET and HEPNET from above.) Anyway, here we go:

  • Janet Abbate, Inventing the Internet is the standard academic history of the Internet. To give all it says on the topic of "where did today's Internet culture come from" would require reproducing a great deal of the book, so you should just get it and read it. (Have you ever seen it?) Here are some excerpts from a hurried look through:
    • the Internet is a product of its social environment. The Internet and its predecessor, the ARPANET, were created by [DARPA] (p. 2) - note no mention of any other "predecessor"
    • [the Internet was] a system that proved to have more appeal for potential "customers" .. than did the overtly commercial alternatives that appeared soon after. (p. 145)
    • On a book with 220 pages of text, the first mention of X.25 appears on page 154
    • UUCP and Usenet appear on page 200, and BITNET and Fidonet on 202
The book also mentions a long list of networking communities that are not yet listed in this article, including things such as the WELL (in San Francisco), AOL and its brethren Prodigy and Compuserve, etc. Some of these even had technical influences on the Internet of today; e.g. chat rooms came from AOL (but someone should research that AOL really were the first to have these before including it anywhere, I suspect some other system may have pioneered them). (BTW, AOL did not originate "instant messaging", though - the ARPANet had that in the 1970s. It only became very popular through the AOL system, though.)
Please feel free to provide extra info on these services in the comercialisation section. I had considered the WELL, while a huge influence on Californian network users, it was not a big influence on the net as a whole. --John R. Barberio talk, contribs 23:31, 14 September 2005 (UTC)
  • Katie Hafner, Where Wizards Stay Up Late: The Origins of the Internet, although a popular history, was carefully researched by a professional journalist. It takes a similar stance to Abbate's:
    • By the end of 1989, the ARPANet was gone. The NSFNet and the regional networks it had spawned had become the principal backbone [of the Internet]. (p. 256)
    • in dozens of ways, the [Internet] of 1994 still reflected the personalities and proclivities of those who built it. (p. 258) followed by a list of people, all of whom were associated with the ARPANET or early Internet
    • In a book of 265 pages of text, the first (and only) mention of UUCP comes on page 244, and X.25 is not mentioned at all

So there's plenty of substantiation in published works for the contention that the ARPANET was the principal source for the Internet (and not just on a technical level, that particular assertion is unquestioned).

This conclusion is easily verified by thinking about what the major applications running on the Internet are. (I looked for, but couldn't find, a table that ranked them all via, e.g. percentage of traffic, so this list is garnered from looking at a number of secondary sources, e.g. [3]) Almost all are either recent developments (such as peer-peer multimedia) of the Internet itself, or go back to the old ARPANET/Early-Internet community.

  • The WWW is the biggest, and that was a "recent" development of the Internet itself (it ran on TCP/IP from its earliest). Predecessors such as Gopher and WAIS also came out of the ARPA/Early-Internet community, although the basic ideas go back to Vannevar Bush's Memex.
  • Email, which seems to be second, is clearly an ARPAnet thing.
  • Peer-peer multimedia exchange is a recent Internet phenomenon.
  • VoIP goes back to early packet voice work on the ARPANET, and a key part of the technical nature of the Internet (separate protocols for IP and TCP) was in fact done in large part to support packet voice (although is has only recently become feasible to use it on any large scale).
  • Similary, "working at home" was a big hit with the early ARPAnet community.
  • IM was first done on the ARPAnet, but was made popular by AOL (not currently mentioned) as a predecessor.
  • Chat rooms are an outside thing, but whether AOL, or earlier, I don't know.
  • Multiplayer online games were first shown on the ARPAnet, with Mazewars
  • Newnews is now fading, but was a USENET thing

Yes, as I already agreed, other network communities brought their own cultures into the Internet when they joining, enriching and modifying it when they did so. But if you look at the relative contributions of each, there's no way you can make a case that any of them had as large a contribution as the ARPANET/Early-Internet community (remember, after 1983, when NCP was turned off, there was no ARPANet anymore, except as a (mostly-invisible) component of the early Internet).

So I have no problem mentioning them as later contributors to the Internet, but they are all secondary. If you want to claim otherwise, i.e. that the ARPANET is not the most significant influence on the Internet, I would in turn ask you for citations for any statements to that effect in major, well-researched publications.

