Talk:Computer network

Ngriffeth and Zac67 have made some lead improvements but there are no useful edit summaries to explain what problem these changes are intended to address. The new lead is not a significant improvement as far as I can tell. It does have a new citation but the lead is supposed to summarize the body and, in a well developed article, the citations should be in the body and therefore are not necessarily wanted in the lead. ~Kvng (talk) 15:02, 3 June 2025 (UTC)

The biggest problem: The article needs considerable reworking. For that reason a summary of what’s here would be extremely difficult and not that helpful to a reader.

The first problem with the lead itself: the lead emphasizes resource sharing, but that’s a high level function that not all networks provide and not all users use. Communication between devices is the basic function of computer networks, so I wanted to emphasize that rather than resource sharing.

Second problem: Communication protocols are mentioned without any preface explaining what they are. It’s simple enough to say they are rules for communicating, so I added that.

Third problem: Telecommunications networks — ie phone networks -- are a special case (albeit perhaps the first network). I don’t know to what extent they even exist any more in developed countries - Verizon, for example, has replaced almost all of its networks in the mid-Atlantic with fiber networks. I’m not sure what the protocol stack is over fiber, but it’s likely not much like the old telecommunications networks. So I think it’s best not to mention them this early.

Speaking of re-working, the history section, while extensive, is confusing because it interleaves multiple tracks on which networks were developed. Until the 1980’s, the ARPANet/internet was almost exclusively American, and Europe was taking a different approach, the ISO OSI 7-layer architecture, which was eventually sort of retrofitted as a 4-layer stack onto the Internet to help explain what was going on. I would suggest separating out developments in the Internet, in the European approach, and the AT&T’s telecommunications network.

Military networks can’t really be separated from the Internet, since it was initially defined for a survivable military network, so that should be made clear.

I hope to clean this up before long. Ngriffeth (talk) 21:20, 16 June 2025 (UTC)

I need to take back my comment about telecom networks, this is a more complicated issue of usage. When I worked at Bell Labs, telecommunications networks and the Internet were viewed as altogether different beasts, but that was probably a parochial view. The great thing about editing Wikipedia is that it forces you to back away from your myopic view of the world and try to see it as a whole! Ngriffeth (talk) 21:37, 16 June 2025 (UTC)

When I worked at Bell Labs, telecommunications networks and the Internet were viewed as altogether different beasts, but that was probably a parochial view. Yup. Once phone networks went digital, especially with stored-program switching equipment, they were rather similar to computer networks, albeit with many of their own protocol stacks - and various copper and fiber communication networks originated to carry digitized voice calls were also used for digital computer communications. Guy Harris (talk) 00:00, 17 June 2025 (UTC)

Remember that we also have Telecommunications network which is arguably a superset of Computer network. Let's try to minimize overlap between these two or at least make clear where the overlap exists. ~Kvng (talk) 14:11, 17 June 2025 (UTC)

Thanks, I am looking at it. And yes, I agree that Telecommunications network includes Computer network. Also the article on Telecommunications is interesting and its style is nice. Ngriffeth (talk) 20:09, 19 June 2025 (UTC)

The telephone I mostly use is a 64-bit multiprocessor handheld UNIX box (well, iOS, unlike macOS, hasn't been run through the Single UNIX Specification test suite, so more Unix-like) that probably sends a lot of phone cals over an IP network (VoLTE, which is, as I understand it, is VoIP over LTE for the first hop), so more and more telecommunications networks are also computer networks. Guy Harris (talk) 22:04, 19 June 2025 (UTC)

That's true if you use the original definition of telecommunications network where it was all about phone calls and the signaling required to support them. The lead of telecommunications network contains a broader definition, exchanging messages. ~Kvng (talk) 15:46, 20 June 2025 (UTC)

By "so more and more telecommunications networks are also computer networks" I mean that more and more of the end nodes of telecommunications networks are devices that could be considered computers in some sense (even IoT devices such as smart meters have microprocessors in them), and more and more of the protocols of telecommunications networks are or are similar to computer network protocols, and more and more of the interior nodes of telecommunications networks are computers doing packet switching.

