Meet the man who gave the world email attachments

We may earn a commission from links on this page.

It was 22 years ago yesterday that Nathaniel Borenstein sent his colleagues the world’s first email attachment: a picture of his barbershop quartet, The Telephone Chords. Borenstein, along with another researcher named Ned Freed, wrote MIME (Multipurpose Internet Mail Extensions) because one day he hoped to get pictures of his grandchildren over email. To the delight of grandparents everywhere, his technology became ubiquitous; MIME is still used for every email attachment, and has spread to host content all over the web.

Like his creation, Borenstein has been one of the web’s utility players ever since. In graduate school, he helped develop one of the first campus-wide networks. He got into digital commerce before eBay, digital payments before PayPal, and digital currency before Bitcoin. He has written dozens of articles, three books, and an award-winning essay on his time advising NATO on how to build war computers. These days he’s the chief scientist at Mimecast, a secure cloud service for business communications.

Email attachments are only three years younger than the web, which turns 25 years old today. Looking ahead at the next quarter-century, Borenstein told Quartz why we’re having the wrong conversations about security, how digital currency will survive Bitcoin, and why technology is easier to predict than human nature.

The first email attachment. Borenstein is the tenor on the far right.
The first email attachment. Borenstein is the tenor on the far right.
Image: Nathaniel Borenstein

The following conversation, conducted over the phone and by email, was condensed and edited for clarity:

Quartz: You’ve been involved in digital currency since the 90’s. Why do you believe Bitcoin is doomed?

Borenstein: I think the principal reason is it’s set up in a way to be extremely speculative, and there’s no way to see where the money is really flowing. The US currency used to be backed by gold. Now it’s backed by the faith of US government. Bitcoin is backed by wasting computational resources. It’s like the tulips in The Netherlands. What I didn’t anticipate, but didn’t make things better, was places like Mt. Gox going under. I think that’s going to deeply undermine confidence in Bitcoin. The fringe will continue to use it, because they think its a way to escape government currencies. But I think it’s built on a false foundation. My biggest concern is that the failure of Bitcoin is going to undermine enthusiasm for digital currencies in general. There are digital currencies that make a lot of sense.

Quartz: Some see applications such as Google Maps as on par with essential government services. Do you think the web will continue to create services that rival or replace what the government has to offer?

Borenstein: There’s a lot of reasons to think that anything that can be done with a market infrastructure can be done better than by a top-down government. What can actually be done that way? I don’t think that is as clear cut. In the case of web services, the government has long been in the business of providing maps. A lot of that original government work underlines services like Google Maps. Now Google’s doing their own corrections, typing over bad government functions. There’s nothing wrong with that, until it becomes a monopoly. This is not a political question, it’s a quality question. When there’s only one player, they inevitably have insufficient incentive to stay as good as they were. If Bing were more credible, I suspect Google Maps would be evolving faster and getting better.

Quartz: How do you see privacy changing?

Borenstein: People worry so much about their privacy going away, but I don’t think anyone has explained how it can be prevented. A lot of that effort of going into the privacy implications of new technology is wasted. People used to talk about hacking in Dick Cheney’s pacemaker, and it could have been done. You’re going to be walking around with a whole host of new devices in your body. They’re there because they’re good for you. You’re better off finding out about cancer when it’s seven cells than when it’s a tumor. This also means there’s info about you that is potentially leakable to the public. As horrible as it sounds, I don’t see any way to have the benefits of technology without having the risks to privacy. There’s never been a system that can’t be hacked. So I guess I get frustrated with all the privacy discussions, because they could be spent doing something more productive.

Quartz: How could these conversations be more productive?

Borenstein: Well, first of all, I think there’s a difference between privacy and security, and it’s something that gets confused a lot in the media. There are issues of real security. If you identify that with a small number of things—like how the government designs security around nukes, for instance—you can design a small system to protect these things. They have a specific threat.

I think it’s worth noting that the smaller these networks are—the less they do, in particular—the more secure they will be.  I wouldn’t want to see a health care-specific network that all providers used instead of their company Intranet.  I’d rather see a special purpose network for sending around, say, patient information and similarly sensitive stuff only.  Now if your office manager accidentally downloads a virus by mail, it’s not on the patient information network.  The next question is, how hard is it to get to the patient information network from the regular one?  You could set it up so that it requires physically going to another machine, you could make it a simple proxy, or anything in between.  The easier you make it, though, the less secure it will be, and the easier it will be for that virus to hop from the manager’s mail to the secure network.

Quartz: There is an arms race between many governments to develop things like quantum computing. How will this technology affect the web and the world?

Borenstein: I’ve not had experience with being the first with a new computing power advantage, but I’ve had other advantages, and I think people fool themselves about what’s important. I have, over the years, become deeply skeptical about first mover advantage. In fact, I have a hard time finding any examples of it. IBM was late to computers, then decades later was late to PC’s. Oracle didn’t invent the database, Lotus didn’t invent the spreadsheet, Microsoft didn’t invent operating systems or window systems, Apple didn’t invent the GUI, Google didn’t invent the search engine, and Facebook didn’t invent social networks. First-mover advantage is probably possible for one-time things. The first bad guy with mega-computation might stage the biggest crypto-related heist in history, but he won’t have a sustainable advantage.

Quartz: Pulling back and taking a broad look at the future of the web, what areas are predictable, and what aren’t?

Borenstein: I’ve never gone wrong with predictions based on Moore’s Law.  I know, there are recurring articles predicting the end of Moore’s Law, but they’re never right, and I wouldn’t be surprised if it continues down to subatomic levels. It’s Moore’s Law that predicts the insanely small size and enormous power that computing devices will have, so I feel safe there. In fact, some of my poorer predictions have come down to a misapplication of Moore’s Law. I predicted in the 80’s that we wouldn’t have decent speech recognition in my lifetime, because I thought we’d have to understand language itself, for which we don’t even have an adequate theoretical framework. It turned out to be amenable to a brute-force approach, which was made practical by Moore’s Law.

The hard predictions are the ones that involve human nature.  It isn’t that human nature is such a mystery, but rather that it’s complex and in conflict with itself.  We want everything to be easy, but we want it to be secure.  We want open access to documents, except when we don’t. Some of us want digital currencies, while some of us—including most governments—are threatened by them. A lot of technological predictions involve social predictions in disguise.

The area where I feel most uncertain relates to wearable computing: how much will people accommodate themselves to such devices? Will they be acceptable in bars? Job interviews? Will the invisible versions be illegal? My guess is that people will end up accepting them, and it will be a big step towards the day we become actual cyborgs. But I wouldn’t care to predict when that day will be.

Quartz: Are you still performing with your barbershop quartet?

Borenstein: Alas, I mostly stopped making music in groups in the late 90’s, when life sort of got away from me. I’ve been talking about taking it up again almost that long. But thank you for asking!