This article originally appeared in TidBITS on 2009-10-29 at 2:54 p.m.
The permanent URL for this article is:
Include images: Off

Why Email Remains the King of Internet Communications

by Adam C. Engst

The most recent "email is on its way out" meme started with an article in the Wall Street Journal by Jessica Vascellaro entitled "Why Email No Longer Rules... And what that means for the way we communicate. [1]" The article makes the obvious point that an increasing number of people - particularly the young - prefer to use Facebook and Twitter and instant messaging and text messaging instead of email.

But like so many articles predicting the death of (or even eulogizing) email, this one misses some important points about why email won't disappear in the foreseeable future (and why it's not even waning now).

It all comes down to two simple facts: email is based on open standards, and it's the lowest common denominator for Internet communication. Any communication system that wishes to supplant email will need to offer both openness and ubiquity, and nothing available today comes even close.

Open Standards -- Email has been around since the very earliest days of the Internet, and even before that, with some claiming the first intra-computer email system originating at MIT in 1965 and Ray Tomlinson of BBN receiving credit for the first inter-computer email system in 1972. Since its inception on the Internet, the basic email standards have evolved some, but honestly, not all that much. Instead, most of the evolution in email has been improving performance and scalability, while continuing to adhere to the basic interoperability standards.

Open standards are important because they allow any programmer to write a new email server or client (in the jargon, a "mail transport agent" or "mail user agent"). Hundreds of email servers and clients have been developed over the years for every operating system known, and while slight compatibility troubles have always existed, the adherence of these programs to the open standards that define Internet email means that any client can work with any server, and all servers can communicate with one another. Interoperability is key.

(To be fair, both Facebook and Twitter have open APIs that enable programmers to write applications that work with the services, but very much in a subservient way. No one can write their own Twitter or Facebook server, though there is an open source attempt to recreate Twitter, called StatusNet [2].)

By relying on a healthy ecosystem of clients and servers communicating using open standards, email gains two huge advantages over Twitter and Facebook.

First and most important, businesses and government organizations need to control their own communications, with as few intermediaries as possible to prevent confidential communications from being seen by outsiders. Can you imagine Apple engineers discussing the next iPhone via Facebook? Why do I suspect that's a firing offense?

Second, an ecosystem based on open standards has no single point of failure. Twitter suffers downtime regularly, though less frequently than in the past, and although I get the impression that Facebook has fewer reliability troubles, it has certainly been inaccessible at times too.

The single point of failure worry exists at the corporate level too. Twitter has no proven business model, much less profits, and while Facebook makes money from display advertising, the company has just crossed into being cash-flow positive, a step on the way to making a profit. Businesses will bet on large, profitable companies sticking around - witness the increased use of Google Apps in the business world - but no company in its right mind will hand essential communications services over to a partner that's not only unprofitable, but also lacks a proven business model.

That said, with Facebook boasting 300 million regular users and Twitter serving another 30 million or so, I'd be shocked if either company were to disappear entirely - some deep-pocketed megacorp like Google or Microsoft would be more than happy to snap them up were they to fail to generate sufficient revenue to survive.

In short, proprietary services are susceptible to all sorts of troubles that simply can't affect more than a small subset of users of an open standard like email.

Lowest Common Denominator -- Let's look one step deeper. Email's open standards, and the ecosystem that has grown up around email because of them, mean that email is the lowest common denominator for one-to-one and one-to-many communication on the Internet today. In contrast, Twitter and Facebook and their ilk are entirely proprietary services.

A basic truism of communication is that any two parties communicating must find a common channel before any significant communication can take place. Whether it's two individuals determining which common language they share or a pair of modems negotiating the fastest common protocol they share (remember those screeching tones before the connection was established?), setting up a shared communication channel is key.

So when it comes to Internet communications, Facebook may have 300 million users, but that's only a fraction of the 1.4 billion people who have Internet access. Heck, just the big four email systems - Yahoo Mail, Hotmail, Gmail, and AOL - may have as many as 650 million users, and I suspect that email usage is a long tail phenomenon, with many more users total at smaller systems.

