What Can the Independent Web Learn From the Internet Technology That Never Really Went Away?
Email is more than fifty years old. It has survived countless technological revolutions, the rise of social media, and an endless parade of supposedly better ways to communicate. Perhaps there is something we can learn from that. A few days ago, I came across an essay by Sergey Kaplich titled Everyone Has Email. Nobody Uses...
mail is more than fifty years old. It has survived countless technological revolutions, the rise of social media, and an endless parade of supposedly better ways to communicate. Perhaps there is something we can learn from that.
A few days ago, I came across an essay by Sergey Kaplich titled Everyone Has Email. Nobody Uses It Right (https://kaplich.me/essays/email).
At first, I wasn’t particularly enthusiastic. Another article telling us how to manage our inboxes? Haven’t we had enough of those? Filters, folders, unsubscribing from newsletters, achieving Inbox Zero. We’ve been discussing these things for decades, and somehow our inboxes are still overflowing.
But Kaplich makes an interesting observation before getting into all that productivity advice. Email is one of the oldest communication technologies on the internet, yet it continues to work remarkably well. It doesn’t belong to a single company. You don’t need to use the same application as the person you’re communicating with. And conversations from twenty years ago can still be sitting in your archive, perfectly readable.
That got me thinking about a rather different question. Why has email survived almost everything the internet has thrown at it? And what might that teach us about the independent web we’re trying to build today?
The technology that refuses to disappear
Think about all the communication technologies that have come and gone since email first appeared on ARPANET in the early 1970s. Bulletin board systems. Usenet. IRC. ICQ. MSN Messenger. MySpace. Google Wave. Google+. And plenty of others that once seemed destined to become the future of online communication. Some disappeared completely. Others survive in smaller communities. And some evolved into technologies that remain important today. Meanwhile, email just kept going.
We now have WhatsApp, Signal, Telegram, Discord, Slack, Teams and countless other ways to communicate. Some are much better suited to particular tasks than email. I certainly wouldn’t want to organise a fast-moving group conversation through an endless chain of emails.
Yet none of these newer systems has made email obsolete. In fact, many of them depend on email. You use an email address to register, recover your password, receive notifications or confirm changes to your account. That’s quite an achievement for a technology that predates the World Wide Web by almost two decades.
According to projections published by the Radicati Group (https://www.radicati.com/), hundreds of billions of emails are sent and received every day. Of course, that includes enormous quantities of automated messages, marketing emails and spam. It doesn’t necessarily mean that people enjoy using email. But it does tell us something about how deeply embedded email has become in the infrastructure of the internet.
Email isn’t an app, tt’s an agreement
I think this is where things become interesting. WhatsApp is a service operated by Meta. Slack belongs to Salesforce. Discord operates its own platform. These companies determine how their services work, which features are available and what happens when their business priorities change.
Email works differently. It’s built on a collection of open standards. SMTP handles the transfer of messages between mail systems. IMAP allows applications to access messages stored on servers. MIME made it possible to add attachments and different kinds of content.
These standards don’t prescribe which application you must use. You can write an email in Thunderbird on a Linux laptop and send it to someone using Gmail in a browser on a telephone or a Windows dekstop. That person might reply using Outlook on a tablet. Someone else might read the same conversation on a phone running Android or Sailfish OS.
Imagine if every telephone manufacturer required you to use the same brand of telephone as the person you wanted to call
Nobody has to install the same app. And nobody has to convince all their friends and colleagues to move to a new platform just because they prefer a different email client. The underlying system simply allows different implementations to communicate. This is such an ordinary part of our digital lives that we rarely stop to appreciate how unusual it has become. Imagine if every telephone manufacturer required you to use the same brand of telephone as the person you wanted to call. That’s more or less how many modern messaging platforms operate.
