Talk:Electrical telegraph

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

Two overlapping articles[edit]

Although this article was started earlier than the Telegraphy article, it can be argued that much of the content of "Electrical telegraph" belongs under the more generic heading of "Telegraphy". What is surely not appropriate is for vast swathes of content to be duplicated in the two articles. What do others think? --TedColes (talk) 15:25, 19 June 2018 (UTC)

I agree. The problem has been made far worse by an editor copying "vast swathes" of material from other articles both to the Telegraphy article and to here. Really, it should have been reverted out of hand at the time, if for nothing else failing to make a proper attribution and not properly preserving the citations. See #Unfollowable citations section above. It has been on my radar for some time to completely rework the telegraphy article, but it is daunting to take on such a large task. I have been putting it off for years.
My idea would be for the Telegraph article to concentrate on historic, non-electric, forms of signalling, of which there is a great deal that could be written, but is not in the article. More modern forms such as Wireless telegraphy more naturally belongs in Electrical telegraph. It uses the same coding (Morse) and requires the same skillset from its operators. Arguably, it should be at least mentioned in both articles, but as it has its own article it should, in any case, be severely summarised in one and mentioned in the other. SpinningSpark 16:03, 19 June 2018 (UTC)
I agree with TedColes that there should be consolidation but I'm not sure I agree with Spinningspark's solution. I would argue that Telegraphy is the WP:COMMONNAME used for Electrical telegraph today; electric telegraphy was by far the largest and most well-known system using telegraphic communication. Just an idea: maybe Electrical telegraph could be merged into Telegraphy? Nonelectric telegraphy systems, like Semaphore and Signal lamps have their own articles and needn't take up much room. --ChetvornoTALK 22:18, 3 July 2018 (UTC)
@Spinningspark: You say "there is a great deal that could be written" about "historic nonelectric forms of signalling" in the Telegraphy article. What forms are you talking about? And how many of those were called "telegraphy"? I am hesitant to see the article expanded to include signalling that does not fit under the term "telegraphy". BTW, I think the sections Internet and Email are ridiculously WP:OFFTOPIC in the Telegraphy article and could be deleted immediately. --ChetvornoTALK 20:28, 3 July 2018 (UTC)
Just noticed this topic has been discussed above, at The merger of two articles is needed --ChetvornoTALK 21:26, 3 July 2018 (UTC)
Sorry for the late reply. First of all, I wholeheartedly support your suggestion that the electric telegraph should have the primary topic name of telegraphy. That raises the question of what name the existing telegraphy article should be under. There are some definitions given in the telegraphy article. If we follow Chappe's definition, the subject is hardly different from signalling. If we follow Morse's definition, then teleprinters and fax machines are about the only things that fit.
On how I was thinking of expanding the existing telegraphy article, I don't agree with you that we should be limited to things actually called telegraphs. But if we don't so limit it, it's hard to avoid the article becoming an article on signalling. Consider first railway semaphore signals. The first commercial application of the electric telegraph was to fulfill such a role. That is, the transmission of a limited set of predermined messages. The electric telegraph was in competition here with systems such as steam whistles. It's hard to argue that railway semaphore and its predecessors does not belong in such an article when it is part of a continuous progression that also included the electric telegraph.
Next consider the marine engine order telegraph. This is actually called telegraph so fits your definition. More than that, it is precisely analogous to some of the early applications of the electric telegraph on the railways. I am thinking here of the signalling between station and engine house in places where rope hauling was used to get trains in or out of a station rather than under their own power.
The heliograph gets a mention in the article, but it could use expanding. It is not called telegraph but it was using the Morse telegraph code and was used by the US Army in Arizona to fill in in a region where the electric telegraph was not available. Smoke signals were used by their Apache enemy and this is an ancient form of signalling. According to that article, there was even a Greek alphabetic version. Smoke signalling was also used on the Great Wall of China, as was a complex form of flag signalling.
On things that are off-topic, I generally agree with you. I think the biggest culprit is the long list of telegram service providers. That falls under WP:NOTDIRECTORY. At the very least, it should be moved to the currently redirected Telegram. SpinningSpark 10:37, 8 July 2018 (UTC)
I don't think it appropriate to remove all non-electrical forms of telegraph from an article called "telegraphy". Marine engine order telegraphs were not electrical initially.--TedColes (talk) 18:07, 8 July 2018 (UTC)
Who suggested we should do that? I certainly don't think that. And yes, I am aware that engine order telegraphs were originally mechanical but fail to see how that is relevant to the discussion. SpinningSpark 19:39, 8 July 2018 (UTC)

