This guide will help you understand how we think about research, how we go about doing it, and how we make each other smarter. These are the expectations that I hold myself to, and that I hold my group members to.
I expect that reading this will take ten to fifteen minutes.
Together, our group aims to make hard problems fun, and to make ourselves better than we ever could be as a collection of individual contributors.
As we start to work together, I will generally focus on mentoring three high level concepts in your research: turning down your idea filter, vectoring & velocity, and articulating the "why"."Nobody tells this to people who are beginners, I wish someone told me. All of us who do creative work, we get into it because we have good taste. But there is this gap. For the first couple years you make stuff, it's just not that good. It's trying to be good, it has potential, but it's not. But your taste, the thing that got you into the game, is still killer. And your taste is why your work disappoints you. A lot of people never get past this phase, they quit. Most people I know who do interesting, creative work went through years of this. We know our work doesn't have this special thing that we want it to have. We all go through this. And if you are just starting out or you are still in this phase, you gotta know it's normal and the most important thing you can do is do a lot of work. Put yourself on a deadline so that every week you will finish [a piece of creative work]. It is only by going through a volume of work that you will close that gap, and your work will be as good as your ambitions. And I took longer to figure out how to do this than anyone I've ever met. It's gonna take awhile. It's normal to take awhile. You've just gotta fight your way through."
Deferring criticism of your ideas is incredibly hard for researchers to do. I have seen graduate students kill their own great insights too many times because they see a flaw in it once they start working on it. Instead, view it as part of a process. We will solve it.
My main strategy for working through the self criticism is to go for volume. In their book Art and Fear, Bayles and Orland (2001) described illustratively what happens when you put your entire identity into finding that One Perfect Idea:
"[A] ceramics teacher announced on opening day that he was dividing the class into two groups. All those on the left side of the studio, he said, would be graded solely on the quantity of work they produced, all those on the right solely on its quality. His procedure was simple: on the final day of class he would bring in his bathroom scales and weigh the work of the "quantity" group: fifty pound of pots rated an "A", forty pounds a "B", and so on. Those being graded on "quality", however, needed to produce only one pot --- albeit a perfect one --- to get an "A". Well, came grading time and a curious fact emerged: the works of highest quality were all produced by the group being graded for quantity. It seems that while the "quantity" group was busily churning out piles of work --- and learning from their mistakes --- the "quality" group had sat theorizing about perfection, and in the end had little more to show for their efforts than grandiose theories and a pile of dead clay."
HCI researchers have even replicated the quality/quantity phenomenon with design. So, when we're brainstorming possible solutions to a problem, I will push you to go broad, generating twenty or thirty ideas before we pick one. If you trust Nobel Prize winners more than you trust me, try Linus Pauling: "If you want to have good ideas, you must have many ideas." Imagine your own creativity as being normally distributed. If you want to have a killer idea — two or three sigma above the mean — you need to generate enough ideas that the Central Limit Theorem kicks in.
I will push us to iterate rapidly. Rather than scoping out a huge project and then digging at it for weeks or months, I will push us to find the core and prototype that core as rapidly as possible. This means working in a loop:
Know that research is hard and full of challenge, and that I will work with you to push through it. In return, I expect that you apply your energy to it. I expect that whenever you get stuck, you come in with not just a description of why you're stuck, but also a set of solutions you tried, and a proposal of what to try next. I am otherwise too willing to brainstorm solutions, and over time, if your reaction is to just wait to talk to me, this doesn't grow you as a thinker.
I will push you hard on the "why" of the project: why is this problem important, why is this a major idea, why is this technical insight novel? This is one of my main skills as an advisor. I am totally happy to start out from a hunch and iterate from there, but I do expect that as we develop the project, we be able to answer these questions robustly. If we can't, I will call it out and we will reconsider.
I will weigh in on which projects or directions I think will be impactful. I expect that we both can explain convincingly that yours is an important project to work on and that the approach is appropriate. My point of view is that if two smart people can come to agreement that a project is amazing, it's very likely that others will see it that way. If I can't do that with where the project is heading, I will make suggestions or ask questions that I believe will push the project in a direction where I see an opportunity. I expect that you consider those suggestions seriously, and if you decide that they're not the right direction to go, I encourage you to tell me so, and why. I am eager to be convinced and see new perspectives.
