can we meet? Sure! During the academic year, I hold weekly group office hours. (Exception: when I'm away for a talk/conference—check the talk dates on my home page.) I don't take appointments for office hours, just come in during that time. During office hours, please walk right in. Do not wait outside (I won't know you're there); join us inside. If your question is about enrolling in a class, see Thanh, not me. If you are a student, and we make decisions about your academic plan, follow up with an email to me summarizing the decisions we made. It's essential to have this in writing—my memory is pretty good, but not perfect. To enable me to focus on research during summer, I don't hold summer office hours.
I don't want to take up your office hours, or have a conflict. Can we meet another time?Classes, degree requirements, recommendations, jobs, research, etc. are all excellent topics for office hours. Over the years, I've experimented with many formats. I've found group office hours offer students the best experience. Often, students have related questions. (Even when ahead of time, they're sure they're the only one.) Other times, it may not be a question someone has yet, but it's valuable for the future. Furthermore, there are many questions for which students can chime in with valuable knowledge I don't have. (Like class recommendations and job advice.) In short, come to office hours: I hold them for you, and your question is welcome. If it's important that our discussion be confidential (i.e., not that you believe it would be boring or irrelevant to others, but the topic is private), come to office hours and let me know you have a private matter to discuss. We can speak privately at the end. If it's important for you to meet specifically with me, I trust you can find a way to make my office hours one week. Alternatively, if other faculty can answer your question, you can speak with another faculty member whose office hours may fit your schedule better.
UC San Diego students
Can I do research with the Design Lab? Will you be my research advisor?One of the best parts of faculty life is working with talented students. The design faculty advise a small number of highly-motivated graduate and undergraduate students who are working on topics related to our group's research. Please apply here. We don't have the bandwidth to advise projects unconnected to our group's focus. (It may be a great project and deserving of support, but successful research requires a tremendous amount of focused effort.) Undergraduates: apply to CalIT2's fantastic Summer Undergraduate Research Scholars program. For research during the school year, email me with your resume and transcript. A prerequisite for joining my group is having taken and excelled in a design course. For undergraduates (and MS students new to HCI), take cogs120/cse170 (offered in winter). For PhD students (and MS students who have already taken an HCI course) take Cogs230/CSE216. I also strongly recommend attending the Design at Large seminar: To earn credit, sign up for Cogs229/CSE219.
I want to learn interaction design; what courses should I take?In our curriculum, you'll learn to think like a designer and gain the skills to design, implement, and evaluate interactive systems. Interaction design students need a "tripod" of Computational Thinking, Design Thinking, and Empirical Thinking. The main HCI courses are the Cogs120/cse170, Cogs102 sequence, and the Cogs187 sequence. To bone up on your technical skills, seek out relevant CSE courses. Take advantage of UCSD's excellent Comm, Visual Arts, and CAT programs. We design interfaces for people; learn something about people and how to do empirical work.
I don't have the prerequisites for your class? May I still take it?In general, no. Two cases where this is possible: 1)If you have taken an equivalent course elsewhere; 2) If you have equivalent extracurricular experience (e.g., in a job) AND agree to take the prereq course at the same time. If either of these apply, Thanh will approve your request (do not email me). If neither of these apply, then you'll need to take the prerequisite class first (do not email me).
How can I find out about design internship opportunities?Students can get announcements of design internships by joining the designjobs mailing list.
advice on picking classes?
Shop Show up to the first day of more classes than you will eventually take. Attending the first day gives you more data about the class (topic, teaching quality, etc.). More data, better decisions.
Challenge yourself Take classes that will deepen and broaden your knowledge. Don't shy away from something because it's "hard". Try it. Build up strong fundamentals. Dive deep into areas you're passionate about. And go out on a limb to try something brand new.
Don't sink yourself Pay attention to how things are going, and adjust realistically and interactively. Adjust your courseload so that you can really learn the material in each course. There's no point to taking lots of classes if you're learning little and performing poorly. If you get in over your head, do something about it immediately. And use that knowledge to plan more judiciously the next quarter. (As opposed to saying, "I bombed last quarter. So this quarter I need to pile on even more.")
Startup XOR classes There's not enough hours in the day to meaningfully do a startup and be a student. Both are valuable pursuits. But not simultaneously. If you want to do a startup, take a leave. And if you want to be a student, save the startup until after graduation.