───────────────────────── Wikipedia groups things by what they are, not what they are called. Electrical telegraph is, for better or worse, the article about the things connected by wires or, maybe, wireless. Engine order telegraph is another "thing" and has its own article, but could be WP:SUMMARYSTYLE at "Telegraphy", it was never word/phrase communication device. I am noticing a problem that also cropped up at Talk:Wireless telegraphy, should we treat one type as the current type and treat the others as "historical". All of these systems are for the most part "historical" - nobody really uses any of them anymore for all intents and purposes. Telegraphy should probably be boiled down to a WP:SUMMARYSTYLE article and purged of most of its content. Its lead definition covers several types of telegraphy. Wired and WP:SUMMARYSTYLE wireless should probably go here. Fountains of Bryn Mawr (talk) 20:01, 8 July 2018 (UTC)

Sorry, AFK. Spinningspark, I was the one who kind of suggested we remove non-electrical signaling from Telegraphy, or at least summarize it. Okay, now I understand the organization better. If there is no article between Signaling and Telegraphy, then I agree with Spinningspark and Fountains of Bryn Mawr that Telegraphy should be a broad article, probably WP:SUMMARYSTYLE, covering historical forms of text signaling, following Chappe's definition. Therefore it can't be the main article for electrical telegraphy, this article should be. As TedColes says, the redundant content in Telegraphy should be removed leaving a summary. The telegraphy articles should be limited to historical methods; nothing after teletypes or radiotelegraphy, no email. (IMHO) --ChetvornoTALK 13:31, 20 July 2018 (UTC)

Some things the article needs[edit]

In the spirit of a brief review, here are some things I noticed that the article could use. The history is excellent, but there is no "How it works" section describing the science of how an electric telegraph line works, giving a typical circuit diagram, and such details as an explanation of "dots" and "dashes", duplex operation, "break-in" signaling, Q-codes, Continental vs American Morse, telegram format and prices. There are images of some early historical systems, but few images of the mature widely used technology of the latter 1800s. There are two redundant images of paper tape recorders at top, only one image of a primitive key, not a standard Morse key, and no images of the other important components of a telegraph line - the liquid battery, relay, telegraph sounder, telegraph pole, telegraph office and telegram form. I'm working on other articles now but could probably contribute at some future date. However I don't really know much. --ChetvornoTALK 14:41, 20 July 2018 (UTC)

  • Statistics: At its largest extent, how big were the world's telegraph networks? How many stations? How long were the world's longest telegraph lines? What was the total length of telegraph wires strung? How many stations per line? How many telegrams were sent per year? Employment? Annual revenue? Annual profit? Were all continents (besides antarctica) linked by telegraph cables? Were all major telegraph networks linked? could you send a telegram between any two people in the developed world? Average delivery time?
  • In "End of the telegraphy era" it says the end of telegraphy in America was associated with the fall of Western Union, but it doesn't say clesrly when that was. What about outside America (besides India)?


@TedColes: Montgomerie may be notable. See this and this scholar search. If you are going to remove redlinks, at least give a rationale why the subject is not notable. If you don't, people will think you have not read Wikipedia:Red link. SpinningSpark 23:56, 23 November 2018 (UTC)

  • Also don't invent nonsense terms like "acoustic multiplexing" just to avoid a redlink to the real technique. Andy Dingley (talk) 00:05, 24 November 2018 (UTC)
    • At least this provoked me into writing William Montgomerie's article so some good has come of it. I note he had quite a few other incoming links even before I started. SpinningSpark 15:03, 27 November 2018 (UTC)

Alter's telegraph[edit]

I have removed the mention of Alter's telegraph. As far as I can tell, nothing was published or reported at the time this telegraph was supposed to have existed. The story is entirely anecdotal, coming to light long after the rise of the telegraph. Ok, there are some claimed witnesses, but the story is still "Oh yes, I built one of those in my barn once". Yeah, and I built an electronic computer in my bedroom in the 1950s (or tried to). It's an extraordinary claim, so needs more scholarly sources than an interview in Popular Science. In any case, even if perfectly true, it had no effect whatsoever on the development of the telegraph since no-one actually involved in the development had heard of it. At best, it is tangential to the story. SpinningSpark 09:58, 16 December 2018 (UTC)

