This article is within the scope of WikiProject Internet, a collaborative effort to improve the coverage of the
Internet on Wikipedia. If you would like to participate, please visit the project page, where you can join
the discussion and see a list of open tasks.InternetWikipedia:WikiProject InternetTemplate:WikiProject InternetInternet articles
This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of
computers,
computing, and
information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join
the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
This article is within the scope of WikiProject Telecommunications, a collaborative effort to improve the coverage of
Telecommunications on Wikipedia. If you would like to participate, please visit the project page, where you can join
the discussion and see a list of open tasks.TelecommunicationsWikipedia:WikiProject TelecommunicationsTemplate:WikiProject TelecommunicationsTelecommunications articles
Protocol Wars is a former
featured article candidate. Please view the links under Article milestones below to see why the nomination failed. For older candidates, please check the
archive.
The text of the entry was: Did you know ... that
Vint Cerf performed a striptease in a three-piece suit during the Protocol Wars?
Current status: Former featured article candidate, current good article
Notability
I placed a notability template on this article. It may be notable (after all, I remember the issues and they seemed important at the time!) but it will need more
WP:RS. At the very least it appears the article is misnamed. Only one of the current references mentions a "protocol war". That is ref 7, the Roger Scantlebury interview. The first three refs don't mention wars at all, and refs 4, 5, 6 and 8 all speak of the "standards war" which seems a more appropriate title to me, but should probably contain a qualifier too, to show which standards war we are talking about. --
Sirfurboy (
talk)
22:55, 5 February 2020 (UTC)reply
This topic is covered is several reliable sources and commonly known as the Protocol Wars. It cuts across several existing articles and is currently not covered on Wikipedia. Let's see how the article develops. Improvements welcome.
Whizz40 (
talk)
04:05, 6 February 2020 (UTC)reply
Hi
Whizz40, per
WP:DEMOLISH, I am indeed happy to watch how this develops, and indeed to help if I can. I won't be taking any precipitate action to undermine what you are doing, but I do think the title needs more thought. A rename is an easy thing to do at any time, so we should think carefully about what is the most approriate title.
You say the topic is commonly known as "the Protocol Wars". Although it is referred to by that name, my analysis of the sources you had yesterday was clear that "standards war" was the more common term, and is the one I would generally use. Protocols are part of the standards, of course, but they are not the only part. There is also a whole ideological war about the standards themselves, particularly at layer interfaces. Additionally, I am not sure this article should necessarily take the unqualified name "Protocol Wars" when there have been many protocol wars, both within the networking community and without. Another protocol war is described in a 2015 article in the IPJ, describing which transport layer protocols should be used in TCP/IP, and that article is titled TCP protocol wars. That is disambiguated from this article's topic by people generally referring to this as the "Internet-OSI Standards War". That would be my preference for the page title, but if you think it is not
WP:CONCISE then how about "Standards war (networking)"? I wonder if
kvng (
talk·contribs), as another subject specialist, might have a view? --
Sirfurboy (
talk)
12:18, 6 February 2020 (UTC)reply
I hope I haven't been misunderstood. Mine was not a rhetorical or snarky question. I don't have a formed opinion about whether this is notable. I was requesting someone highlight evidence supporting notability; bthere are a lot of sources cited. ~
Kvng (
talk)
13:39, 7 February 2020 (UTC)reply
I too am happy to keep discussing if you wish. There was a standards war - I was just checking that (a) it was notable and (b) we are calling it the right thing. So happy for you to withdraw the deletion if you wish, or we can allow deletion but maybe come back and try again to create the article in the future. --
Sirfurboy (
talk)
14:21, 7 February 2020 (UTC)reply
There is some discussion of a standards war in
OSI model and that is clearly the topic being covered by this article. My assumption at this point is this is a notable subject. I will find time later to do a
WP:BEFORE-style assessment. If protocol wars is not the right title, we can always rename. ~
Kvng (
talk)
14:27, 7 February 2020 (UTC)reply
"Standards Wars"(PDF). Student project at Department of Computer Science and Engineering, University of Washington. 2006.
Thank you for these. Citations 1 and 2 are both very good, but they have the same author. Is there perhaps one more we can find? Sorry! --
Sirfurboy (
talk)
15:14, 7 February 2020 (UTC)reply
Ok, Riaz, 2016 does not seem to mention protocol wars or the standards war and the last ref is more of a passing mention, But Davies, 2010 has a significant section, and speaks of Protocol Wars (plural, stating there was more than one war). That is a good ref, but then it suggests something subtly different to the first three refs which are all focused on the OSI-Internet standards war. Should this article be about that standards war in particular, or about the protocol wars in general (which include the OSI standards war?)