I have served as the academic advisor for more than 200 undergraduates. I've seen my share of letters from the administration that a student has been placed on probation because of poor academic achievement. Every single time, this probation was caused by students making unrealistic overcommitments, and then not being able to deliver. Don't make this mistake.
recommendation lettersWriting recommendation letters is one of the important things I do, and seeing students succeed in the next chapter of their lives is deeply rewarding for me. It also take a lot of time. I'm currently getting multiple requests for letters per week, every week. Unfortunately, a decent fraction of this time isn't the intellectual effort of composing a thoughtful letter, but the administrative tasks of gathering all the resources I need to write it, keeping track of the students/locations/status table, and dealing with the idiosyncrasies of all the web-forms. (It may be surprising to some, but the web has made letter-writing more time-consuming, not less.) Please...
- Send all correspondence about recommendation letters to srk-admin@ucsd (not srk@ucsd), so that it goes to my admin also. Put "recommendation letter" in the subject of your email. This helps insure that nothing falls through the cracks. Also, use the srk-admin address on all Web sites. I will not respond to questions about letters except through that address. Why? Both your letter and my other responsibilities are very important. The admin address insures both are handled well.
- Ask for recommendation letters at least a month before they're due. While I can often get to them sooner, this lead time lets me work around travel, big deadlines, etc., so that I can block out a free, quiet time to write a thoughtful and substantive letter. Please don't send email asking, "Sorry for waiting until the last minute, but I have an imminent deadline -- could you write this right away?" Follow the instructions below, and I will do my best.
- Ask your TA(s) for courses and/or your PhD mentor (for research) to send me an email describing the work you did. When you do, summarize your work for them to help jog their memory. I will not write your letter until I receive a note from your TA/mentor.
- Send one email that contains all of the information required to write the letters. This email should:
- Summarize your work in my research group and/or class(es). Include a particular example where you demonstrated creativity, determination, or excellent performance, or recognition. My memory is pretty good, but external aids are always helpful. Please send me content free of grammatical errors (if you need the help, ask a friend or colleague to do an editing pass). Summarize your work beginning with the most important contributions, not chronologically.
- Attach your transcript, CV/resume, and (if applicable) a draft of your statement (doesn't need to be final).
- Include a link to a Google spreadsheet of all schools (if you applying anywhere other than the Stanford CS co-term). The first column should list schools. The second column deadlines. The third email addresses, if at all possible. If email is not possible, include a link to a Web form. That helps my admin and I insure no school falls through the cracks.
- Provide a consolidated request for my sending your letter. Please avoid sending a new email every single time something new pops up. Three options for achieving this.
- List the recipients in your request email and/or
- List the recipients in the spreadsheet listed above and/or
- Set up an Interfolio account, because you might very reasonably find new jobs, schools, or fellowships to apply for. You can then have Interfolio send it anywhere you need.
- To convey that my letters are candid, I don't release letters to you.
- If you are applying to any Stanford program, do not give me any paper. Really. Despite the fact that some claim to require paper (like the co-term), it works just fine to submit them electronically. If a non-Stanford program requires paper, please give it to my admin, Teenah Eco, on the 5th floor of Atkinson Hall.
- Please do not worry that I will drop the ball and neglect to submit your letter on time. I am acutely aware of how important these letters are. I have written hundreds of recommendation letters, over more than a decade. Every single time that I have received the request at least a month in advance, I have submitted the letter in time. Sometimes -- especially during December's heavy season of PhD and faculty applications -- I submit my letter a few days after the applicant deadline. This has never caused a problem. When I submit your letter, I will notify you. Please email me one reminder 2-3 days before your first deadline. Please do not send more than one reminder. Email overload is the primary impediment to my writing your letter; please don't make it worse.
- No thank-you gifts. It's very kind, I appreciate the sentiment, and I understand it's unlikely that chocolate would sway a letter. But to make clear to your letter recipients that my kind words are based on your deeds, no gifts. There is one very important way to thank me: work to be a mind-blowing success wherever you go! That way, the letter I write for the next student will be received enthusiastically.
- Lastly, and most importantly, after you've been out in the world a while, let me know how you're doing! I'd love to hear from you. Share the exciting things you're up to, and also share any ideas you have about new things we should be teaching.
- Who should you ask to write you a letter? There are two main criteria to balance. The first is to find people who can say substantive, strong things about your work. The second is to have letter-writers that have previously written letters to the program you are applying to, so that the recipients know how to calibrate them. For that reason, I suggest against having PhD students or post-docs write you a letter themselves. Instead, have them share their insights with a faculty member writing a letter.