I would also add that saying he predated Morse is a false comparison. Alter had an experimental system in his barn. Morse's system was actually in use. There were many working experimental systems that predate Alter's claim, and over much larger distances; Ronalds (1816), Schilling (1832), Gauss (1833, and a system in use by 1835), and even Cooke had an experimental system by March 1836. SpinningSpark 10:25, 16 December 2018 (UTC)

Defining opeing paragraph[edit]

I have heavily edited down the recent change to the opening paragraph in the lead. It read,

First of all, I don't think it's a good idea to use modern telecomms terms to describe telegraphy when they just were not used, or even existed, in the industry. "Digital on/off" is not correct, or not correct in all cases, so it is not a defining feature – see quadruplex and harmonic telegraph for instance. "Combinatorial logic to encode and decode data" didn't exist at the time. It just didn't happen, unless you want to count what's going on in the operator's head. On, radiotelegraphy, I thought we had reached a consensus (at least by default) that that doesn't belong here and should be covered in its own article and summarised in the telegraphy article. We would certainly get into difficulties of scope if we included all electromagnetic radiation transmissions as "electrical". Fibre-optic, for instance, would then suddenly be in scope. SpinningSpark 11:46, 3 April 2019 (UTC)

Agree absolutely; the modern communication jargon, like "digital", is going to mislead general readers.
However, I feel the existing introduction is too vague; for example it should be explained in the lead that the word telegraph means a point-to-point text messaging system.
Although the article describes many alternate telegraph technologies, the most widely used systems, commercial telegraph networks, generally worked the same way. A telegraph line was two or more stations connected by an overhead wire supported on utility poles. At the sending station an operator tapped on a switch called a telegraph key to spell out text messages in Morse code. The switch connected a battery to the wire, sending pulses of current down the telegraph line. At the receiving station the current pulses activated a telegraph sounder, which made a clicking sound. A receiving operator who knew Morse code translated the clicks back to text and wrote it down. That is a bare-bones description of what an electric telegraph was, and I think something like it should be in the introduction. --ChetvornoTALK 18:22, 3 April 2019 (UTC)
No issue at all with your assessment that the lead could use improvement, but we should be very cautious with equating the Morse telegraph (which by the way, badly needs its own article) with telegraph generally, or even saying that the Morse telegraph was the "normal" system. That was the case in the US, but not elsewhere. It certainly wasn't in Britain and the British Empire which was a not insubstantial part of the world. There, the Cooke and Wheatstone system and other non-Morse systems were the norm. In France, a different needle telegraph system was used. The Morse system did not become the standard in Britain until after nationalisation, and even then other systems lingered on. By the time Morse had actually become standard, teleprinters were coming in and the Baudot code was rapidly taking over as the most used code. None of the elements of the Morse system can be claimed to be defining; not the code, not the sounder, not even the ubiquitous telegraph pole – the second largest company in Britain standardised on buried cables. That company also did not use batteries, so even that is not defining.
The lead is supposed to do two things: define the subject and summarize the article. Concentrating on Morse does neither of those things. SpinningSpark 22:28, 3 April 2019 (UTC)
I thought teleprinters used Murray's 1901 modification of Baudot's 1870 code, so wouldn't "Baudot-Murray" be more accurate?--TedColes (talk) 11:13, 4 April 2019 (UTC)
That doesn't significantly change my point. SpinningSpark 13:24, 4 April 2019 (UTC)
Okay, I didn't realize how wide the use of needle telegraphs were (which I would have if I had actually read the article instead of just skimming it!). You guys wrote an excellent, comprehensive history. I still think there should be some sort of brief description of how telegraphs work in the introduction. Maybe a generic explanation that covers both Morse and needle systems? Their core technology was the same: At the sending end switches are pushed or turned in sequence to spell out characters of a text message. These send pulses of current down the line or lines. At the receiving end an electromagnet-actuated indicator displays the sequence of characters either aurally or visually. Yes, there were many elaborations like multiplex and harmonic telegraphs that fall outside this description, but we can mention that. --ChetvornoTALK 22:34, 4 April 2019 (UTC)


Spinningspark has disputed that telegraph lines are inherently half-duplex. Just in case there is confusion here:

  1. Simplex - communication proceeds in one direction only
  2. Duplex - communication can proceed simultaneously in both directions
  3. Half-duplex - communication can proceed in both directions, though not at the same time.

Examples: (1) broadcast radio, (2) telephones, (3) "walky-talky" type radios.

