the User Interface Design group: time for tea

  1. Prof. Rob Miller describes the beauty and wisdom of crowd computing to Campus Preview families in April, 2011 in the Grier Rooms.
    Prof. Rob Miller describes the beauty and wisdom of crowd computing to Campus Preview families in April, 2011 in the Grier Rooms.
  2. Prof. Rob Miller describes the beauty and wisdom of crowd computing to Campus Preview families in April, 2011 in the Grier Rooms.
    Prof. Rob Miller describes the beauty and wisdom of crowd computing to Campus Preview families in April, 2011 in the Grier Rooms.
  3. The User Interface Design group develops and studies new user interface techniques, primarily focusing on the following areas: Crowd computing, UI Automation & Customization, and Software development. Visit Rob Miller's User Interface Group website: http://groups.csail.mit.edu/uid/index.shtml
    The User Interface Design group develops and studies new user interface techniques, primarily focusing on the following areas: Crowd computing, UI Automation & Customization, and Software development. Visit Rob Miller's User Interface Group website: http://groups.csail.mit.edu/uid/index.shtml

Time for Tea? Yes! You’re invited to Professor Rob Miller’s User Interface Design Group – daily meetings.

What does it take to entice a large group — a dozen graduate and undergraduate students on average — to work collaboratively? EECS faculty member Rob Miller has come up with the paradigm for his students: daily ‘tea’. Though tea may not really be served, students in the User Interface Design (UID) group have come to gather each day from 1 to 2pm to collaborate. Talking informally but informatively about each other’s projects builds not only a sense of community, but a supportive way for each student to progress toward his/her individual goals. (In fact, Rob Miller finds that students meet for ‘tea’ whether he is there or not–a very positive measure of the effectiveness of these meetings.)

As a former PhD candidate in computer science at Carnegie Mellon and earlier an undergraduate and M.Eng. student at MIT, Rob Miller notes: “Certainly my mentors at CMU and MIT strongly influenced what I hoped to achieve. Brad Myers’s Amulet group at CMU, and Charles Leiserson’s Cilk group here at MIT, were models for the way I run my group.”

Miller also credits his luck in being able to work with “some truly outstanding students, who attract and bring in other fantastic students, because we strive to have an open, welcoming group.”

Finding the time for working individually with a lot of students is challenging Miller admits. In fact, part of the motivation for daily group meetings is to simplify his schedule by relying on a set time for each daily meeting. He is pleased that a lot of mentoring ends up happening in these group meetings — in a natural way, and, not just mentoring by him but by senior students. He also meets one-on-one with students when needed or desired.

Another motivation for the daily meetings is the pure excitement that seems to generate from the direction of the research. The UID group mostly focuses on problems that involve people and programming. They build novel tools, interfaces, and systems that make software easier, more productive, and/or more powerful. Miller describes the nature of the group’s work: “Building systems is one defining characteristic of MIT’s approach to HCI: we’re engineers. We don’t just study the way things are today; we build new things to make the world better. Robert Kennedy (quoting George Bernard Shaw) put it well: Some people see things as they are and say why? We dream things that never were and say why not?”

Working with crowd and cloud computing and seeing the way ahead in these systems

Most of the group’s recent work focuses on collaboration and interaction, learning how to orchestrate contributions from a group of people over the web. “We’re living in an exciting time,” Miller notes. “Not only do we have cloud computing — elastic, highly-available, on-demand computing and storage resources available over the net — but also crowd computing — elastic, on-demand human resources, who might be acting for pay (Mechanical Turk), for fun (Facebook), or for the good of humanity (Wikipedia). We’re still in the early days of understanding how to design and take advantage of these kinds of systems.”

Miller points out that on an individual basis, each student also has intellectual ownership of a project — a thesis that is wholly his/hers to design and carry out. So the group’s building a core project (such as a toolkit that provides a common infrastructure to which many people contribute), balances well as each student develops a system or experiment of her/his own that uses that infrastructure. Miller said about this paradigm, “ It’s an effective model, which our group has repeated several times, and we’re hardly the only group in computer systems research that follows it.”

  1. Prof. Rob Miller gives the Campus Preview prefrosh and family members an animated view of the User Interface Design Group that he runs in CSAIL. April, 2011 in the Grier Room.
    Prof. Rob Miller gives the Campus Preview prefrosh and family members an animated view of the User Interface Design Group that he runs in CSAIL. April, 2011 in the Grier Room.
  2. Prof. Rob Miller describes the beauty and wisdom of crowd computing to Campus Preview families in April, 2011 in the Grier Rooms.
    Prof. Rob Miller describes the beauty and wisdom of crowd computing to Campus Preview families in April, 2011 in the Grier Rooms.

He also notes with enthusiasm: “Our whole group is bursting with ideas. The four projects highlighted in this article are PhD thesis projects, so they were conceived by the students, not by me. But the development and focusing of the ideas is always a collaborative, give-and-take process.”

Considering the pace of research in Human Computer Interaction, it is impossible not to ask Rob Miller about his sense of the direction in which this field will move.

“Looking out a few more years, I see at least three interesting research directions for HCI. One is crowd computing — learning how to design and understand systems involving a large group of people interacting over the network. Another is multimodal interaction — interfaces that use more than just traditional display output and keyboard or pointer input, but accommodate a far wider range of natural communication behaviors, like speech and gesture and body language and emotional affect. A third is accessibility — interfaces that can be used fluidly and effectively by a variety of users in a variety of contexts, without regard to physical or situational disabilities. It seems probable that these three directions will actually converge to produce computer interfaces that draw on both crowd intelligence and artificial intelligence to make life better for people.”

Read about the UID projects and the students who develop them in the articles that follow: Soylent | Collabode | Sikuli | TurKit

Leave a Reply

Replies which add to the content of this article in the EECS Newsletter are welcome. The Department reserves the right to moderate all comments. If you would like to provide any updated information please send an email to newsletter@eecs.mit.edu.