As I said, the IEEE sources are good. Ideally we would have sources from different authors, but either of the Russell sources are good because the IEEE is a respected source and these are secondary sources. Wikipedia, as a tertiary source, prefers secondary sources, so I think they are good. Not sure if
kvng (
talk·contribs) agrees, but hopefully we all agree on that.
The paper from University of Washington is interesting. It is a student's final project. It is not clear to me the basis on which it is published though. Is this a submitted and corrected paper that the University published? Is it a sample project that they put up for student guidance? or is it published in some other way? A student project may contain errors, but the paper does do two things: (1) it supports the view that this subject is notable, and (2) it provides a bibliography so should lead to more sources (although whether those will be secondary sources is unclear).
So I think we are getting closer. Something does appear to be notable here, but we need to be clear on that question of whether this article focuses only on the OSI standards war or whether it is a description of a series of wars as per Davies, 2010.
One possible approach to get started, perhaps we can define an outline of the topic area and then identify the scope for this article and the article title? For example:
Standards in computer science
Protocol Wars
Internet-OSI Standards War
Open to other approaches as well. I'm still open to deleting this article on the basis the content is better covered in other articles to which it is relevant. Is there a concern this article will become unwieldy and poor quality, duplicating other content or orphaning content that would be better found on other articles to which it relates?
Whizz40 (
talk)
08:58, 10 February 2020 (UTC)reply
It could be discussed in
Internet protocol suite or
OSI model but since it pertains to both, it would be hard to decide where and hard to avoid duplication. As long as we're happy notability requirements are met, I think it better to cover it in this separate article. ~
Kvng (
talk)
18:16, 11 February 2020 (UTC)reply
Okay so thinking about the article title. I think what started as datagrams vs virtual circuits ultimately became the Internet-OSI Standards War. If the latter is the main topic and the former can be covered in the article as background then I can see no reason not to rename the article to Internet-OSI Standards War or something else. Any other views on the best article title? Along the way there were pioneers (RAND, NPL, ARPA, CYCLADES) vs PTTs (X.25), Open (TCP/IP on Unix) vs proprietary (IBM and DEC), US DoD/NSF/NASA/DoE (TCP/IP) vs Europe (X.25), IP over X.25 vs either/or, culminating in Internet-OSI (including the US Dept of Commerce supporting the latter).
Whizz40 (
talk)
08:53, 11 February 2020 (UTC)reply
I would leave the title alone for now and concentrate on developing the article. When we're happy with the content, it should be clear whether the title needs to change. Changing the title once is not a big deal. It can get messy if we end up changing it multiple times. What we have now is a good working title. ~
Kvng (
talk)
18:21, 11 February 2020 (UTC)reply
Made a start, welcome editing from others and views on quality and approach. I am short on time and there is plenty to cover but if anyone thinks it is heading in the right direction and wants to help improve and put it up for DYK then please do (might just make the criteria with input form others, started on Feb 5th or 5x expansion).
Whizz40 (
talk)
11:48, 13 February 2020 (UTC)reply
Initial edits to the article are done. If anyone has any suggestions for how to incorporate the images mentioned in the DYK discussion below please do let me know.
Whizz40 (
talk)
07:18, 20 February 2020 (UTC)reply
Hi there, you put the article up for AfD instead of speedy in error. No big issue. AfD takes a week but otherwise it should be uncontroversial. If you want a speedy deletion, the following tags need to be placed on the page, replacing all the page content with these:
As this is an author request deletion, I don't think I can do it (the Db-self template not being correct for me), so if you want a speedy deletion, just paste those onto the page. Thanks, and thanks for all the new information at History of the Internet too. --
Sirfurboy (
talk)
11:25, 7 February 2020 (UTC)reply
The following is an archived discussion of the DYK nomination of the article below. Please do not modify this page. Subsequent comments should be made on the appropriate discussion page (such as
this nomination's talk page,
the article's talk page or
Wikipedia talk:Did you know), unless there is consensus to re-open the discussion at this page. No further edits should be made to this page.