Let's talk about what I advise, how I advise, and how I give feedback.
I advise projects that are in the space of social computing systems. This means that I'm interested in the design of interactive systems that connect people, or draw insight from the interactions between people. There are lots of different viewpoints on this space: we've worked on projects in social media; interactive systems powered by large-scale behavioral or community data; teams, organizations, crowds, and society; the intersection of AI and online communities; I'm always interested to push into new directions. Most of our work is in the design+engineering of new systems, but some is social scientific or more theoretical in nature. I'm happy to have conversations about the boundary points of my scope, and whether your interests overlap it. If you want to do work outside of that scope, I am not the right advisor: there are lots of amazing people at Stanford to work with. Typically, students and I kick around ideas at the intersection of our interests to find a point of resonance.
Generally, for our first project together, I will encourage a project closer to the core of the social computing systems that we have been working on recently so that we can develop a strong collaboration and mentorship rhythm; later projects may push further on the boundaries. I am more likely to pitch concrete projects to you early on to help as a scaffold, but as you grow, I expect that this relationship inverts.
In picking projects, we work toward societal good. Understand that technologies can be misappropriated, so we work on problems where the good is clear and substantial, and where the expected negatives are small in comparison and can be reasonably expected to be controlled by design, norms, and regulation. If you're unsure, bring it up and let's discuss. If I'm not sure, I will discuss it with you too. These discussions are important.
There are many possible metaphors for what we do as advisors: you've probably worked with advisors who view themselves as sage, or manager, or intellectual sparring partner, or something else. I view myself as an amplifier — that my goal is to understand your hunches, intuitions, and ideas, and to work with you to amplify them to produce the most amazing outcome possible. In order to be an amplifier, I expect that you come to me with opinions: you guide the project that we are working on. There exist many great projects in the world; I don't a priori have an agenda of which one within my scope we pick, just that we find one that has excellent chances of impact. I am happy to brainstorm projects with you and will eagerly gesticulate about problems I think are exciting. But I need to get your honest take on where you view the problems at. I need to deeply understand The Why of your project to be an effective amplifier. I will push you on The Why if I remain unconvinced.
I am not an armchair advisor. This perspective places me onto the hands-on end of the spectrum. I am a good fit if you are OK with this kind of relationship, as opposed to a more distant sage advisor who only checks in occasionally.
Academia can be a place with very little feedback, so I hold quarterly meetings with my PhD students and postdocs to reflect on the past quarter's progress and identify goals for the next quarter. This is not an evaluation meeting! It's a reflection meeting: a time to look back on areas we've been working on, and what we should work on next. If you are an undergraduate or MS student, I suggest setting these up with your graduate student mentor as well. When we hold our quarterly meetings, I will work with you to identify one area of focus and growth that you want to work on. During that next quarter, I will focus my feedback in that area, and may push you much harder on it than I would normally.
In contrast, if we both decide that something isn't an area of focus or growth, I will be more likely to step in and support it as a collaborator. For example, in the past, students and I have decided that framing papers and ideas are an area of focus, so I push the student hard to iterate on writing and iterating the intro of the paper with few direct edits from me, but I'll jump in and write other sections of the paper. If they're not one of your current focus areas, I might jump in and edit your paper's text, contribute code or statistical analyses, create graphics, or other contributions. I will try to signal in advance if I am going to do this, and if I don't, please call it out. If I signal it and you want to own it, let me know. This strategy is how I ensure that you are constantly stretching and growing as a researcher without needing to boil the ocean.
In general, I hold myself to a very high standard, and so I expect that my students try hard as well. Much like with the group culture, I believe that putting in more than the baseline academia level is what's needed to have an effective advising and group environment.
I work hard to create an engaged and positive group culture. Engaged means that we work hard to make each other better researchers. Positive means that we frame critiques, questions and suggestions so that the recipient leaves each meeting more excited about their project than when they entered.
We commit ourselves substantially to the group, much more than is common in this department or in CS more broadly. I pay in a lot, so I expect that you do too. I do this myself, and expect this of you, because I believe that the individual and collective benefits of this effort are substantial.
Most concretely, this means being present at group meetings and on Slack, both in terms of literal presence and in terms of active engagement. It means tossing ideas into the meeting or Slack channel, giving people feedback on their ideas to try and help make their ideas better, and finding opportunities to deepen both our insights and our social connections.
