Skip to content

Latest commit

 

History

History
51 lines (34 loc) · 5.29 KB

2021-11-05-what-i-learned.md

File metadata and controls

51 lines (34 loc) · 5.29 KB
layout title subtitle author tags publish_date summary
post
What I Learned from Doing Research
Take-aways not specific to HCI
Ziwei Gu
Reflection
11/05/2021
Some of my thoughts on research, project management, communication, and others.

I loved every bit of my 2+ years of research experience at Cornell. I was also grateful for the opportunity to participate in top conferences including WWW'21, CHI'20, and CHI'21. Overall, the experiences are life-changing, and I'd like to summarize some of my learnings here (constantly updating) as a reference to others and to myself.

Research Project Management

  • When a project involves significant system building, it's a great idea to spend enough time designing and brainstorming before implementing. It's tempting to start coding early and deliver a prototype quickly so that you can cheerfully declare "we are almost done!" However, it's highly risky as it often leaves you in a dangerous situation where things don't work as expected and you are stuck with hours and hours of wasted work. Plan well and only start implementing when the team has a clear vision of the system.

  • "Start simple, fail fast, fail early."

  • HCI is centered on humans. Put users' need first! It's easier said than done because we developers tend to get lost in what we can do, what's new, and what's cool. For example, when displaying data, we must resist the temptation to show everything we have computed so that users don't have to suffer from information overload. One method I like to use is to stand in users' shoes and ask what they’re gonna do with all those information.

  • Pick impactful research ideas worthy of your pursuit (e.g those that have a potential to fundamentally change some previous beliefs, to influence a large number of people, to inspire a new class of systems, etc.). Usually, "it's as easy to do something big as it is to do something small."

  • A procedure I found effective to start and complete a research project (after having a rough research idea in mind):

    1. Have a mental model of what a successful project looks like. (Yes, this is the first step!)
    2. Do a lot of literature review and write a research proposal.
    3. Have your advisor read the proposal and have an open and objective discussion with them.
    4. (If both of you are passionate) Start carving out the general framework in Overleaf. (Writing is a great way to get you started and help organize thought.)
    5. Make plans on tasks. To-do list, timelines, milestones. Prioritize!
    6. Design and Implementat. Code, test, iterate.
    7. Fill out the empty sections in Overleaf from Step 4 and finish writing.
    8. After submission, reflect on the whole process and take notes of any thought or feedback.

Communication

  • Be bold. If there's someone you admire, email them to ask for advice or a meeting. You never know who will be willing to meet with you. If there's an opportunity you are excited about, take immediate actions to get closer to it. You never know what you can take home. I used to be shy about reaching out, but by forcing myself out of my comfort zone, I have met and talked to board members of Cornell, CXO of Lyft, chairs of CHI, learned important life lessions and formed long-term connections.

  • A tip on reaching out to people: people are always more interested in their own problems than your problems. So if you want to reach someone, start with what they are dealing with. A practical example: email a professor by starting with their research, not yours.

  • I found it crucial to be candid when discussing research ideas and each other's work. I used to be leery of saying anything negative about others' or our collaborative work. After witnessing the power of a group of smart people talking openly with one another in a CHI session, however, I realized they could overcome any obstacles this way. Now I talk about failures openly and objectively. (But be sure to be friendly and inclusive!)

  • When presenting your research, sell the problem you are solving, and accent the difference your are making! Be concise! Skip any part that is not adding value to your message. Be very careful when adding long sentences to your slides: the only texts on slides should be the "take-home" messages you want your audience to remember.

Motivation and Productivity

  • I felt somewhat complacent for a while after landing an internship offer from Lyft, and I deeply regretted it. It turned out that nothing is forever. Our goals and expectations are constantly changing, so we need to constantly seek ways to improve or even reinvent ourselves.

  • Worthy pursuits require 100% effort. If there's a contest (or anything competitive), and you are not 100% certain whether to participate, then you'd better not participate. While you are not going to all in it, your competitors are, many of whom are set to devote everything they have to it. Thus, you are already out of the game unless you can beat them from a higher dimension.

  • Work is hard, but "inspired actions" are not. My friends often ask me how I manage to work long period without being burned out. To be honest, I don't think I "work", instead I try to take "inspired actions" every day. Truly inspired actions are effortless.

  • Be grateful every day and let others know your gratitude.