The secret might be evolution rather than revolution
There’s a fascinating technical document from 2009 called Internet Mail Architecture, written by Dave Crocker and published as RFC 5598 (https://www.rfc-editor.org/rfc/rfc5598). It describes the architecture of internet email and how its different components work together. One of the things that struck me is how email developed through a long series of evolutionary changes. When people wanted to send attachments, additional standards made that possible. When they wanted to access messages from different devices, email clients and servers evolved to support that. When spam and fraudulent messages became serious problems, new authentication mechanisms were developed.
Not always elegantly, I should add. Email’s security architecture is a complicated collection of mechanisms accumulated over many years. And some of its problems remain frustratingly difficult to solve. But the important point is that new capabilities were generally added without requiring everyone to abandon the existing system. An email address created years ago can still communicate with one created yesterday. Old messages don’t suddenly become unreadable because a company has decided to launch an entirely new version of its platform.
Email has been able to change without constantly asking the world to start again. That seems like a valuable design principle.
Open doesn’t automatically mean independent
Before we get too enthusiastic, there’s an important complication. Email may be based on open standards, but the modern email ecosystem is heavily influenced by a relatively small number of large providers. Google and Microsoft, in particular, have considerable influence over whether messages are successfully delivered. Their spam filters, authentication requirements and reputation systems can make life difficult for people running smaller mail servers. Technically, you can operate your own email server. Practically, getting your messages reliably delivered can be surprisingly challenging. Another issue: if you’ve used the same Gmail address for twenty years, your digital identity may be much more dependent on Google than you realise.
There’s an important distinction here between an open protocol and a genuinely decentralised ecosystem. The first makes independence technically possible. It doesn’t guarantee that independence will be easy to achieve. Email also has serious problems with spam, phishing, privacy and security. Its original design emerged in a much smaller internet, where the threat environment was very different.
So I don’t think email should be treated as some perfect example of how everything ought to work. It’s much more useful as a real-world experiment that has been running for over fifty years. And one of the lessons of that experiment is that open standards alone aren’t enough. We also need healthy ecosystems, multiple independent providers and practical ways for people to move between them.
We’ve been here before
There’s another interesting part of email’s history that I hadn’t really considered before. SMTP wasn’t the only attempt to standardise electronic messaging. During the 1980s and 1990s, X.400 was developed as part of the OSI family of standards. It had substantial institutional backing and was designed to support sophisticated messaging requirements. Yet internet email eventually became the dominant approach.
Researcher Kai Jakobs examined this history in a paper titled Why Then Did the X.400 E-mail Standard Fail? (https://aisel.aisnet.org/jit/vol28/iss1/5/). The explanation is more complicated than simply saying that one technology was better than the other. Timing, implementations, adoption and the way standards were developed all played important roles.
That suggests something worth remembering when we discuss today’s alternatives to dominant internet platforms. A technology doesn’t necessarily succeed because it has the longest feature list or the most sophisticated architecture. It also needs to be accessible, useful, widely implemented and capable of growing without breaking what already works. And once people start building applications and communities around it, something important happens. The technology becomes infrastructure.
The internet doesn’t need another everything app
One of the things I find slightly exhausting about modern digital services is their ambition to become the centre of everything. Messaging apps want to become social networks. Social networks want to become shopping platforms. Productivity tools want to become communication platforms. And almost everybody wants to add an AI assistant. There’s nothing inherently wrong with adding useful features. But it often means that more and more of our digital activities become dependent on the decisions of a single company.
Email took a different route. At its core, it transports messages between addresses. What those messages mean, how they’re written, how they’re displayed and what people do with them are largely separate questions. A personal letter, a newsletter, a purchase confirmation and an automated server notification can all use the same underlying infrastructure.
And people are free to build different applications around it. That separation between infrastructure and application may be one of the reasons email has lasted so long. It doesn’t need to be the perfect application for every purpose. It only needs to provide a sufficiently useful foundation on which other things can be built.