- If you are reading this and you are involved with graduate admissions, you can help! First, do not have additional questions that are unique to your university. I work hard to write thoughtful, informative letters that are self-sufficient. University-specific questions do not provide significant incremental value, but do consume significant incremental time. It is not a scalable solution. Second, please encourage your institution to participate in a central clearinghouse for letters. That is a scalable solution.
If you find a problem with course materials, please flag it.
If you experience technical issues while taking a Coursera class, please visit these troubleshooting pages in Coursera Learner Help Center and get in touch with Coursera Learner Support team from there.
phd committeesLearning about research in diverse areas is something I really value about university life. Design, as a multidisciplinary field, provides many such opportunities. Consequently, I enthusiastically accept invitations to participate on student's PhD committees when the research has a design-relevant component — and this can be construed broadly. For both of our benefit, I need to decline invitations when my expertise won't have much to offer. For you, it's really important to assemble a committee that can provide you with meaningful feedback on your work. For mine, I have lots of research that I'm excited about doing — if I can't meaningfully contribute, I'd rather be doing research. If you invite me to join your committee, please include a sentence about what feedback you hope I'll offer on your work.
In general, I take my role on your committee to be a coach, not a judge. I presume that by the time you have scheduled your orals/defense, you and your advisor both believe that you are ready. As such, my feedback will be about how to make your dissertation stronger. Please help me provide valuable feedback:
- It's essential that your entire committee be present at the same time because you need to insure that your entire committee is on the same page and has a single, joint discussion. Also, spending time with other faculty is a big part of the draw to participate.
- To find a time that everyone can attend, make sure to schedule your orals at least two months in advance.
- If possible, schedule your presentation in the afternoon: I try hard to keep mornings free for focused research. Scheduling into an existing group meeting is ideal, as that'll already be on many faculty and students' calendars.
- Before your defense (or proposal), please email me a draft of your dissertation (or proposal) and directions of how to get to your orals. I may not know that the "conference room" in your department is room 203, where on the floor room 203 is, or even which concrete & glass building is yours :) Here are a few questions to think about as you put your presentation together:
- Is your title the broadest accurate description of your (intended) contribution? For example, how would your claim & contributions change if you removed qualifying words from your title?
- Can you give us a worked example?
- What are your measures? Have at least one slide titled ‘measures’. In general, using APA section titles – methods, measures, results, discussion – will help both you and us.
- What are the 2-3 most related pieces of prior work? How does your work relate to that?
- What feedback do you want from us? and where would you like to head after graduation? (For example, if you’re headed on the academic job market, you’ll may be best off doing different stuff to finish your thesis than if you’re planning to do a startup.)
- At your presentation, please
- Provide me with a paper printout of your slides. It's much easier to take notes by annotating slides.
- Remind everyone to turn off their mobile phones and WiFi so that we can focus on your presentation.
- Clearly explain what you did, what the results were, how your work relates to prior work, how your results compare to alternative approaches, and what you take the import of your work to be. Well-designed visual aids (such as graphs) can help; Edward Tufte's books provide a great resource for this. In particular, remember to label the axes of graphs. Well-chosen photos, diagrams, and pictures help convey both content and setting. Remember the point of visuals is information, not decoration. If it distracts rather than adds, delete it.
- Make sure someone takes notes of all the questions your committee asks, and your responses to them. Not you, as you'll be immersed in the Q&A. Your advisor is generally a good person to ask to take notes.
- Allow enough time in your schedule to receive and integrate comments from your reading committee. Your dissertation readers' time is best used serially. (When people read in parallel, they may waste time suggesting the same corrections.) Please don't ask everyone to read your dissertation a week before the filing deadline. Instead, if your schedule slips, extend your registration by a quarter. If you ask me to read your dissertation, I presume you seek (and plan to address) comments. If all you seek is a quick signature and don't want feedback, please ask someone else.
- After your defense, please send a revised version of your dissertation that integrates both in-person feedback from your defense, and written comments from your readers.
prospective studentsYes, my lab is taking new students!
what department should I apply to?Design is a popular focus area for several graduate programs. Graduate admissions are handled at a department level. I'm jointly appointed in CogSci and CSE. Students seeking to work with me should apply to one (or both) of those departments. How should you decide between the two? The main differences are degree requirements and your cohort. If you seek an engineering-oriented program and cohort, apply to CSE. If you seek a social science program and cohort, apply to CogSci. I also work closely with faculty and students in Communication and Visual Arts.
