Email In the 18th Century

morphovar forwards a writeup in Low-tech Magazine recounting an almost-forgotten predecessor to email and packet-switched messaging: the optical telegraph. The article maps out some of the European networks but provides no details of those built in North America in the early 1800s. Man-in-the-middle attacks were dead easy. "More than 200 years ago it was already possible to send messages throughout Europe and America at the speed of an airplane — wireless and without need for electricity. The optical telegraph network consisted of a chain of towers ... placed 5 to 20 kilometers apart from each other. Every tower had a telegrapher, looking through a telescope at the previous tower in the chain. If the semaphore on that tower was put into a certain position, the telegrapher copied that symbol on his own tower. A message could be transmitted from Amsterdam to Venice in one hour's time. A few years before, a messenger on a horse would have needed at least a month's time to do the same."

Spam?

[AlphaDrake (1104357)]

Did spam make it across these networks as well?

"Having trouble with the smell of thine donkey? Only have the one mistress? Try friar pete's ol' fashioned elixer de skunke, it's new lead based formula works wonders like that Jesus guy over there"

Re:

[K. S. Kyosuke (729550)]

Ah, yes, Claude Chappe's optical telegraph. :-) Nice that people still remember these. You can also read about them here [wikipedia.org]. The part about the system cost compared to the electric telegraph is really interesting. It is not very suprising that this system was ultimately replaced soon after electrical telegraphs had become available. (One has to ask why Czech Post - providing virtually the same quality of service - has not yet seen the same fate? ;-))

Re:Spam?

[AKAImBatman (238306)]

It is not very suprising that this system was ultimately replaced soon after electrical telegraphs had become available.

Actually, it wasn't. The electrical telegraph had a very rocky start. Both France and Britain had optical telegraphs in place and were uninterested in investing in this new "electric" form of telegraph. Especially since those who worked on electric telegraphs were often untrained quacks.

It took a relatively new nation that lacked a telegraph (i.e. the United States) to cause the electric version to catch on. Even there, it took a while before the possibilities were really explored. Once it caught on, though, it caught on like wildfire. Didn't take long for an international telegraph to get setup, and for ticker-tape machines to appear.

For those interested in the topic, I highly recommend the book The Victorian Internet [amazon.com]. It is well written, well researched, and tells a fascinating tale of the telegraph development that parallels the development of the Internet. On top of that, it sheds light on how the telegraph affected computer design and the communications protocols we use today. (e.g. ASCII is derived from the telegraph codeset called "Baudot Codes". Named for the inventor, Émile Baudot. He also has a measure of transmission speed named after him called "Baud". As in, a "300 Baud Modem". )

Re:Spam?

[Alioth (221270)]

Terry Pratchett did - his recent book "Going Postal", one of the main "characters" of the story is the clacks - the Discworld optical telegraph network. It's a fun book.

Re:Chappe's telegraph and buiding of a fortune

[AI0867 (868277)]

actually, the story was more interesting
-Rothschilds get information early
-other people know rothschilds get the information early
-rothschilds dump all their stock
-everyone else dumps their stock
-stock crashes
-rothschilds buy everything

massive stock manipulation, but I guess that was legal back then.

(or at least this is the version I heard)

Re:Chappe's telegraph and buiding of a fortune

[instarx (615765)]

actually, the story was more interesting
-Rothschilds get information early
-other people know rothschilds get the information early
-rothschilds dump all their stock
-everyone else dumps their stock
-stock crashes
-rothschilds buy everything

massive stock manipulation, but I guess that was legal back then.

Actually this would be perfectly legal today. Getting public information faster than everyone else is smart, not illegal; and there is noting illegal about selling stock to drive the price down and then snapping up deals. Market-makers do it every day to shake out margin traders.

Re:

[Foobar of Borg (690622)]

I thought the Rothchilds use carrier pigeons, a competing form of packet based communication.

Yes, but their packets were more susceptible to malware, especially of the Hawk variety. A Beowulf cluster of Hawks was the ultimate in DOS attacks.

Re:

[Tablizer (95088)]

Did spam make it across these networks as well?

In an 18th-century British accent: "Oh bloody hell, I shall not need my wanker any bloody bigger! May the Queen assign lasting damnation upon your deplorable message."
       

Re:Spam?

[by Anonymous Coward]

Note: A wanker is the term for a person who masterbates. "A wanker wanks".

I live in fear that this may be marked informative.