In essence, everyone who pays for Internet access or receives it from a school, business, or governmental organization has an email address. The main exception I can think of would be kids who use their parents' Internet connection. Email being the lowest common denominator of Internet communications means five things.

I'm not saying that email is the best imaginable solution in each of these situations, merely that it is the best and most ubiquitous tool we have to solve them today. And realistically, Twitter and Facebook aren't likely to fade away any time soon either. New and old technologies usually coexist happily for a very long time unless one can completely replace the other. Email and SMS text messaging have eliminated the need for many phone calls, but no one's suggesting that phone calls will stop happening entirely.

Improving Email -- While I am arguing that email won't fade away in the foreseeable future, I also think that Internet communications systems need to evolve to graft the key aspects of email with the lessons learned from Facebook and Twitter, while working to eliminate problems that have long dragged down email.

The main way Twitter and Facebook differ from email is that they evolved largely from the concept of blogging. Email is essentially a bidirectional communication medium in which you have a pretty good idea who reads what you write, whereas blogging is a publishing medium in which your audience is largely unknown to you.

The genius of Facebook and Twitter is that they combine publishing with bidirectional communication - in both you can post something that all your friends or followers can read, or you can direct a message at a specific person or group of people (the latter is trickier in Twitter, but is possible; that's another article).

And in fact, the particular genius of Twitter over Facebook is that it better honors the publishing model of blogging by allowing asymmetric following. That is, I can follow bike racer Lance Armstrong [3] because I'm interested in what he has to say, but Lance doesn't have to follow me back.

That's a far more comfortable relationship approach than on Facebook, where people who want to follow what I have to say automatically have to be my "friends," and by allowing them to read what I write easily, I have to open myself up to what they write as well. The fact that Facebook has had to add the capability to hide updates from particular people and applications is a hint that their system doesn't map well to the way relationships work in the real world.

The closest we have today, I'd argue, is Google's Gmail, which attempts to rethink what email is all about. Gmail's Web interface treats email as a constantly connected stream of information, and its constrained text-editing environment encourages the kind of quick, concise responses that you see on Facebook and Twitter. Plus, its approach of grouping messages in the same conversation provides the same sort of historical context as Facebook posts with their comments, and with the generally weak threading provided within some Twitter clients.

Obviously, Gmail is a service and suffers from lack of control and somewhat from the single point of failure problem, but because it also supports email's open standards, you can at any time switch to a different client or even use a different ISP. And at least Google is a large, profitable company that can't disappear without warning.

Although it isn't yet available to everyone, Google Wave [4] is attempting an even more significant rethinking of how we communicate on the Internet. The question is, will Google make Google Wave sufficiently generic and open that it could possibly approach the ubiquity of email? Or perhaps Google Wave will absorb classic email, layering its new ideas on top of the older protocols for backward compatibility?

No matter what, the other huge challenge that email systems have at best swept under the rug is spam. The extreme openness of email has allowed it to be abused in ways that are either difficult to replicate via Facebook and Twitter, or which programmers at those services have been able to stop, thanks to their complete programmatic control over the system. Spam filtering software at both the server and client levels has improved to the point where it's possible to avoid seeing much spam in your Inbox, but it's still common to lose the occasional legitimate message to an overenthusiastic spam filter.

Will email systems evolve to meet these needs? Or will we simply keep patching the old protocols, servers, and clients because we don't have the willpower to force a major change that would disrupt everyday email communications across the Internet? My money is on the latter, which is a shame, because as much as I don't see email disappearing any time soon, I would like to see it improve and learn from more-modern services like Facebook and Twitter.

Oh, and that Wall Street Journal article predicting the end of email's reign as the king of Internet communications? The most telling part comes at the end, in the byline, which makes the ultimate case for why email isn't going away any time soon. How else would you contact Wall Street Journal reporters?

-Ms. Vascellaro is a staff reporter in The Wall Street Journal's San Francisco bureau. She can be reached at [redacted]