Skip to Content

E-Mail You Can’t Ignore

A new program from Microsoft learns who’s important in your life and puts their messages at the top of your inbox.
December 12, 2005

Many workers will return from their holiday vacations to an avalanche of unread e-mails. And sorting the important ones from the trivial might just exhaust any holiday goodwill – especially now that three-quarters of all incoming office e-mail is junk, according to research firm Gartner.

There are new solutions to manage one’s mailbox, however, that combine software and sociology. Going beyond existing measures, such as spam filters and blacklists, these newer applications prioritize incoming e-mail by studying the patterns of human interaction.

Microsoft Research released one such program on November 30. The free download is called SNARF, for Social Network and Relationship Finder. It runs alongside Microsoft Outlook (2002 and newer versions), poring through e-mail histories and following chains of communications to ferret out the unread messages it deems most important.

SNARF measures a sender’s importance based on two key factors: the number and frequency of messages sent and received. The program then sorts unread e-mails into three fields: messages where the user is listed in the To or CC fields, group e-mails, and all messages received in the last week. SNARF lists messages by senders, rather than subject lines, and puts a user’s most important correspondents on top.

Danyel Fisher, a researcher in Microsoft’s Community Technologies Group and a member of the SNARF development team, believes SNARF’s effectiveness lies in its simplicity. “We’re just counting e-mails,” Fisher says. “Some people might call it a brain-dead algorithm, but the messages you send someone is a pretty good proxy for how well you know people,” he says. “It can be very detailed.”

Microsoft has no plans yet for incorporating SNARF into a commercial program such as Outlook, but Fisher says social networking will continue to be a theme in the company’s product development. “Even though this is just demonstration software, it’s telling us there’s a hunger for new ways of thinking about e-mail,” Fisher says. “We’ll be taking into consideration what’s important, and how people use it.”

Not surprisingly, Microsoft isn’t the only company that see the benefits of adding social networking intelligence to productivity applications. LinkedIn is a web site that leverages the six-degrees-of separation principle to help professionals network. It expects to release an update to its Outlook toolbar application in January that will prioritize e-mails by finding personal relationships among other LinkedIn members.

“E-mail is an essential tool, and people are having a problem managing it,” says Konstantin Guericke, co-founder of LinkedIn, who believes the network of contacts users have already made on his company’s service can also help organize their messages. “We have a network and we want to use it to be applicable to people’s lives – not just every two years when they’re looking for a new job,” he says.

To Guericke, the LinkedIn toolbar will improve on SNARF by addressing broader social criteria than e-mail histories. “Our view is that you need a mix of implied relationships,” he says, stressing the importance of less-obvious contacts, like new business leads, that e-mail histories can’t divine.

Guericke noted that the new toolbar will take advantage of the LinkedIn web site to recognize messages from new senders within one’s existing social network. “Leveraging e-mail patterns is a good source of data, but profile and relationship information is also needed to get it working well enough that users will find it useful,” he says.

To some technologists, however, building social awareness into e-mail systems is overkill. Several existing applications already organize and prioritize incoming e-mail better than algorithmic filters, says Ross Mayfield, chief executive of Socialtext, a networking application that uses community-editable bulletin boards, called wikis, to entirely replace e-mails in project management.

“There are times when you come back from vacation and you’ve got two thousand e-mail messages, and [SNARF] could provide one dimension to sort that data,” Mayfield says. “If I were to advise users, however, I’d have them use this as a filter of last resort. A ranking of what might be more important to you, based on past behaviors, may only lead to the creation of a different echo chamber.”

While Mayfield believes group-based wikis are one way to reduce interoffice e-mails, he also thinks that a focus on e-mail searching, rather than sorting, can help sift through bloated inboxes, while minimizing time reading irrelevant e-mails. “You don’t need a filter up front if you have a strong search to fall back on,” he says, citing the integrated search capabilities of mail systems like Google’s Gmail and Yahoo Mail.

Of course, no system so far can identify and appropriately emphasize messages from the connections you make outside the electronic world – the person with whom you just swapped business cards, the long-lost cousin, or your friend with a new e-mail account.

Coye Cheshire, a professor in the School of Information Management and Systems at the University of California, Berkeley, who studies social exchanges online, believes imposing rules – even those derived from previous habits – limits communication. “We can probably make good assumptions based on e-mail use, but we have other interactions they’re not taking advantages of yet,” he says.

Cheshire believes these types of problems will be addressed by the next generation of socially-aware tools, which he hopes will integrate networking intelligence into everyday communications, such as phone conversations and instant messaging. “As we see things moving more towards electronic-based forms of communication, telephone lines moving to electronic broadband lines, it raises the possibility of greater interactive communication,” Cheshire says.

Keep Reading

Most Popular

It’s time to retire the term “user”

The proliferation of AI means we need a new word.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Sam Altman says helpful agents are poised to become AI’s killer function

Open AI’s CEO says we won’t need new hardware or lots more training data to get there.

An AI startup made a hyperrealistic deepfake of me that’s so good it’s scary

Synthesia's new technology is impressive but raises big questions about a world where we increasingly can’t tell what’s real.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.