Only on slashdot

[Colin Smith (2679)]

The irony of having to define the word "Wanker" to a bunch of mostly American nerds.

 

Re:Spam?

[ArAgost (853804)]

Well, I can assume that once you reach a certain proficiency, you can be called a MASTERbator.

Re:Spam?

[morgan_greywolf (835522)]

Indeed. A guy named Isaac Bayes would stand between two of the towers and every time he spotted a reference to making your penis larger, he would create a lot of thick black smoke so as to block the transmission between two towers.

And to this day, most spam filters are still called 'Bayesian filters.'

Re:

[petermgreen (876956)]

Did spam make it across these networks as well?
I doubt it for simple economical reasons. Theese networks were probablly more expensive to use than the postal service and unsolicited bulk messages aren't really very urgent.

Light the Fires

[coaxial (28297)]

Gondor needs help.

Postal mail used to be pretty good, too.

[Kadin2048 (468275)]

I was reading something recently that discussed the US Postal Service in the late 19th century. In some major cities, like New York and Boston, the mail used to come as much as five times a day. That meant you could write to someone (local, served from the same Post Office) in the early morning, have it picked up in the first round, delivered in the second, have their reply picked up in the third, and delivered on the fourth. (And you could even send a reply back in the final pickup for delivery the next morning.) That's pretty good -- some people I know don't even check their email that often!

If you wanted service and delivery times that good these days, you'd need to go with a courier service.

Re:Postal mail used to be pretty good, too.

[iocat (572367)]

I remember reading an article a few years ago, on various companies' ettiquette for the term "email." Some caled it 'email,' some called it 'electronic mail,' some called it by a quaint brand name ('QuickMail', anyone?). The article noted that at Micorosoft, it was simply refered to as "mail." So the author asked the inevitable question: "What do you call something that comes in a physical envelope?" The answer? "FedEx."

Anyway, there is a good book called The Victorian Internet that, despite its suspect name, is extremely well written and goes into great and fascinating depth on the telegraph (optical and electronic), as well as the pro-tech savvy of the Victorian age. I'm too lazy to put in a link for you, but I assure you, the google or the amazon can give you all the details.

Re:

[c6gunner (950153)]

With such a poor understanding of economics, it's surprising you were ever able to afford a computer!

Ah, Clacks

[The Grey Ghost (884000)]

Apparently where Terry Pratchett got the clacks - http://en.wikipedia.org/wiki/Clacks [wikipedia.org]

Re:Ah, Clacks

[by Anonymous Coward]

No worries. Antibiotics will clear that right up.

Re:

[DynaSoar (714234)]

The wife reminds me that quantums, being able to traverse the multiverse(s?) must therefore be able to travel any direction in time. Therefore Robert Hooke may have gotten the idea from Pratchett. Given that more people have read Discworld books than have read Hooke's works, and that any of them may emit idea quantums, she is most likely correct.

but

[Sobieski (1032500)]

If it was "wireless and without need for electricity", then it was not electronic mail

Re:but

[fbjon (692006)]

This was enterprise mail, hence not available to the plebes.

"Minor" mistake but...

[RobertM1968 (951074)]

"provides no details of those built in North America in the early 1800s. Man-in-the-middle attacks were dead easy"

The "early 1800's" is the 19th Century - not 18th.

Re:"Minor" mistake but...

[hpa (7948)]

[1] I've read they did it for efficiency because internally it multiplies the index to get the starting offset in an array of equal-sized elements. If you start at one, then indexing requires a subtraction, or else waste an element, which may have mattered in the 60's when RAM cost an arm and a leg.

The compiler is more than capable of doing this transformation. The real reason is because the vast majority of algorithms are easier to describe with the first index as zero -- this was a lesson learned from FORTRAN, which started indexing at 1.

Semaphores and smoke signals are ancient

[davidwr (791652)]

Native American smoke signals date back to pre-Columbian times.

Torches and and other forms of optical telegraphy date back to ancient times.

Thanks to the seminal work of J. Hofmueller and his colleagues, modern flag semaphores can also be used to encapsulate IP datagrams [ietf.org]. Presumably, this is more efficient than delivering the same traffic by animal transport [ietf.org] but less efficient than by wire or radio.

Telegraph Hill in San Francisco

[ortcutt (711694)]

Telegraph Hill in San Francisco was at one time the site of an optical telegraph. Hence the name.