Can I discuss my application with you?No. (So that faculty can spend their time with current students, teaching, and doing research.) UCSD is a great place to be an undergraduate or graduate student. Put together a strong application. The admissions committee looks at a range of factors, including grades, test scores, and recommendations. One particularly important point is evidence of ability to do research -- if you have done research, your chances of admission are far better. Especially if you have worked on published research. Stress this in your application. If you'd like to get a feel for our program, we encourage you to:
If you've been admitted; congratulations! We hope you'll come, and the faculty are happy to talk with you and answer questions.
What should I put in my statement of purpose? Paint a rich picture of the work you have done and the work you seek to do. Avoid vague claims of interest like "I have been fascinated by computers since childhood" or "UCSD is a great university" in favor of concrete descriptions of work you've done and topics you find interesting.
Non-UCSD students interested in internships Generally speaking, we do not have the resources for non-UCSD interns. We will not respond to internship requests from students. If you have research experience in design, and are interested in a project related to our group's work, please have a faculty member at your university email me, and have them include an explanation of your particular qualifications for design research.
Can I do a Post-Doc with your group? When there's a great fit, a talented student, and funding, we sometimes have Post-Doc positions. If you write us about this, please explain how your background prepares you for this particular group, and what particular joint research opportunities you see. The volume of Post-Doc requests we receive prevents us from responding to generic inquiries.
colleagues & industrial partners
Can your students help me with my design project?Interest in design collaborations and students is high (currently I'm often getting multiple collaboration requests a day), and we love to help. Here are two readymade paths. The first is to hire a student; instructions on how to connect are below. Second, we're always looking for good seeds for class projects; send us your idea at this link, and we'll connect you with student project teams. Student projects can help colleagues and industrial partners explore speculative projects that they don't have time to, design outside the box, push the envelope, and translate UCSD research into practically-oriented design sketches. Sponsoring a student project is a great way to engage with the university community, find talented students to hire, and gain inspiration. It's not a venue for production code or private IP. Student work is earlier-stage than those things and obviously focused on educational rather than commercial goals. Currently, all class project work is completely open—students own their work and we don't have a mechanism for anything to be proprietary or secret. We are open to discussing more sponsor-directed models, as long as its in students' best interest. If you can help provide a great educational experience—and want to have your thinking expanded by student exploration—we'd love to talk. By contrast, if you see students as a way to avoid paying market rate for software development, please look elsewhere.
How can we begin a collaboration? We greatly value working with partners outside academia. If you would like to support our research, hire our students, or get involved in any other way: contact srk-admin@ucsd with your interest, join us online or in-person for the Design at Large seminar, and sign up for Design Lab email announcements. Once a quarter, we'll send out final project presentation announcements. This is the best way to meet students, talk with them about their work, and discuss common interests.
Can I take your class for professional development? Yes. Provided you have the required background, you can take design classes through UCSD Extension. You can also join the 200k+ students who have signed up for my free online HCI course.
We would like to hire your best student. Can you put us in touch? We get several requests per day seeking to hire design students and developers, and this outpouring of enthusiasm is awesome. Given this volume, please follow these instructions to insure the best students read your email.
The best way to send internship requests is to contact the Corporate Affiliates Program. They are extremely effective at getting the word out—much better than us. If you do send us email that you would like shared with students: Please send one email to srk-admin@ucsd that can be forwarded as is. No attachments; no preamble; just what should be forwarded. I encourage you to provide a concrete morsel that can serve as a draw—students rarely respond to vague requests. As mentioned above, join us for the seminar and sign up for quarterly announcements—the quarterly final project presentations are the best place to meet students and see their work firsthand.
Employers interested in hiring PhD interns Internships are a great way to expose students to real problems and learn how people really use interactive systems. Internships also provide students with valuable experience about what it's like to work outside the university. We encourage PhD students to seek internships when it will likely contribute to the student's dissertation research. Consequently, it's tremendously import that students be able to publish their internship work and use the resulting data and insights to further their research. When appropriate, code that interns write may be proprietary to the employer. However, employers should anticipate -- indeed relish -- the idea that students will continue working in the area of their research internship, and that after their internship students will continue to publish and produce code that is often open source. Interns provide their employer the opportunity to closely collaborate with someone immersed in the culture, insights, and innovations of their university. These internships often result in long-term collaborations with the student and with faculty.
Use our software, data, & experimental materials We work hard to share our software, experimental materials, and data whenever possible. For our data and experimental materials, please contact us. Here are links to use PeerStudio and Talkabout; contact us for source code. To use other software, please contact the first author of the research paper.