Next, what is meant by a "line"? A communications path might consist of several wires, such as the Wheatstone system, or a single wire such as the classical Morse system. In the latter case signalling is always simplex or else half-duplex using a "turn-around" character (think of radio operators using the proword "over"). The only mechanism to permit full duplex operation would be modulation of carrier waves, and I'm not aware that telegraph systems ever used this (though could be proved wrong). Regards, Martin of Sheffield (talk) 15:43, 20 July 2019 (UTC)

You certainly can be proved wrong. See Quadruplex telegraph. Bare copper wire lines are not inherently half-duplex. I would have thought that was obvious. Your example #2 shows that. SpinningSpark 15:50, 20 July 2019 (UTC)
But that's a half-duplex system. For the duplex part consider that the current can only be flowing in one direction at any instant. The Stearns part stops the local receiver responding to the local key, but to activate the remote receiver current must be passing from local to remote. The utility of the system was not to permit simultaneous transmission which would be difficult for an operator to handle, but to allow recording mechanisms to just record inbound traffic. The diplex part relies on using opposite polarities. Consider what happens if local A holds the line positive and local B is trying to pull the line negative. Martin of Sheffield (talk) 16:07, 20 July 2019 (UTC)
Every source I have ever read on the quadruplex telegraph says that four messages (two in each direction) can be sent simultaneously. See for instance Bray, Innovation and the Communications Revolution. Please provide a source for your claim. In any case, quadruplex notwithstanding, duplex telegraphs were in existence prior to Edison's invention. It is an utterly standard analysis of transmission lines to consider two currents flowing simultaneously on a line representing signals travelling simultaneously in opposite directions. No carrier wave modulation is required to either generate such signals or to separate them. SpinningSpark 17:13, 20 July 2019 (UTC)
I'm suprised that you gave a link to chapter 6. The first paragraph mentions duplex and quadruplex systems just in passing, the whole of the rest of the chapter is "The telegraph-telephone frequency-division multiplex transmission engineers"! The first two paragraphs of section 3.3 would have been a better quote, but even then it rapidly leads to Baudot and his time-division multiplexing. As regards a source, you provided one - Quadruplex telegraph. Early telegraphs worked at slow speeds over relatively short distances, a DC analysis is therefore more appropriate than a transmission line analysis using step functions or AC theory. According to Morse_code#User_proficiency the fastest keying ever recorded was 35 wpm. Using a 50 dot duration word that comes out at around 30 baud. Assuming 1 baud = 1 Hz, that is a wavelength of 10,000 km. Martin of Sheffield (talk) 19:37, 20 July 2019 (UTC)
I could cite any number of books that say the same thing. Do your own gbooks search if you doubt it. The bottom line is that your addition has been challenged. Per WP:V you need to cite a source before it can go back in (and one that has more authority than all the dozens of sources that contradict your claim – which does not include a Wikipedia article, even if I accepted that it supported you, which I don't). Arguing from your own analysis doesn't cut it if you don't have a source to back it up. SpinningSpark 13:31, 26 July 2019 (UTC)
I thought you'd given up arguing this. Before I waste more time, perhaps you could explain "The bottom line is that your addition has been challenged. Per WP:V you need to cite a source before it can go back in", since the only time I've edited the article was a clarification on Morse. The reference to the Wiki article is that it clearly explains the mechanism, have you read it in detail? Martin of Sheffield (talk) 14:40, 26 July 2019 (UTC)
My apologies, it was not you who inserted the text. I was discussing the text I removed here which is the same edit you were referring to in your opening post. However, that does not change the position that a reliable source is required before it could go back in. I don't care what it says on the quadruplex telegraph page, we don't accept other Wikipedia articles as reliable sources, but in any case, nowhere on that page does it say that the system is not true duplex. SpinningSpark 16:46, 26 July 2019 (UTC)

This site has some simplified diagrams of the quadruplex telegraph which may help understand its operation. SpinningSpark 13:58, 28 July 2019 (UTC)

New lead[edit]

First of all, congratulations to Chetvorno for providing a badly needed much better, and more detailed lead.

I have extensively reworked the second paragraph for two reasons. Firstly to get it into historical order; needle telegraphs came first by a wide margin and remained important in the UK, British Empire, and France for some time, even after the Morse system became widespread. Secondly, it was too Morse-centric, especially after Mark's additions. Certainly, the Morse system is important, but the little details of the way the Morse system worked don't belong in the lead of this article. They belong in an article on the Morse system which Wikipedia currently sorely lacks. The Morse code article stands in for this lack with redirects, but the code is not the same thing as the system.