The hill owes its current name to a semaphore, a windmill-like structure erected in September 1849, for the purpose of signaling to the rest of the city the nature of the ships entering the Golden Gate. Atop the newly-built house, the marine telegraph consisted of a pole with two raisable arms that could form various configurations, each corresponding a specific meaning: steamer, sailing boat, etc. The information was used by observers operating for financiers, merchants, wholesalers and speculators. As some of these information consumers would know the nature of the cargo carried by the ship they could quickly predict the upcoming (generally lower) local prices for those goods and commodities carried. Those who did not have advance information on the cargo might pay a too-high price from a merchant unloading his stock of a commodity -- a price that was about to drop. On October 18, 1850, the ship Oregon signaled to the hill as it was entering the Golden Gate the news of California's recently acquired statehood.

Telegraph Hill [wikipedia.org]

patent trolls

[yabba-dabba-do (948536)]

In other news, NTP is now looking for someone to sue over this infringing technology.

The Victorian Internet

[blamanj (253811)]

Tom Standage's book covered this quite well [amazon.com].

i have a great comment

[circletimessquare (444983)]

please watch this space for 3 hours in order to view it

my comment is currently being transmitted from schenectady to poughkeepsie and the bad weather is interfereing with the candles staying lit

Common in Italy in the middle agaes

[by Anonymous Coward]

The whole cost of Southern Italy is full of towers that were used a light based communication/alarm system, especially against the raids of the so called saracens (people from the Islamic nation from the south) in the middle ages. I believe that a similar system was also used in Roman and possibly Greek times. The distance between the towers is also similar, 5-20Km.

So...

[Rip Dick (1207150)]

Was the Optical Telegraph networked described by the clueless politicians of the time as a "series of flags"?

Wow!

[sunspot42 (455706)]

Looks like the Victorians could copy and transmit data faster than Windows Vista!

My semaphore tower sucks

[Jesus_666 (702802)]

I don't want to start a holy war here, but what is the deal with you semaphore fanatics? I've been sitting here at my freelance gig in front of a semaphore tower (a 1860/300 w/64 flags) for about 20 weeks now while it attempts to copy a 17 Meg file from one city on the east coast to another city. 20 weeks. At home, on my dovecote running Columba livia domestica, which by all standards should be a lot slower than this semaphore tower, the same operation would take about 2 weeks. If that.

In addition, during this file transfer, the newspaper will not work. And everything else has ground to a halt. Even my inkwell is straining to keep up as I type this.

I won't bore you with the laundry list of other problems that I've encountered while working on various semaphore towers, but suffice it to say there have been many, not the least of which is I've never seen a semaphore tower that has run faster than its dove counterpart, despite the semaphore towers' faster signalling architecture. My pigeonry with 8 Columba palumbus' runs faster than this 300 flag-position machine at times. From a productivity standpoint, I don't get how people can claim that the semaphore tower is a superior machine.

Semaphore addicts, flame me if you'd like, but I'd rather hear some intelligent reasons why anyone would choose to use a semaphore tower over other faster, cheaper, more stable systems.

Sempahore towers

[Uomograsso (968695)]

There is a reconstructed tower at Chatley Heath near Guildford, England, which was part of the route from the admiralty in London down to Portsmouth.

There are still some left in Barbados:

http://photo.clifford.ac/2007/Barbados.October/tn/dscn2211.jpg.index.html [clifford.ac]

and here is what you see when looking at Cotton Tower from Grendade Hall:
http://photo.clifford.ac/2004/Barbados.April/tn/p4130674.jpg.index.html [clifford.ac]

--
Alan clifford

They're in an old movie too

[belmolis (702863)]

I remember first seeing these in an old movie, which I remember as being in black-and-white. It may have been an old version of The Count of Monte Cristo.

Re:The Count of Monte Cristo

[tolworthy (1205778)]

It's not just the movie. These message towers play a key part in the novel. The Count ruins one of his enemies, a banker, by sending a false message about a foreign war.

Sorry, but...

[djupedal (584558)]

The Great Wall in China put similar means to use hundreds of years earlier.

Colored flags, whistling arrows, fires & hand signals all worked as part of a communication chain that spanned greater distances as well (6,400 km).

And 'man-in-the-middle' attacks were usually over before they began :)

taggers are fucking illiterate

[toby (759)]

BEACONS of Gondor, for Sauron's sake.

BEACONS.

If you can't afford a dictionary, rednecks, at least Google.

Re:taggers are fucking illiterate

[halcyon1234 (834388)]

Mmmm, bacons of Gondor. Sizzling fatty meats of Frodo!