Now look at the independent web
This brings me back to something I’ve been writing about quite a lot on Just Me lately. There’s a growing collection of projects and communities trying to make the internet a little less dependent on large, centralised platforms. The IndieWeb movement encourages people to publish on their own websites. RSS allows readers to follow publications without depending on a social media feed. ActivityPub provides a foundation for federated social networks. Markdown makes it easier to keep written content in a format that isn’t tied to one particular application.
I’ve also been exploring older and alternative internet technologies such as Gopher, Gemini and Yggdrasil. These projects are very different from one another. They solve different problems, operate at different levels and have very different communities. But many share an underlying idea. The infrastructure should give people the freedom to choose their own tools, rather than requiring everyone to live inside the same platform. And that is precisely what email has been doing for decades. Not perfectly. Not always conveniently. But successfully enough to survive enormous changes in the internet around it.
The value of being boring
There’s a wonderful irony here. In technology, we tend to celebrate disruption. New products are exciting because they promise to replace something old. New platforms are interesting because they introduce entirely new ways of doing things.
But perhaps the technologies that matter most are the ones that eventually become boring. Nobody gets particularly excited when an email travels successfully from one provider to another. We simply expect it to work. The same is true of DNS, HTTP and many other foundational internet technologies. They’re not always visible to ordinary users, but enormous parts of our digital lives depend on them. And their longevity gives people something that fashionable applications often cannot: a reasonable expectation of continuity.
That matters to me. It’s one of the reasons I prefer keeping my writing in Markdown files. I love Markdown because I don’t want my words to become permanently trapped inside an application simply because that application happens to be popular today. Markdown is also particularly convenient when working with AI, whether you’re using a cloud-based service or running a language model locally. Its simple, readable structure makes it easy for AI tools to process, analyse and work with your content, without depending on proprietary file formats.
The same principle applies to communication. A conversation shouldn’t have to disappear because the company operating the platform has gone out of business, changed direction or decided that an old feature is no longer worth maintaining. Of course, email doesn’t automatically solve that problem. You still need to manage your archives and choose your providers carefully. But its architecture at least gives you options. And options are valuable.
A small, surprisingly complete internet
Here’s a thought I rather like. Imagine someone who has a personal website, publishes articles there, follows other websites through RSS and communicates directly with people through email. That’s already a remarkably complete personal internet. They can publish, discover, read, respond and build relationships. They don’t need an account on a particular social media platform for any of those activities. Of course, they might still choose to use social media, messaging apps or online communities. There’s no reason to turn digital independence into a competition about who can avoid the most platforms.
The important difference is that those services become optional additions rather than essential foundations. A personal website provides a place to publish. RSS provides a way to follow others. Email provides a way to communicate. All three are based on technologies that have been around for decades. Perhaps building a more independent internet doesn’t always require inventing something new. Sometimes it means recognising what we’ve already built, understanding why it survived and finding new ways to use it.
Maybe survival is the most interesting feature of all
I started thinking about this because of an essay about email productivity. I ended up somewhere rather different. Email’s survival doesn’t prove that open standards always win. Nor does it prove that decentralised systems are automatically better. They have their own problems, and email demonstrates several of them very clearly.
But it does show that an internet technology can remain useful across generations of devices, companies, applications and user expectations. It can evolve without requiring everyone to migrate simultaneously. It can support competing implementations. And it can allow people who have made completely different technology choices to communicate with one another.
Those are not particularly fashionable features. But perhaps they’re exactly the features we should be paying more attention to. Because when we talk about building a more independent web, we often focus on the next exciting project, the next new protocol or the next alternative to a dominant platform. Meanwhile, one of the internet’s oldest technologies is still quietly demonstrating what long-term interoperability can look like.
Maybe the real question isn’t why email is still here after fifty years. Maybe it’s why we keep building new communication platforms that forget so many of the things email got right.
Did you enjoy this article?
Recommend it — Standard Reader surfaces well-loved writing to more readers across the network.