... that
Larry Roberts described the
ARPANET model as "oversold" during the Protocol Wars? Source:
page 1309, "Hopefully, the extensive publications on the ARPANET have not oversold the particular variety of packet switching used in this first major network experiment. "
ALT1:... that
Vint Cerf performed a striptease in a three-piece suit during the Protocol Wars? Source:
page 55, "As Cerf addressed the IETF, he slowly removed the layers of his signature three-piece suit, performing a striptease that revealed a Tshirt: “IP on Everything.”"
Comment: This article was created on Feb 5th. At the start of the day on Feb 6th, the article was 3,569 bytes, at the end of the day on Feb 12th/start of day Feb 13th it was 21,595 bytes, 5x expansion and I posted on the talk page
here on the 13th to seek assistance with the DYK nomination and hit submit on the DYK nomination on the 14th due to real world responsibilities. The article is rated C-class by Wiki Project Computing.
Hi @
Whizz40:, taking a look at this as you requested. This is far from my area of expertise, so can you say why the first hook would be interesting to a broad audience? The second hook is certainly catchy but the first isn't standing out to the casual reader.
Kosack (
talk)
09:16, 16 February 2020 (UTC)reply
Thank you Kosack. Agree the second hook is catchy. Larry Roberts played a central role in building the ARPANET, which was the forerunner of the Internet. He 'switched sides' in the Protocol Wars when he later built
Telenet using a different approach. We could rephrase the first hook as:
Review as follows. Article was created on 5 February, nomination was marginally outside the window but I think we can allow a little slack there. Article is well over the minimum required prose size and has no copyvio concerns. ALT1 and the revised original hook are catchy enough for promotion and have inline citations to reliable sources. QPQ is also complete. @
Whizz40:, from your comment above, you're happy with ALT1 being the main hook for promotion?
Kosack (
talk)
16:42, 16 February 2020 (UTC)reply
Yes, agreed, let's go with ALT1 for the DYK nomination. Vint Cerf is better known, and this hook relates to the Internet-OSI Standard War which is the main component of the Protocol Wars. Thank you Kosack.
Whizz40 (
talk)
01:22, 17 February 2020 (UTC)reply
In case anyone reviewing DYK or the Talk page is more familiar with this than me, there are three images which could be included in the article but I'm not sure about the fair use or copyright requirements to reproduce them. These images would be for further article development rather than the DYK nomination, but any suggestions on how to include them would gratefully received:
The Davies graphic from the 1975 Computer world.[1]
a Vint Cerf "IP on everything" t-shirt photo for which reference 70 (on page 60) in this source mentions a link to the image but that is not working.[2] However, there are pictures found on the web
here.
@
Whizz40:, I'm happy to promote this now. If you do add an image before it reaches the main page, I'll double check that too. For now, this is good to go with ALT1 as the main hook.
Kosack (
talk)
05:24, 20 February 2020 (UTC)reply
References
^Enterprise, I. D. G. (1975-10-22).
Computerworld. IDG Enterprise. pp. 17–18.
@
Cwmhiraeth:, @
Kosack: Thinking about the images discussed on the DYK nomination, I plan to be bold and add these based on fair use rationale. I should be able to do this tonight before if goes on the main page, but Cwmhiraeth, happy if you can move it back in the queue to allow time to do this and for a review. My plan for developing the article is to get plenty of readers and editors looking at it when it goes on the main page and I think the images may hold people's attention on the article for longer. This might bring out some edits about aspects that I had not anticipated or read about because it is a broad topic that covers many years. With this and images added, I am thinking about working towards GA.
Whizz40 (
talk)
10:07, 21 February 2020 (UTC)reply
OK, probably not, but what does PTT stand for? Standard style would be to define it on first use. Public Telephone and Telegraph? Telegraph? Surely the Protocol Wars did not involve Morse code?
IAmNitpicking (
talk)
14:46, 26 February 2020 (UTC)reply
Thanks to
Whizz40 (
talk·contribs) for providing a very good article looking at this aspect of Internet history. This has grown far quicker than I expected, thanks to the Whizz40's efforts. I have read through and provided a copy edit, explained in the edit summary. Nothing major, and I am not precious about the words I added to the lead - please revert or change those as you see fit. My thoughts about the article at this stage are as follows:
The lead can now be expanded a little to make it clear what the actual war was about. The detail is in the history, but the majority of readers will only read the lead and scan the remainder, so summarising the key debate might make sense;
The history is all good stuff but takes a while to find any actual war. Perhaps the battles shuld be more prominent (but it is there - I like the addition of the Vint Cerf T shirt recreation image illustratibg that; and
On reading this, I still feel that the key battle ground is probably the Internet/OSI standards war, so I still think that may be the better article name, but would value the thoughts of other editors. --
Sirfurboy🏄 (
talk)
10:57, 28 February 2020 (UTC)reply
SNA is mentioned twice in the article, along with
DECnet and
XNS. The history of SNA is covered at
SNA § History. In terms of this article, what are the summary points we want to include on SNA and the other proprietary standards?