The example I gave, the voice part of the telephone network, was a telecommunications network but was clearly not a computer network for a long time (for one thing, it predated computing devices that communicated) that has, to a large degree turned into a computer network.

As for the broader sense of telecommunication network:

The introductory paragraph of telecommunications network says

A telecommunications network is a group of nodes interconnected by telecommunications links that are used to exchange messages between the nodes. The links may use a variety of technologies based on the methodologies of circuit switching, message switching, or packet switching, to pass messages and signals.

The introductory paragraph of computer network says

A computer network is a collection of communicating computers and other devices, such as printers and smart phones. In order to communicate, the computers and devices must be connected by wired media like copper cables, optical fibers, or by wireless communication. The devices may be connected in a variety of network topologies. In order to communicate over the network, computers use agreed-on rules, called communication protocols, over whatever medium is used.^[1][2]

The differences I see between the two leads are:

1. the telecommunications network lead speaks of the purpose of the network - "exchanging messages";
2. the computer network lead speaks of the components of the network - "computers and other devices".

I think the low-level purpose of a computer network could be seen as "exchanging messages"; the higher-level purposes such as "sharing resources" are achieved by exchanging messages.

So are computer networks just telecommunication networks where the nodes are computers - by some definition of "computer" - and computer peripherals? Guy Harris (talk) 20:40, 20 June 2025 (UTC)

Interesting question. The problem with giving the purpose of a computer network is that there are so many purposes - whatever a network architect can dream up. Maybe the purpose should be stated as “enabling the computers to be used together on one or more applications or services." Ngriffeth (talk) 15:34, 2 July 2025 (UTC)

The opening paragraph of the article on telecommunications networks doesn’t distinguish them adequately from computer networks:

A telecommunications network is a group of nodes interconnected by telecommunications links that are used to exchange messages between the nodes. The links may use a variety of technologies based on the methodologies of circuit switching, message switching, or packet switching, to pass messages and signals.

Perhaps telecommunications networks are more general, not requiring computers (up until the 1960’s, the telephone system was controlled by mechanical switches, not computers). Ngriffeth (talk) 21:43, 16 June 2025 (UTC)

the telephone system was controlled by mechanical switches, not computers ...and was mainly carrying analog voice on the data plane. Once the "voice" part went digital, the digital parts got used for digital communication as well (ISDN, T-carrier, E-carrier, SONET/SDH, etc.), and the switches were computerized and communicated with each other over their own digital protocol, the network was definitely similar.

There may be characteristics (circuit-switching vs. packet-switching, local vs. wide-area, etc.) that distinguished telecom network from LANs, at least at one point, but the two types of networking have converged to a large degree. Guy Harris (talk) 00:05, 17 June 2025 (UTC)

Agreed. As I review these articles, I am more and more convinced that they are now the same thing, Ngriffeth (talk) 18:58, 24 October 2025 (UTC)

The “Network science” sidebar seems unfair to the general reader, because it provides so many rabbit holes that are irrelevant to computer networks. Is there a process to go through before removing or changing a sidebar?

Also, I spent some time working on a sidebar for “Computer network” since that seems like a big enough topic that a sidebar would actually be useful to someone. But researching how to create sidebars ran me into a wikipedia policy of discouraging sidebars in the lead, and in fact it seems that sidebars are of questionable use. I agree with that policy. So… how should I proceed? Ngriffeth (talk) 15:40, 2 July 2025 (UTC)

Does anyone agree/disagree that the network science sidebar does not belong here? It seems rather like having a "Mathematics" sidebar for Accounting or a Game theory sidebar for Poker-possibly some topics would be of interest but a user that needed guidance on where to look further would end up in an intellectual version of a maze!!

If there are no responses to this within a couple of weeks, I will simply delete it. Ngriffeth (talk) 15:13, 4 July 2025 (UTC)

Can you link to where you found the assertion that sidebars are of questionable use? I think you may be overstating the unfairness of including this sidebar. ~Kvng (talk) 12:15, 13 July 2025 (UTC)

Wikipedia:Manual of Style/Lead section#LEADSIDEBAR

Apparently, the use of sidebars in the lead is discouraged but not prohibited:

"there is a moderate consensus in favor of introducing guidance that in general, editors are discouraged from placing navboxes in the lede, with about two thirds of participants in support. This consensus does not authorize the mass removal or deletion of sidebar navboxes. Several participants emphasized the need to make case-by-case determinations; if editors seek the deletion of a particular navbox, it should be brought to TfD as usual."