Semaphores weren't the first

[Jesus_666 (702802)]

Actually, the semaphore-based network wasn't the first on in Europe. Before it, there was a simpler network based around mutexes, but it wasn't very popular because it got quite bothersome once you had more than two people communicating. Still it was a major step forward from the previous concurrent networks where the non-locked shared message space meant that if two people broadcasted at the same time they'd overwrite each other's messages.

Much later, North America would see an experimental monitor-based optical messaging network, but the cost of keeping hundreds of big CRTs powered on all the time quickly put an end to it.

Re:

[Watson Ladd (955755)]

Later Reppy implemented a rendezvous based messaging system, but only the French really understood it.

"Virus"

[Tablizer (95088)]

network consisted of a chain of towers... placed 5 to 20 kilometers apart from each other. Every tower had a telegrapher [worker], looking through a telescope at the previous tower in the chain...

Back then when a "node was infected with a virus", it was literal.
   

The First Time Information Outpaced Man

[Hubec (28321)]

Before the semaphore telegraph a man could travel faster than information. Am I the only one who thinks that's just really cool? The whole concept of being able to race across the globe faster than events is completely alien to our current existence.

Hmmm... Let me put it this way; Before the semaphore telegraph, the world was split into a very large number of simultaneous but completely separate realities. As soon as that telegraph came into existence those realities began merging into one.

Re:

[complete loony (663508)]

Certainly the latency was lower, but the bandwidth sucked. Don't underestimate the bandwidth of a state coach full of parchment.

Horses versus humans

[Tablizer (95088)]

Article: Humans or horses can maintain a speed of 5 or 6 kilometres an hour for long distances.

It may defy common sense, but a runner in top shape can almost match the pace of a horse over long distances. There used to be a yearly contest in England, and a human sometimes won. Our ancestors used to chase down pray by outlasting them in the heat (some isolated tribes still do). Our sweating system keeps us cooler than hairy animals. However, it may be more economical to wear out a horse than a human. Plus, a horse can carry more.
   

Re:

[demonlapin (527802)]

That's a nice story, but experience [thislife.org] disagrees with you. Quadrupeds move much more efficiently than we do. We're smarter than they are, so we take advantage of their behaviors to kill and eat them. Driving herds off cliffs, e.g. However, the experience of the Plains Indians with horses pretty clearly shows that people will take any advantage they get and use it to master their surroundings. If people on horses were inferior to people on foot, they wouldn't have bothered to become expert horsemen.

Re:

[ucblockhead (63650)]

Plus, a single person can switch horses. That's how the Pony Express worked, and it's how people could make 200 miles a day even in classical times.

Man-in-the-middle attacks?

[Dan East (318230)]

Man-in-the-middle attacks were dead easy.

No they weren't, and the article doesn't say that they were. Man-in-the-middle attack means that transmitted data can be modified, or entirely new data can be introduced. Think about it. You have a telescope permanently aimed at the next station in line, viewed by a person who has spent thousands of hours staring at that station. Now don't you think if someone, somehow, got in that exact line of sight with their own semaphore in attempt to transmit their own data, that it would be extremely obvious to the operator that something was very wrong?

What the article does say is that the system is vulnerable to eavesdropping. However, a number of solutions would be available. Shutters could be used to restrict visibility of the semaphores to the line of sight of the next tower. Since they were elevated, it would be difficult to get into that line of sight in most terrain. Obviously, the messages themselves could be encrypted as well. The semaphore operators did not have to understand their message. They simply moved the position of their signaling arms to match the position of the sending tower. The sending tower would visually verify that the receiving tower had properly copied the data. The operators did not need to know what the data meant to relay the information - only the initiator and consumer of the information needed the ability to encrypt / decrypt, which is still where we stand today.

Telegraph was very much open to eavesdropping - in fact, I believe it was much easier. Simply pigtail off of any of the thousands of miles of wire, and run a line to a comfortable listening post out of sight of the railway or road. With radio it became even easier!

Dan East

Re:

[kobotronic (240246)]

Your telescope is trained at the next semaphore tower, yes. But can you tell whether the operator sitting hidden beneath and pulling the levers is the person it is supposed to be, or perhaps some impostor who by use of force or bribery took over the controls? Isn't this a plausible injection vector for a man in the middle attack?

Re:did china do this as well on the great wall?

[AlphaDrake (1104357)]

Link [wikipedia.org]