Whizz40 (
talk)
14:19, 3 March 2020 (UTC)reply
I consider the article overly "internet" centric, neglecting packet switching networking done in the Mainframe World, starting at least with the
IBM System/360 and the accompanying
IBM 270x Communication Controllers. --
Poc (
talk)
15:22, 11 July 2020 (UTC)reply
This isn't the
packet switching article, it's an article about the battle between the Internet protocol suite and the OSI protocol suite, both of which were not vendor-specific protocol suites, with vendor protocol suites, including but not limited to IBM's SNA, being of lesser relevance to the article's topic.
Guy Harris (
talk)
18:09, 11 July 2020 (UTC)reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Note: this represents where the article stands relative to the
Good Article criteria. Criteria marked are unassessed
Pioneers vs. PTTs: I recommend you add a new first sentence to provide broad context, before zooming in on Baran and Davies. Something like this, stolen/adapted from
History of the Internet: With limited exceptions, the earliest computers were connected directly to terminals used by individual users, typically in the same building or site. Wide area networks (WANs) emerged during the 1950s and became established during the 1960s, using telephone lines to form connections between physically distant computers. (I made up that last part about the telephone lines, not sure if it's technically correct but I'm trying to understand why Baran and Davies were worried about the PTTs.)
ARPANET: When you first mention ARPANET, I think it would be appropriate to spell out the acronym and say briefly what it is; you say that it was intended to enable time-sharing of computers, but it think it would help to mention that it was a DoD program.
Technical aspects: I like Postel's ancient mystics quote, it's colorful. I'm confused on what he meant though - was he sarcastically mocking the OSI model for having too many layers? Or was he saying that TCP/IP did not have enough layers?
He was mocking the number of layers in OSI. Clarified that seven was more layers than the network engineering community was anticipating.
Whizz40 (
talk)
12:56, 16 May 2020 (UTC)reply
Practical and commercial aspects: The sentence In 1990, CERN established a transatlantic TCP/IP link with Cornell University in the United States and academic institutions in some European countries and organizations had adopted or signaled their acceptance of TCP/IP is confusing, I think you either need to remove "and organizations" or move it before "in some European countries". Right now it reads like some of the academic institutions are in European countries and some are in organizations.
Images: Outstanding use of images, you found some very engaging ones and fair use rationales are present. When I was trying to figure out the difference between datagrams and virtual circuits a diagram would have helped, but I poked around some other pages and couldn't find one. Do you think a diagram of the OSI or TCP/IP models would be appropriate? Something like
File:Internet layering.svg?
Spot checks: Spot checked references #11, 22, 64. For #64, which part of the sentence is it supporting? I couldn't find anything in there about Cornell on a link being established in 1990.
Fate of OSI: Does anyone still use OSI? I remember having to memorize the seven layers in a networking class a few years ago, but for the life of me I can't remember why. If it's still around, would be worth mentioning in the article.
The OSI model is still used as a teaching tool, and there are some OSI protocols in existence. Added more context to the OSI Reference Model section and added a Legacy section.