I didn’t read this all that carefully before. I’m not proposing to get rid of the Network Science sidebar completely, just questioning its relevance for this particular article. Does that require going through the TfD process? Ngriffeth (talk) 22:05, 13 July 2025 (UTC)

No, TfD would be for removing the sidebar everywhere. I doubt such a proposal would get anywhere. You seem to be offended by the sidebar covering theoretical/mathematical as well as practical topics. I personally see no problem with this. Good luck. ~Kvng (talk) 23:04, 13 July 2025 (UTC)

Thanks for your attention to this, and other computer networking edits.

I notice that no one has answered my central question, which is whether there is a process for removing a sidebar from an article. On the one hand, it seems like a simple edit, like any other, so I could just remove it; on the other hand, it’s a relatively big thing, so maybe not.

I was surprised that you think I’m offended, although on reflection maybe I am a bit offended at the waste of time this can introduce for a reader that knows little about computer networks.

As to being offended by theory/math topics, I personally like them, that’s where I started out, in math and theoretical computer science. My concern about this sidebar arises from the fact that as I gravitated to working on computer networks, I found my math and theory background minimally useful. Ngriffeth (talk) 14:41, 14 July 2025 (UTC)

@Ngriffeth the process is the talk page discussion we're engaged in. We need to see a consensus of editors here supporting removal ~Kvng (talk) 15:15, 14 July 2025 (UTC)

Keep for relevance. --Zac67 (talk) 20:23, 13 July 2025 (UTC)

I don’t see many topics of relevance to computer networks in the sidebar. I say this as someone who worked with and taught computer networks for about 20 years.

Can you explain which topics in the Network Science sidebar are relevant to a reader trying to learn about computer networks? Ngriffeth (talk) 22:07, 13 July 2025 (UTC)

In light of Wikipedia:Manual of Style/Lead section#LEADSIDEBAR, what would you think of making it a navbar instead of a sidebar in the lead? Ngriffeth (talk) 21:42, 4 August 2025 (UTC)

WRT WP:LEADSIDEBAR, there is no lead image or infobox in this article so why not put the first navbar at the top? If you want to change the form of {{Network science}}, I think you'd have to do that, not just here, but for all articles that use this sidebar. ~Kvng (talk) 22:47, 4 August 2025 (UTC)

I agree. I don’t want to change the Network Science sidebar. I just think the current sidebar has too many topics of no relevance to compiuter networks and too few that are relevant. If you think a sidebar is important to have here, I suggest a sidebar that has more topics of relevance to computer networks (honestly, a Computer Network sidebar would seem reasonable, it’s a big complicated area of computer science). Ngriffeth (talk) 17:23, 5 August 2025 (UTC)

Yes, you've said that. I understand your point but I had a look around and I could not a suggestion for improving organization; I just can't find a clean place to draw the line you're looking for. You don't have anyone agreeing with your proposal to remove the template. We need a better proposal to proceed. ~Kvng (talk) 23:05, 5 August 2025 (UTC)

Ok, thanks for your attention to this.

I’m dropping it as I don’t view this as sufficiently important to work on right now. Ngriffeth (talk) 13:22, 6 August 2025 (UTC)

In looking for a citation for this, I found this quote at https://www.coursehero.com/file/p3vdhdfc/Computer-networks-may-be-classified-by-many-criteria-including-the-transmission/

"Computer networks may be classified by many criteria, including the transmission medium used to carry signals, bandwidth, communications protocols to organize network traffic, the network size, the topology, traffic control mechanism, and organizational intent."

This is near-plagiarism from a not very authoritative source and also none of the classifications is used later in the article, which makes the inclusion of this paragraph of questionable use. Ngriffeth (talk) 19:42, 24 October 2025 (UTC)