Philip Guo (Phil Guo, Philip J. Guo, Philip Jia Guo, pgbovine)

Inception: the moment when a new research project is born

Summary
What does it feel like at the moment when a new research project is born? I try to describe one such example here.

People sometimes ask me how I come up with new research project ideas that eventually lead to publications. And what does it feel like when such an idea is born?

The details differ for each project and are especially complex if I'm co-developing an idea together with my students or collaborators.

The simplest case to explain is when I come up with an idea on my own, which happened with several of my Ph.D. projects (read The Grind for details). Now as an assistant professor, most of my projects are co-developed with others, but occasionally I still get to do solo work.

For many such projects, there's often an inception moment when the germ of an idea crystallizes into my head almost wholly-formed, as if it came out of nowhere in a Eureka-like spark of inspiration. But the full story is always a lot more nuanced – usually involving months or years of incubation time when I had no clue that was the idea that would eventually form.

This morning, I had one of those inception moments when a new research project was born. I want to briefly walk you through what led up to this moment of clarity and what it feels like ...

  • Almost 8 years ago when I was in grad school, I started the first version of Python Tutor. Back then I had no idea that it would eventually help me get two faculty jobs (in 2013 and 2016) and turn into a scalable platform for conducting research. I just wanted to make something cool and share it with the world. But without building and growing that site, I would've never been in the position that I'm in today to come up with my current idea.
  • Over 2.5 years ago back when I was a first-year assistant professor at the University of Rochester, I co-developed an idea with a first-year Ph.D. student that eventually turned into a successful Google faculty research grant proposal. That idea was very distantly related to the one I just came up with, but it was in the context of learning from YouTube how-to videos rather than learning to code on the Python Tutor site. Still, the seeds of my current idea were there in that old proposal. By the time the funding arrived, that student had already decided to drop out of grad school, and I was also on my way out of Rochester. We never got to implement that idea, so I used the funding for another student to work on a related idea.
  • Also around that time (2.5 years ago), loosely inspired by the high-level ideas laid out in that Google proposal, I decided to augment my Python Tutor site to collect a specific form of data from users' coding activities that may or may not have gone anywhere. Back then I didn't have a concrete plan, but I had a vague hunch that adding this kind of data collection may open up interesting possibilities down the line. And even if it was an intellectual dead-end, no harm done. It was only a few days of coding work.
  • Around 6 months ago, I looked back at the collected data and saw that indeed it looked promising, so I decided to augment the data collection a bit more along those same lines. I still didn't have a concrete project idea yet, but I had an even stronger hunch this time around that it may lead somewhere promising. And again, even if it didn't, I wouldn't lose anything except for a few days of coding time.
  • Now with 2.5 years of data from the original Python Tutor intervention and 6 months from the augmented intervention, the stage was potentially set for something ... I still didn't know what yet. Something was bound to spark sooner or later, but I had no idea when or where that moment would come.
  • I was traveling for the past few days so my mind was clear from the usual day-to-day work responsibilities. Being in a different mental context let my mind clear a bit from short-term goals and subconsciously churn on longer-term speculations ... this was 100% subconscious; I wasn't trying to come up with anything new.
  • I got a ton of sleep last night to recover from the fatigue of traveling, which further helped clear my mind.
  • Today was a Sunday, so I didn't have any immediate pressures of work work work like I usually do on weekdays. I had no specific plans for the day. (This part is critical, I think!)
  • When I woke up this morning, I had an inception moment where the current idea came to me. I totally didn't expect it; it just came out of nowhere. I suddenly saw everything so clearly tied together – the Google research proposal from 2.5 years ago, how that didn't end up working out with YouTube videos but the essence of that idea could transfer to Python Tutor, how all that data I had collected from the new Python Tutor interventions in the past few years may lead into a brand-new idea, and most importantly, that there was a set of upcoming paper submission deadlines to venues that would be a good fit for this work. That way, even if it doesn't get accepted on the first attempt, there would be ample opportunities for adjustments and resubmissions down the line.
  • The feeling I currently have is one of intense yet simple clarity that's hard to put into words. It's as though everything from the past few years that was dancing around this idea but was all jumbled and fragmented in my mind all of a sudden clicked into place to form into a coherent thesis.

I don't think it's a coincidence that this spark came after I took a step back away from my day-to-day work (this time when relaxing after traveling for work, but at other times in the past, during vacations or other moments away from the daily grind).

OK, so will this new idea really turn into a publishable project? I can't say for sure since I can't see the future. Maybe it won't pick up steam, maybe it'll totally flop, or maybe I'll get too busy with other projects to make headway on this one. Who knows?!? But the one thing I do know for sure is that many of my successful research projects in the past started with one of these sudden inception moments. Like I described above, though, they never come out of nowhere. These moments of inspiration come only after months or years of incubation, dead-ends, and weird turns of serendipity.

So the only actionable advice I have if you want to come up with such ideas is to continually stumble in the right direction.

For more thoughts along these lines, I highly recommend Steven Johnson's book, Where Good Ideas Come From: The Natural History of Innovation.

Created: 2017-11-19
Last modified: 2017-11-19
Related pages tagged as research:
Related pages tagged as assistant professor life: