The Email Fave Button Could Actually Happen
An Internet-standards expert says the miracle feature is “perfectly plausible.”
In the dark days of 2011, it was a common refrain: Why isn’t there an Instagram for video? Now we have Vine, and all is right in the world.
But there’s another feature, actually, that’s somehow discussed everywhere but implemented nowhere. I speak of course of the fave—or “like”—button for email.
The Atlantic’s own senior editor, Dr. James Hamblin, proposed the idea in a video last week:
But the proposal has a long history: Buzzfeed senior editor Katie Notopoulous recently listed the fictional feature as one of 92 free ideas. (It was Idea #64, right after #63, “Shazam for plants.”) Tweets requesting the feature appear as early as 2011.
Why? My colleague makes the case better than I ever could, but—for me—the optimal fave-for-email would allow for a fast, easy reply. On both Twitter and Facebook, a fave or like can symbolize a stern nod, a warm hug, an IOU, or a farewell wave. A fave can be many things, and an email fave would contain them all.
It would be, in short, the best feature ever.
I, too, had long dreamed of the email fave, but I also figured it was impossible. Email is pretty old, even judged against its fellow core components of the Internet. It existed in rudimentary form in the 1970s. Simple Mail Transfer Protocol—the specification that’s still used today to hustle modern emails between servers—was in place by 1982. As such, I believed email was locked in, finished, feature-complete. And even if email could be changed, I knew it didn’t work like other parts of the web. Though it sounds obvious, email can’t be changed like Facebook: It’s not a singular product that can be unilaterally altered by a single institution. Email is a set of standards, ratified by an engineering group and implemented independently by private developers. How can you change that?
I talked to Pete Resnick to find out. Resnick is one of two directors of the applications area in the Internet Engineering Task Force (IETF). The IETF is the engineering standards group in charge of a swath of Internet technologies, including the crucial Internet protocol suite, TCP/IP, and... email. If email were to get a fave button, Resnick would be involved. When we spoke, he said implementing an email fave was “perfectly plausible.”
“Email has been around for a long time, and it’s exceedingly extensible,” he added. He even has an idea about how to implement such a feature.
Two main specifications dictate the basic behavior of the modern-day email infrastructure. The first is RFC 5322, “Internet Message Format,” which defines what an email is, how individual emails should be formatted, and what information they need to contain. The second, RFC 5321 or “Simple Mail Transfer Protocol,” specifies how servers should send and receive those messages. Between the two—one that defines email, another that says how to move them—the basic functions of email emerge.
Those two specifications are widely modified and extended to allow email to do and be more things. And there’s an existing extension, Resnick said, which would adapt well to the email fave: RFC 3798. RFC 3798 is designed to send read-receipts—little messages to certify that a message was read, which, crucially, the sender usually asks for beforehand—but it could easily be adapted to allow anyone to reply-all with an “acknowledgement” message.
Resnick quickly sketched the hypothetical: Hit the “acknowledge” button—or the fave, like, or the A-OK—and a quick, auto-generated email would go out to everyone on a message’s thread. Older software would show that message as a regular email, with text like, “At 12:34, Robinson Meyer acknowledged your message.” But newer clients could render it as a single line at the end of the sender’s message, perhaps with all the acknowledgers folded into a sentence as they are on Facebook: “Robinson Meyer, Adrienne LaFrance, and Megan Garber liked this.” In other words, every fave wouldn't hit your inbox like one of those annoying auto-reply vacation messages. The integration could be seamless for both sender and receiver.
“We don’t specify user interface stuff except in broad terms,” said Resnick. So decisions like the ones I just sketched would be left up to individual email clients. But decisions like that could be made—and if the feature was approved, they would be.
How to get the feature approved? You’d have to go through the IETF, the organization Resnick helps to lead, makes Internet standards. It also has an open membership. That means anyone can join it—all they they have to do is sign up for a listserv. (A fitting irony that, to shape the future of email, you have to sign up to get more email.)
Because the IETF’s membership structure is so open, it doesn't use voting to decide which technical specifications to adopt or how those specifications should be written. (If it did, literally anyone on the Internet could show up for any election.) Instead, it convenes working groups around possible new features and appoints a chair for each. The group then makes decisions by coming to “rough consensus,” which it’s the chairperson’s job to suss out and arbitrate. While the entire group doesn’t have to agree completely on every specific, the chair must feel that any dissents have either been addressed or technically dealt with.
Resnick mentioned a saying of the influential American computer scientist, Dave Clark: “We reject kings, presidents, and voting. We believe in rough consensus and working code.” The key to a successful standard proposal, he said, is working code, and preferably, working code written in different implementations by different people.
When a working group submits their proposal, it goes to the IETF’s senior committee: the Internet Engineering Steering Group, which Resnick sits on. That organization can either approve proposed standards or send them back down to the working group for further plans. (Notably, they can’t reject them.)
It’s not like email hasn’t been extended recently. Just two years ago, the IETF finally “internationalized” email addresses and permitted them to include letters and symbols beyond the basic (and English-centric) 128 ASCII characters. An email fave button would be much, much easier to write and specify.
What’s holding us back, Internet?