Whizz40 (
talk)
12:56, 16 May 2020 (UTC)reply
That's all I have! Thank you for this article, it was fun to read. I'll place it on hold for seven days so you can work through the comments above. If you disagree with any of them, let me know. For some, like explaining ARPANET and Pouzon, maybe it's not necessary since the reader can just use a wikilink to learn more; I tend to think articles should be self-contained as much as possible, but maybe I'm wrong and it's just unnecessary detail. --
Cerebellum (
talk)
02:57, 16 May 2020 (UTC)reply
Thanks for the improvements, I'd still like to see page numbers for the books but it's not the end of the world, since the Google Books links are for the relevant page. Pass as GA. --
Cerebellum (
talk)
01:22, 23 May 2020 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
01:00, 14 May 2020 (UTC)
1982 BSD DECNET
DECNET for at least one of the 4.1c BSD variants exists "in the wild". ("varient" as in tapes with a "4.1 BSD" label sticker on them can have different contents as 4.1* BSD appears to have been some sort of "rolling release")
The DECNET code is also in the BSD SCCS version control archives Those SCCS archives were converted to the fossil version control system and available on online at
Files in sys/deprecated/netdecnet/. Note that the "deprecated" directory is where it ended up after being in a "current" directory back in 1982. The same DECNET code is available on the CSRG ISO #1 in the directory 4.1c.1/sys/netdecnet. --
Jamplevia (
talk)
01:53, 10 June 2023 (UTC)reply
I got as far as building a working RA81 disk image with 4.1c.1 BSD, a version that is really a backup tape of source code. I grafted the source code onto binaries from a newer version of BSD and then downgraded the binaries by recompiling them. All the executables including the kernel are authentic in that they are the result of compiling the source code. However, configuration files like /etc/passwd are not authentic because they aren't on the backup tape. The result has the DECNET source code in /sys/netdecnet but I have yet to get that to compile.
In an article on protocol wars, missing to clarify conspicuously that the NPL Davies' model was datagrams and that of Khan for ARPANET was virtual circuits, would AFAIAC be missing a major information. In order to deal with your concern that the caption shouldn't bee too technical, I will delete the "connectionless" and "connection-oriented" mentions (likely to be less explicit for many readers than "datagrams" and "virtual circuits" in the context of the two major packet switching models). From that, please discuss what you wish before just deleting what I find appropriate for a good quality article.
RD2017 (
talk)
10:52, 31 October 2023 (UTC)reply
The datagram vs virtual circuit confrontation (DG vs VC) has been in my understanding one of the summums of the "protocol wars". Understanding this confrontation requires IMHO to know (a) how and why operators of the public packet switching networks of the 80s unanimously chose VCs for the CCITT X.25, (b) how and why DARPA chose DGs for the TCP/IP of the Internet, (c) and how and why the IETF Internet overwhelmed in the 90s the worldwide X.25 network.
Agreed, the text I proposed for (a) deserves to be improved and preferably shortened, and at least some of its substance is welcome in the X.25 article.
Yet, some coverage of (a) remains IMHO necessary in this article. If other’s comments don't object too much, I envision working again on the subject.
This article doesn't need all of the information added about the process through which point a came about. It's too much detail for readers who already have plenty to get through on this topic. This article already has a paragraph describing how "the X.25 standard was agreed by the CCITT in 1976" and an image of the main contributors, both in the section on
TCP vs CYCLADES and INWG vs X.25. For readers who are looking for more detail on how that came about, we can provide a link to another article, per
WP:Summary style; or, if the topic is notable, then a new article could be started covering that topic.
Whizz40 (
talk)
15:16, 11 November 2023 (UTC)reply
One of my recent contribution tended (a) to find and include direct references when available, (b) to prefer freely reachable references to those that need to spend money to read them, (c) for each assertion, to limit the number of references to what is enough to be convincing (readers' time is precious).
To cover Whizz40's point that a reference to the (well-known) seminal TCP paper of 1974 doesn't per se proves that it is indeed seminal, a reference to the first TCP standard RFC (i.e. RFC 793 of 1981, freely accessible) is IMHO largely sufficient for this (it says “TCP is based on concepts first described by Cerf and Kahn in [1]”).
To legitimate the next sentence of the article, i.e., “The paper drew upon and extended their prior research, developed in collaboration and competition with other American, British and French researchers”, the quotation of the paper itself I had selected is IMHO appropriate and sufficient: "The authors wish to thank a number of colleagues for helpful comments during early discussions of international network protocols, especially R. Metcalfe, R. Scantlebury, D. Walden, and H. Zimmerman; D. Davies and L. Pouzin who constructively commented on the fragmentation and accounting issues; and S. Crocker who commented on the creation and destruction of associations”.
Whizz40 (by mistake I suppose) replaced it with a longer citation containing irrelevant text, namely '... [6] R. Despres, "A packet switching network with graceful saturated operation," in Computer Communications: Impacts and Implications, S. Winkler, Ed. Washington, D.C., 1972, pp. 345-351.)'.
Thank you for your comments. On your first point, RFC 793 would perhaps be considered a primary source. Per
WP:RS, reliable secondary sources are preferred. Therefore, I restored these to the article. On your second point, I have made this change in the article. When I get time, I'll add links to the key primary sources in the References section.
Whizz40 (
talk)
20:40, 17 November 2023 (UTC)reply