Thanks again to Chetvorno for something that should have been done a long time ago. SpinningSpark 16:44, 20 July 2019 (UTC)

American Civil War: USASC and USMT[edit]

Just a question about the American Civil War section. The final paragraph may give the impression that the USASC's telegraph system replaced the USMT's. This was not the case. Meyer tried the logical move of absorbing the USMT into the USASC but was unsuccessful. He also favored the adoption of the Beardslee telegraph as USASC's main technology but that instrument was abandoned by 1863. The USMT remained independent and, with its Morse-based telegraph network, retained a more or less complete monopoly on telecommunication throughout the war. I think the USASC's main contribution was the idea of the "telegraph train" as a more efficient and military way of deploying a system. You probably don't want to go into so much detail in this article, but perhaps a re-wording would clarify the situation.

In the third paragraph is the sentence "Many Northern victories were found due to the use of the telegraph." I'm not sure what that means. Also not sure the articles linked to about Antietam, Chickamauga, and Sherman's March support the idea that the telegraph played a major role in their outcomes.

18th or 19th century invention?[edit]

Recently there has been a difference of editorial opinion as to whether the electric telegraph should be categorized as an 18th century invention or a 19th century invention. --ChetvornoTALK 21:19, 8 August 2019 (UTC)

My opinion is that it should be categorized as a 19th century invention. According to the "Early history" section, the idea of an electric telegraph was conceived by several people in the 1700s, who built working short range prototypes, so I can see the argument for the 18th century. However these were all impractical because they were based on electrostatics. The article doesn't go into all the reasons, but it's pretty clear electrostatic discharges could never carry signals through wires a practical distance. The first successful electric telegraph systems, as well as all the direct current technology which was necessary for them, the electrochemical battery (1800) and the electromagnet (1824), was invented in the 1800s. IMHO, just having an idea does not constitute an invention; you also need the technology to make it practical enough to be useful. --ChetvornoTALK 21:19, 8 August 2019 (UTC)
Hmm.., I kind of agree, but what you describe would be the invention of the electromagnetic telegraph. The article has a broader scope than that. Nowadays, publication of a patent would be enough to claim invention. It would not be necessary for it to be practically succesful for that claim to stand. SpinningSpark 21:59, 8 August 2019 (UTC)
@Spinningspark: I've run into this issue on other articles, as I'm sure you have also. Many editors take the position that the "invention" occurs when the first person has an idea, writes about it or does experiments, and he is the "inventor". In my opinion this can lead to ridiculous statements giving undue weight to minor players. In this view Mahlon Loomis invented radio in 1866, Otto von Guericke invented the electric generator in 1663, and Gottfried Wilhelm Leibniz invented the computer in 1674. The history of most inventions starts with a series of "dead end" technologies; impractical solutions to the problem. It seems to me that in the development process the "invention" doesn't occur until some degree of practicality is achieved. I'm not suggesting this as an absolute principle, just that when deciding when something was "invented" (always a value judgement), common sense and due weight be considered.
In the case of the electric telegraph, as far as I can see the only thing "invented" in the 1700s was the idea; the electrostatic technology was a dead end. The electrostatic experiments were important steps and are valuable additions to the article; these experimenters invented "a" electric telegraph, or telegraph precursor, but not "the" telegraph. It seems to me the electric telegraph is synonymous with the electromagnetic telegraph, it is the only kind that transmitted beyond a single building, the only kind that got beyond prototype stage, the only kind that had any practical use, and it is the common meaning of the term 'electric telegraph' today. --ChetvornoTALK 19:34, 20 August 2019 (UTC)

Wheastone ABC telegraph[edit]

I think the Wheatstone ABC telegraph has been given WP:UNDUE prominence in this article. It has more words written than both the Cooke & Wheatstone, and the Morse systems. It was basically a minor sideshow compared with the latter two. As far as I know, it was never used in public telegraph offices, it was entirely for private use. There are other systems that were much more widespread that aren't mentioned at all. The Henley-Foster telegraph and the Foy-Breguet telegraph are examples that were taken up by major organisations. Also, it has been inserted under the Cooke and Wheatstone heading, but it never was an offshoot of the Cooke and Wheatstone system.

I propose that the material is cut out into a separate article, leaving just a mention and a link. SpinningSpark 16:52, 25 August 2019 (UTC)