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)

No. There are grounds for an opinion that the ARPANet is the principle source of Internet Culture. Your asertion that since the modern techs used to support internet culture are either new or ARPANet originated are not only spurious and ilogical (That the technologies currently used are ARPANET technology based is no reflection on the origin of the culture), but it is original research, and not suitable for the article.
As I pointed out with the growth numbers of UUCP, simply going by populace alone, the rest of the networks overwhelm ARPANET. I'm getting concerned that your input is showing a bias towards ARPANET. Can you please step back, since your involvement with ARPANET may colour your opinion.
I have provided cites to UUCP growth history. The Usenet article provides sufficiant cites to demonstrate historical significance of its input. If you need more citation that the culture of the Internet is hugely influenced by UUCP, then I suggest you read the Jargon File, and its colection of cultural references. Comercial use of near global scope networks also predates NSFNet's opening of the networks, which is why X.25 is significant both in a tech and culture basis. --John R. Barberio talk, contribs 23:31, 14 September 2005 (UTC)
I can provide citations to the content about the back side of the moon, but it will provide nothing useful on this topic. Citing figures about the growth of the UUCP proves nothing about the size of the influence of the UUCP network on the Internet. Your comment about the Jargon file made me laugh - that was originally an ARPANet thing, which the UUCP guys added to. And for you to complain about "original research" is a joke of high order indeed: what on earth do you think your info on "growth numbers of UUCP" is, but an attempt to sustantiate your 'novel historical interpretation' through doing original research?
I repeat again that X.25, which wasn't even defined until 1976, had zero technical influence on TCP, which dates to 1974 - check out RFC 675, from December, '74. (Why do I have to keep repeating this?)
I had next to nothing to do with the ARPANET; I did use it, but then so did hundreds of thousands of other people. I show a bias towards the ARPANet because I happen to like historical fact. I was involved with the Internet - which this article purports to be a history of.
And I will ask you, again, for citations for any statements in major, well-researched publications that state that the ARPANET is not the most significant influence on the Internet. Noel (talk) 00:05, 15 September 2005 (UTC)
This is the major problem Noel. You want to claim that ARPANET was the most significant influence on the internet. You have no direct cites to support that. I have cites that lots of internet culture came from elsewhere, that I Have Already Stated. All evidence I have read and cited supports that the Internet Culture came from a melting pot of the cultures from Usenet, ARPANET/NSFNet, Fido, Compuserve/AOL... It seems absurd to then claim that to support that I have to find a cite saying 'ARPANET is not the most significant influence on the Internet' to support this. I do not need to cite to support an absence of evidence. (Incidently, yes, I have read 'Where Wizards Stay Up Late' et al. It just seems I came away from them with substantialy diferent impresions than you did.)
Again, your implication that networks are insignificant because they had no impact on technical development is spurious. This article is not solely about TCP/IP development, or the technologies currently used. It would be strange if an article on the history of steam ships did not mention paddle boats because propeller boats 'won'. --John R. Barberio talk, contribs 00:50, 15 September 2005 (UTC)
I saw some discussion above about how the article mixes technical development with the broader development, and how it would be good to separate them. So perhaps it would help to do that.
And the ARPANet was the most significant external influence on the Internet. (The most significant thing on the Internet these days, the Web, was a product of the Internet itself, it didn't come from any outside source, other than not-widely-deployed ideas such as Memex and Hypertext.)
I'm not disputing that lots of cultural influences came from other networks. However, that is not necessarily inconsistent with the statement that the most important influence on the Internet was the ARPANet/early-Internet (the two merge into one another so seamlessly in the '82-'85 time period, as all the ARPANet hosts transitioned to TCP after NCP was disabled in early '83 that it's hard to distinguish them).
(BTW, since the Network News Transfer Protocol, RFC 977, dates from February 1986, a lot of "Usenet" history from not too long after that date is actually Internet history, as Usenet became accessible to everyone on the Internet, and the pre-existing Internet user community started contributing.)
I have provided several quotations (above) which testify to the primary importance of the ARPANet to the Internet. And there is plenty more; e.g. Abbate's book has several chapters on the ARPANET - and about three pages on UUCP/Usenet.
So I will ask you, yet again, to provide us material from major, well-researched publications which dispute the conclusion that the ARPANet was the primary influence. Noel (talk) 02:00, 15 September 2005 (UTC)
It is a POV that the ARPANET was the most significant influence on the internet. There is lots of evidence to sugest otherwise. Without significant cites and a consensus to support that ARPANET was the sole major contributor, this view can not and will not be expressed in this article. Yes, the ARPANET was a primary source of the technical history of the Internet. This article is not a technical history, and other issues of popualtion and culture are to be taken into account.
The article currently states that ARPANET was the most significant technical contributor to the internet. All your cites provide support for ARPANET being the most significant technical contribution. They do not support ARPANET being the primary culturaly influence. Nor do they support ARPANET being the most popular network.