As a starting point, you should assume that the CSCW Code of Conduct is the baseline set of behaviors and level of respect that we expect from each other. And honestly, most of the time, the group is a very happy and collaborative place to be, so a code of conduct serves mainly as a background set of shared norms and expectations.
But, it's important to have explicit plans for what happens if something does go wrong. It might be, for example, an interpersonal collaboration issue: e.g., disagreements, misaligned expectations, or inadequate levels of dedication. If the issue is in these categories, I ask that you make a sincere effort to manage it first through discussion with each other. If that fails, please tell me: I am here for you and will help resolve it. In the worst case, if it can't be resolved, I still expect everyone to act professionally toward each other. Treat them as someone you are trying to help succeed in their careers.
More extreme are issues that affect marginalized groups: this group is not a space that tolerates harassment, talking down, sexism, racism, or other behaviors that marginalize. If these categories of behavior are directed toward you, or if you see this happen, tell me. Microaggressions integrate over time to become as toxic in aggregate as more extreme behaviors, so let me know about these if you observe a pattern as well. Unlike the cases in the previous paragraph, I don't expect you to try and work these out interpersonally before coming to me—just come and let me know. The hardest thing as an advisor is finding out afterwards that I didn't even know when something bad was happening. Tell me. I will work hard to do something about it.
If there is a complaint, I will generally start by having a private conversation with each of the people involved. If the event is not a massive breach of our code of conduct, and it's the first instance, I will generally aim to improve the situation. If the event is a more substantial breach or is a pattern that remains after prior feedback, I will develop consequences dictated by the nature of the violation that are within my jurisdiction as a faculty member.
I take your time very seriously. I ask more structured time commitments of you than many faculty. However, I work hard to organize the commitments so that your time investment brings big benefits to you and to the group. If you feel that an element of the schedule is underperforming, tell me! I am always experimenting.
Know that research is hard and full of challenge, and that I will work with you to succeed. In return, I expect that you apply your excitement, energy, and creativity to it. This does not mean working every waking hour. I can tell you from personal experience, and those of friends and colleagues, that pushing yourself too hard only results in burnout and uninspiring research. Instead, remember that creative work such as research requires that you have the mental capacity for creativity. So, find the rhythm that brings you the most energized to the research that you're doing. Do you best to navigate the tradeoffs, and I'm happy to have conversations about it if you're unsure. While it's rare, if I have concerns about your level of commitment to the project or about how you are prioritizing, I will air them directly with you in a weekly or quarterly meeting, so you don't need to wonder whether I'm concerned.
If you are a visiting or rotating PhD student, masters student, or undergraduate, make sure you do all of these. We will approve your membership on these lists once you complete this page:
students-msb@lists.stanford.edu list by filling out this form. This is all of Michael's group members.hci-gates3@lists.stanford.edu list by filling out this second form. This list is for all the CS HCI researchers, not just Michael's group.#students-msb and #rollingtopics channels. I also recommend #lolchi for memes and #hci-community for community amongst PhDs and postdocs. We often use Slack for quick chats and questions.Rotation students and visiting students: there are a few other perks I'd like to get you set up with. First, the HCI Group has a Dropbox org — if you're going to be aligning with the HCI Group somewhere, ping a faculty member about joining it so you get more space. Second, we have a GitHub org. Ping another PhD student in the group to get added as an admin to our GitHub org. You can use it for your research repos. Third, the faculty here are generally happy to pay for items that make your work environment better. Ergonomics (e.g., chairs, keyboard trays, keyboards) and hardware (e.g., monitors, mice) are fair game. If you need a laptop or machine to work on, also ask for that. Just prep a request list and let us know.
To succeed in research, you need to be able to dedicate significant time to it. Each unit at Stanford is roughly three hours of work per week outside of class/meetings. Making this concrete: for my group, you need to be willing to dedicate at least three or four units of time — about nine to twelve hours a week — to it in a given quarter. I am very happy to support research for credit if you'd like this reflected in your transcript! It's not uncommon for students to wind up as coauthors on papers, or eventually do an honors thesis or even a PhD eventually. Happy to chat about this!
Don't view yourself as someone who waits for instructions. The most successful CURIS students actively question, contribute, and add their own ideas to the project. You are the most promising student in your cohort; that's why we chose you. Add your voice and energy to the project!