Technology News Technology News

Another Perspective


Organized message delivery systems are as old as civilization itself.  The Egyptians had some kind of postal system in place as far back as 2000 B.C.  More than 2,500 years ago, Cyrus the Great set up a kind of pony express in his Persian empire.  The Chinese had a postal system in place long before that, and maybe even before the Egyptians had theirs.  (We'd know a lot more if more records had been preserved.) The Romans did a bang-up job that included the extensive construction of post roads.  Fast-forward a few millennia, to e-mail.  It's hard to say whether this is the future of the post or the end of it.

Until this month, I thought the British had the best postal system in the world.  Maybe they still do.  But during the past few weeks, it became clear that twice-a-day mail delivery, for decades a feature of the Royal Mail, is finished.

Postal service in Britain used to be even more impressive.  There was a time when you could mail a letter anywhere in London early in the morning and have it delivered by the end of the same day.  Until the latest round of service cutbacks, it was still overwhelmingly likely that any first-class letter mailed anywhere in Britain would be delivered the next day for about the equivalent of 50 American cents.  Imagine that!  FedEx for half a buck!  This wonderful quality of service is probably going to fade away.

It's not just that I'm a fan of the way the British do a lot of things, although that is certainly the case.  The Brits are proud of their postal system, and put their legislation where their pride is.  The superiority of the Royal Mail is embodied in law.  Under English contract law, if you accept an offer by mail, the legal time of the acceptance is the moment you mail that letter.  That's right.  Not when it's received.  Not a day or two after you send it.  The second you pop it into a mailbox, it is deemed to be delivered to the offeror.

Pillar box
Pillar Box
Nothing will ever fill its slot

English law is far older than American law, and is its dominant progenitor.  So if you think the lawyers you know are weasels, imagine how high an art form English legal weaselry has become.  They have had hundreds of extra years to work out the angles.  Yet they still stick with the postal acceptance rule.  Even the smartest lawyer in England (which probably means the Prime Minister's wife, noted barrister Cherie Booth) can't weasel out of the timing on a deal accepted by post, unless both parties agree to an exception in advance.

British second-class mail, which costs just over the equivalent of 30 cents, has generally been faster and more reliable than first-class mail in the USA.  And Canadians grumble that they would be happy if they even got the kind of service their neighbors to the south received.

Even with its special legal status, in England (or anywhere else), when it comes to speed, you just can't beat e-mail.  It's close to instant.  It's also very cheap (and it's seen as free by users whose e-mail costs are bundled into Internet services deals).  The basic function of e-mail, getting a message from point A to point B, is the same as that of physical mail, the service that Americans call snail mail with considerable justification.  But e-mail is not like sending a letter, and in some ways it can never be.

Originally, e-mail was more like telegraphy or telex than postal mail, and looked it.

In 1971, when Ray Tomlinson got a mail transfer agent and a receiving client running under TENEX, a third-party operating system for the Digital Equipment PDP-10, which academics and researchers preferred to DEC's own TOPS-10 software, there was no Internet.  Messages were just strings of text rendered on a teletype terminal or, rarely, a CRT display.  Tomlinson, a researcher at Bolt, Beranak and Newman, one of the nation's top technology innovators in those days, and the outfit that wrote TENEX, achieved immortality.  He was the one who put the at sign (@) in e-mail, to allow senders, receivers, and the computers that helped them to separate the recipient from the host.

As Ray Tomlinson burned up countless hours getting brief messages to go from one computer to another (notwithstanding they were in the same room), other boffins were developing ways and means of interconnecting computers that were more than a few yards apart.  The most significant ancestor of the Internet, Arpanet, was in its infancy, but what a precocious baby it was!  TCP/IP simply did not exist, and would not come to life for several more years.  This year is only the 20th anniversary of TCP/IP, which was unleashed in 1983, when Vinton Cerf and Robert Kahn got the scheme into tolerable shape and persuaded their friends and colleagues to take it out for a test drive.  If Arpanet was fire, TCP/IP was central heating.

The technologies that are most widely used to move e-mail around the globe may depend on TCP/IP, but they were actually developed somewhat independently, and would have been available to us even if other deep communications protocols had triumphed.  SMTP, now the standard way to transmit e-mail, didn't appear in a modern form until about 1981 and the POP3 protocol used by most of us to pick up e-mail messages came on the scene several years later.

You may take POP3 for granted, but it was quite an achievement.  Prior efforts to develop services for e-mail recipients were surprisingly inept.  POP3's pops were poop.  So, even though it seems as if e-mail has been with us a long time or even forever, it's simply not the case.

Ray Tomlinson
Ray Tomlinson
He knew where it was at

This is not the only popular illusion about e-mail.  The other widespread misconception is that e-mail has become very much like postal mail because it can transmit formatted messages, which can look like written letters or printed documents and can even include (as attachments or embedded objects) sound, video, and other media clips.

There is no question that e-mails are pretty versatile.  They can even be three-dimensional, in a sense.  Engineers and designers routinely exchange by e-mail STL files that direct three-dimensional printers (also called rapid prototyping systems) to create solid objects as output.

The misperception all this pretty (and pretty complex) e-mail creates is that an e-mail message is some kind of original document, when it is actually a facsimile, a copy of the material created at the sender's desk and, for the most part stored there (or wherever the sender chooses to keep a repository of outbound messages).  Furthermore, the original may never have all the properties of the complete delivered message.  A music file may have never been heard by its composer, which makes the writer resemble in a small way the deaf Beethoven.  An engineer's e-mail that creates a piece of an automobile dashboard or a replacement hip joint may have originated as a two-dimensional display depicting a three-dimensional object or as a figment of imagination.  It can still end up as a physical object generated by a three-dimensional printer or rapid prototyping system if that's what the recipient uses for an output device.

A very important distinction between e-mail and snail mail is its degree of privacy.  When you send a physical letter, only the envelope is exposed to view within the postal system, which includes people as well as machines.  E-mail is entirely exposed at every step along its path from sender to receiver.  It's all a postcard.  Encrypting e-mail is like putting a coded message on a postcard.  The content may be hidden, but the message simply screams that it is an encrypted object.  While the exposure of content makes it possible for spam catching software to work and for lots of other handy things to happen, it also can be a drawback if for some reason you really want that message to be private.

Another difference that distinguishes e-mail from snail mail is that e-mail postmarks are notoriously easy to copy.  Postmarks can be forged, but it is simply not a common practice.  Spam with spoofed headers is ubiquitous.  So while you can probably trust the header information in an e-mail from a correspondent you know or trust (with the qualification that the sending system's clock might not be perfectly accurate), you are silly if you put a lot of faith in the header information of an e-mail from some random sender.  E-mail desperately needs a simple, reliable mechanism--an optional one--that would assure recipients that a header is trustworthy.

It's too bad nobody has figured out how to create a Royal E-mail.

— Hesh Wiener November 2003

Copyright © 1990-2019 Technology News of America Co Inc.  All rights reserved.