Can I also ask you to please respect the contraversy tag, and discuss significant changes to the article before you make them. If you check above you will find that we had a discussed and concencus agreeed rewrite of the article that took place on a temporary page with time for people to make changes. Unanamous editing is not a good thing on Wikipedia.
Incidently, you are wrong on several counts over things you have raised here. E-Mail predates ARPANET. Compuserve offered the first online chat rooms. USENET predates RFC 977, and dates back to 1979. Claim that 'Working From Home' came from ARPANET is baseless speculation or anecdotal, you could work from home over X.25 too. MAZEWARS was developed at the NASA Ames Research Center and Xerox PARC in 1974 as part of the development of Ethernet networks, not ARPANET. [4] (This should go in the article! Network games developement seems a prety major culture thing to me.)
Can I again ask you to back off from this article. You are getting aggresive. --John R. Barberio talk, contribs 10:41, 15 September 2005 (UTC)
If this article is not the technical history of the Internet, then what is it? We do not (yet) have a separate Technial history of the Internet article.
A Technical History focuses on the technical developments. Since technical details are better served on the pages of the technologies, the detail of technical history belongs there too. The primary focus of this article has been the groups of people who furthered the progress towards the Internet. --John R. Barberio talk, contribs 20:08, 15 September 2005 (UTC)
Your comment about "[no] significant cites" is bizarre: how exactly do describe quotations like the [Internet] of 1994 still reflected the personalities and proclivities of those who built it, from one of the major histories of the Internet, not count?
Please note that quote is not the [Internet] of 1994 still reflected the personalities and proclivities of ARPANET. You're inteperatation of it as such is a POV. --John R. Barberio talk, contribs 20:08, 15 September 2005 (UTC)
Again, your lack of knowlegde of this subject you claim to have a better understanding of is showing. If you look at the book, the paragraph that quotation is from goes on to discuss the following people: Roberts, Heart, Crocker, Kahn, Cerf: all of whom were associated with the ARPANET. So, once again, you're shown your lack of knowledge, and shot yourself in the foot thereby.
No doubt you'll try and find some way of wiggling out of this too. Once again, I stand ready to ask Katie herself what it was she meant by this. Want to make a bet on who she's going to say has it right? Noel (talk) 04:30, 19 September 2005 (UTC)
Also, your are again showing your lack of knowledge of the history of networks with your Mazewars comments. The original Mazewars, the one the Xerox/Ethernet one copied, was developed on the Imlacs, some years before the Xerox one. The Imlac one ran years before the Xerox one was even written; see this page. The Imlac one ran over the ARPANET (not well, because of delay/bandwidth issues).
And I put nothing about "working from home" in the article, as I have not had time to look for discussion of this in the histories; I have no idea where you got that idea - the article doesn't even contain the word "home"! (But it would have been hard to do it over X.25 first, when X.25 was only developed in 1976, years after the ARPANet was operational.)
Your comments about email make no sense at all - where on earth do you think the article said that the ARPANet had the first email? (It did have the first network email.) All my text said was The applications run on the Internet in its earliest days were exactly those already running on the ARPANET: email, and later: e-mail actually predates the Internet; in fact, existing e-mail over the ARPANET was a crucial tool (and I think that text pre-dates my revision, actually).
Ditto for Usenet - I cannot imagine what in my revision you read to get the impression you state, since I wrote The Internet quickly developed a native mechanism for carrying Usenet material .. in February 1986 - and since a few lines above, it says UUCP was done In 1979, clearly the Internet taking up News was years after it became popular through UUCP.
Please make sure you read the text that's actually written before you throw all sorts of confused charges around. Noel (talk) 17:15, 15 September 2005 (UTC)
I was responding to what you have brought up on this Talk page, which I thought was obvious.
Also, the Mazewar article you cite does not mention ARPANET at all.
Can I also remind you that I had to strugle to get you to accept an expanded mention of UUCP at all?
--John R. Barberio talk, contribs 19:52, 15 September 2005 (UTC)

Mazewar

The reason:

the Mazewar article you cite does not mention ARPANET at all

is because the page I cited was support for this statement by me:

The original Mazewars, the one the Xerox/Ethernet one copied, was developed on the Imlacs .. The Imlac one ran years before the Xerox one was even written; see this page.

statement, and indeed and the page cited, by the person who actually wrote the Imlac Mazewar (Steve Colley) says "I believe the first Maze and the two machine version happened in '73" - i.e. well before the Xerox one, which according to this page, by the person (Jim Guyton) who actually wrote the Xerox version, happened in '77 or so (based on his hiring sometime around late '76, and his report that "Over the next year I spent a lot of nights working on getting Mazewar running". That page also notes that the Xerox one was inspired by, and based on, the Imlac/PDP-10 one.

As to the ARPANet usage, this page says "When Greg Thompson appeared at [MIT] in 1974, he brought with him .. some of the Imlac games that he had encountered at NASA Ames. Among these was Maze." .. "Early in its life, we played Maze across the Arpanet with players at USC" which shows that play across the ARPANet preceded the Xerox version (which was '77 at the earliest, as indicated above). Noel (talk) 14:17, 16 September 2005 (UTC)