<< Back to September 2005 CRN Table of Contents

[Published originally in the September 2005 edition of Computing Research News, Vol. 17/No. 4]

Expanding the Pipeline

Anywhere, Anytime—or Just Where is Your Office Anyhow?

By Susan Landau, Sun Microsystems Laboratories

My morning routine is to stop in the office early and see what has come in during the night. Then, over yogurt, cranberry juice, and The New York Times, I let issues sift and settle. Afterwards I reverse my commute—all thirty-four steps of it—and return to my study. I am a Distinguished Engineer at Sun Microsystems and I work from home.

I work ninety miles from the closest Sun office and across the continent from the one to which I report. Late afternoon, I might be walking my dog in conservation land at the end of my country street; minutes later, I might be developing the company's stance on digital-rights management with the Chief Technology Officer.

I do almost all my collaboration from home. I travel to Sun less than a dozen times a year; the rest of the time I use email, phone, and the Web. I call in for talks—indeed, I run a lecture series—and for meetings. I've even run a security review for a Sun product, spending a single day at the company and doing the rest of the work remotely.

The coming business trend is dispersed workforces. That's a natural fit for Sun (which “makes the net work”). Many companies are developing technologies for remote workers. Because I have first-hand experience with Sun's program, that's the one on which I'll focus.

When I was hired six-and-a-half years ago, the program didn't exist, and my manager and I worked everything out from scratch. It was not easy. DSL wasn't available, and the local phone company had never heard of Solaris (“What version of Windows is that?”). Updating my Linux box to accommodate a new AT&T calling plan could take hours. Getting the new Cisco router running was a challenge. But management was committed to my working effectively from home, and they made it happen.

Roughly half of Sun's employees, sixteen thousand people, spend at least twenty percent of their time working from home; fifteen hundred Sun employees spend at least three days a week at home. Some are hired into this situation; others move into it from a regular Sun office. Sun Labs, the research arm of Sun in which I work, has offices in Menlo Park, California, and Burlington, Massachusetts, and employees who work from home in Beaverton, Oregon; Seattle, Washington; Amherst, Massachusetts; Ottawa, Canada; Zurich, Switzerland; Wivenhoe, Essex, England; and Craigfad, Scotland.

Tools

The program works because Sun built the needed infrastructure. This includes educating and training managers and employees on everything from how to supervise remote employees and connect them to the company, to how to manage and effectively participate in a distributed meeting (Plan the meeting well. Create time for social chatting at the beginning or end of the meeting. Address behavior problems as they occur: say “We hear typing noise, please mute your call”). Sun provides “real estate” for its remote employees to use when they come to a work site: reservable offices, drop-in centers (no reservation needed), and iWork cafes (both types of Java available). Most interesting to me, though, are the technology solutions.

Accessline is a follow-me phone product: your phone number follows you from home office to office drop-in space to conference meeting room—wherever you want to get calls. You may not be in one place, but your callers reach you as if you are.

“Sun Forum” provides small groups of users (fifteen or fewer) shared access to electronic whiteboards, chat facilities, and audio and video conferencing.

Sun Rays are Sun's thin client/fat server solution. Evolved from “memoryless workstations” (do the computing locally, but memory is elsewhere), Sun Rays enable “hot-desking”—authenticate yourself to a machine and start your session, move to another office, reauthenticate, and there's your session. When you move offices, you don't move machines, you just move your ID card. The Sun Ray is a thin client; the guts of the system are in the Sun Ray server. Once there is bandwidth built up—and the United States has done so—supporting remote users is easy. Since updates, patches, and so on are all done centrally, Sun Rays are a great solution for remote users.

And Does the Program Work?

Sun likes the Remote Work from Home program because it saves money. You don't need an office for an employee who doesn't use one (the company has saved more than 7,700 office spaces). Employees enjoy the flexibility the program offers.

The experience is not all positive, however. The Labs' sysadmins—I’ve had three during my time at Sun—have been wonderful, but problems often occur when the sysadmins are not available. So I've learned to handle minor system administration myself.

Another complication is that I'm on the road a lot. I make about six week-long trips cross-country to Sun annually, and several shorter ones to the Sun office two hours away. This is in addition to my usual travel related to conferences, committees, and other professional activities.

In my time at Sun, I've done collaborative work with engineers on the Liberty Alliance specifications, developed Sun's digital-rights-management policy stance with the Chief Technology Officer, advised the public policy office on cryptographic export control and other security issues, and worked closely with Sun's Chief Security Officer on a variety of issues. I put effort into my work relationships and, as a result, I have a number of close ones. It is possible to build such relationships, but it is harder to do from a distance. The moral: working remotely is not for the socially impaired.

One problem I have not solved is that I don't participate in the informal discussions—the random idea brought up over latte, the talk someone heard at a meeting the week before and mentions at lunch—that are the lifeblood of a research lab. I was never one for the water cooler, but I've always enjoyed long dinners and discussion about new ideas, old ones, politics, science, and technology. I arrange this when I travel to Sun, but my connections to my colleagues are not as close, our conversations not as frequent, as they would be if I saw them daily.

Remote Work: A Solution to Many Complexities

These are minor cavils. The real point for me is that, like many professional women, I have a family and a “two-body” problem.1 Sun's willingness to have me work from home has enabled me to have a dynamic scientific career while living with my husband and kids. I feel very, very lucky.

The two-body problem is a serious roadblock for many women scientists who are often married to other scientists. (A 1992 article in Science stated that 69% of married female physicists were married to scientists, as were 80% of female mathematicians and 33% of female chemists.2) There are couples who solve their problem by finding two positions at the same institution, but there are not many such solutions. There are couples who solve their problem by finding two jobs nearby, but there are not a lot of research institutions close to one another.

Remote work from home will not work for everyone or in every situation. And it has its costs. But it is not just industry that can benefit from it. If the program works for industry, why not for academia as well?

Professors probably need to teach classes in person, at least some of the time. But Sun managers run engineering meetings remotely; why can't faculty run their research group meetings the same way? Not all meetings with students have to happen in person; college faculty could conduct some “one-on-ones” remotely (it would be a good industry learning experience for the students). Just as Sun's program expects employees to be available at times worked out with their managers, academia could have similar expectations for remote faculty. If industry can run successful team projects with participants in India, China, the U.S., and the U.K., surely academia can learn to work with professors who are remote three days a week. (Note the two days in the office shouldn't be fully scheduled, e.g., some meetings can happen remotely. Time for informal interactions—for the random conversation over tea and the chance encounter in the hallway—should be nurtured.)

The problem of the lack of women in science and engineering research has myriad causes.3 In order to make work from home effective, Sun built the necessary social and support infrastructure into its program. Similarly, our colleges and universities should be using creativity and energy to increase the participation of women in science and engineering. Enabling geographic flexibility for faculty and students is one more tool that academia can use to fuel that increase.

End Notes

1. S. Landau, “Universities and the Two-Body Problem,” Computing Research News, March 1994, p. 4.
2. Ann Gibbons, "Key Issue: Two-Career Science Marriage," Science, March 13, 1992, pp. 1380-1381.
3. See, for example, the recent report of the Harvard Task Force on Women in Science and Engineering, http://www.news.harvard.edu/gazette/2005/05.19/01-taskforce.html, which gives recommendations for all steps of women's careers: undergraduates, graduate students, postdocs, and junior and senior faculty.

Susan Landau (susan.landau [at] sun.com) is a Distinguished Engineer at Sun Microsystems Laboratories, and a member of CRA-W.