2014 Annual Review

February 22, 2015   

Recently I subscribed to James Clear’s blog and his writing and advice has resonated with me. The last time I felt like someone was speaking directly through me was Cal Newport’s Study Hacks blog, which I discovered during undergrad, and I felt it directly addressed the issues I was having with keeping up with my intense courseload.

James Clear posted his 2014 Annual Review in December, and it reminded me of how we do annual self-evaluations for work. I read through my evaluation for 2013 and enjoyed reflecting back on that time, how much I did then, what didn’t go well, and what I’ve learned since. So I’m going to follow his template with the following three questions.

  1. What went well this year?
  2. What didn’t go so well this year?
  3. What am I working toward?

1. What went well this year?

Fellowships. This year, I was the first and sole recipient of the NumFOCUS John Hunter Technology Fellowship. I was thrilled to receive it, especially because I thought I had no chance of getting it. From the wording of the announcement, which said “postdoctoral scientists or senior graduate students,” I thought I wasn’t seasoned enough, and I figured people who work on scikit-learn and IPython would apply and definitely get it over me. Plus I couldn’t think of anyone who would be willing to be my Software Mentor, as required for the application. Thankfully, at PyCon, I was talking to Titus Brown and mentioned that I wanted to apply to this fellowship but thought I wouldn’t get it, and that didn’t have a Software Mentor. Immediately, he dismissed my doubts and offered be my Software Mentor, on the spot! I almost cried because I was so happy and surprised to hear someone believe in me so strongly, especially because for the previous year I had been getting major shade and negativity from certain peers back home, which majorly impacted my mental state and made me feel constantly inadequate.

Python involvement. I’m very proud to have contributed to the seaborn statistical data visualization Python package, specifically creating the first-ever (that I know of) single-line clustered heatmap function in Python! It was months of work, and couldn’t have been done without the input from Mike Lovci and Christopher DeBoever. Plus now I have more clout to evangelize Python for bioinformatics research :)

Passed Qualifying Exam! On May 15th, 2014, I proved my competency to a group of three professors, who all agreed that I can move on to the next stage of graduate school! I spent a solid month writing and working on my presentation and was very nervous and stressed about it, so it was very satisfying to be over.

Travel and conferences Last year I traveled to:

  • Strata 2014 in Santa Clara, California
  • PyCon 2014 in Montreal, Canada
  • Post-quals vacation in Cabo San Lucas, Mexico
  • Single-cell genomics conference in Stockholm, Sweden. Plus Italy (Rome and Venice) afterwards
  • PyData 2014 (gave a talk!)

It’s been a great experience to visit these places and make new friends in the Python, data science, and single-cell communities. Looking forward to more this year!

Photography While I haven’t publicly shared much photography, I’ve taken lots of photos on these different trips. The Europe trip was especially good, because I purposely tested out different settings in low light and other tough situations, and I spent a LOT of time watching LightRoom tutorials on editing, and learned a lot of new techniques that I’ve been practicing with new photos.

Personal relationships. I had some tough interpersonal issues this year and in talking through it with a friend, they recommended the book Difficult Conversations. I highly recommend this book to everyone, because difficult conversations happen every day, and this will provide you the tools and perspectives you need to understand them. I can honestly say that the book has changed my life and helped me critically evaluate everyone’s role in a tricky spot, and come up with strategies in how to approach these situations tactfully. Tact has never been my strong suit, so these lessons were eye-opening.

2. What didn’t go so well this year?

Not being afraid of the unknown or difficult. These first few years of graduate school, if some task took steps that I couldn’t clearly visualize, or I could see them but I thought they were too hard, I wouldn’t do it, and my research would suffer as a result. For example, I’ve been putting off doing a complicated operation on a bunch of sequences that would help my research in ways I can’t comprehend now, but it was so nebulous to me that I didn’t even want to attempt it. Or I’d get some result and not know how to interpret it, and the solution was to go to the literature and read a bunch of papers about those genes, but that’s too hard so I didn’t want to do it.

Blogging. I published 10 total blog posts in 2014. I switched to hosting my blog using github.io rather than Tumblr in 2014. I thought my lack of blogging was a result of my blog being on Tumblr and being ugly and annoying to write IPython notebook style blog posts for it, rather than using markdown/IPython entries. I did have 6 blog posts after September, compared to 4 pre-September, so the change helped. But I still feel I could have blogged more, even small random things I found helpful like that bit about using different git branches in different IPython notebooks, which was a great discovery for me in my development workflow. I think part of my lack of posts was also the misguided belief that every post has to be a gigantic, perfectly crafted tome, when it’s fine to have a concise 100-word post.

Publishing. I didn’t complete any scientific publications this year. While it’s not common for a 2nd-year graduate student to publish a paper, I thought I was working hard towards that goal, but in reality, I didn’t understand the magnitude of work that goes into a publication. I also think my being afraid of the unknown, not knowing how to do something and being afraid of being wrong, also played into this.

Regular progress reviews. Every once in a while I put together weekly progress reviews of what I did that week and what’s the plan for the next week. It makes me feel good, and it works for a bit, but then I give it up because I think “ehh, I’ll just remember what I need to do.” I feel I spend too much time on trivial things like reformatting code or making some plot look just slightly better (but not necessarily more readable), rather than moving on to the next thing.

Python involvement. You may wonder why this topic is in both the “went well” and “didn’t go so well” categories. While I’m happy with my seaborn contributions, I haven’t been as happy with other things. There’s been features I’ve wanted to add to pandas (read gzipped files over the internet) and other visualization features to seaborn, I haven’t gotten around to doing them. I did start subscribing to matplotlib-devel and watching the seaborn repo for all issues, and replied to a few, but I’m still afraid that I don’t have anything to add to the conversation.

Sustaining interest. As both James Clear and Cal Newport have discussed, it’s consistent, sustained effort and focus over a long period of time which generates results. I find it really easy to start something fresh and new, and I struggle with maintaining interest in something long enough to reap the rewards at the end.

Cello Last January I performed a small cello concert for my friends, and worked really hard practicing a lot for it. But this last six months I haven’t practiced much. Part of it is my teacher had personal commitments that prevented him from teaching for some time, but I think the more major part is that I didn’t pay for my lessons. I was supposed to help him with his Cellosophy cello app, but honestly I just bit off more than I could chew. The coding was going to be for the instructor website in D3.js and that’s not part of my core competency, so it was a lot of work for me to just make a bar chart. And so I didn’t do a good job of collaborating… and as a result I didn’t ascribe as much value as I should have to my lessons, and didn’t invest as much time in them. Once I get back into cello, I’m going to pay for them. As expensive as they are, they are worth it.

Fitness I was very good about going to yoga and exercising 4+ times a week, but something happened around August and I stopped working out as much. I gained some weight and lost fitness, and I’m now trying to change that (see next section).

3. What am I working toward?

Regular progress reviews + Time-boxing. I’ve started planning my workday carefully in a Moleskine daily planner (in green, my favorite color). This is part of what my colleague Colleen Stoyas likes to call #projectGETONYOGAME. Writing down how I spend my hours has been a very effective way for me to keep track of where the time goes, especially when it turns out that what I thought was a 30 minute break turned into an hour and a half of eating and chatting… yikes. I’m taking the Now Habit approach of not trying to change my habits dramatically, but writing everything down and paying attention to what I’m doing. Below are some example schedules I’ve created, for days that went as planned, and others that did not. The images have been blurred for privacy.

Here’s a sample day that went quite well, and I mostly stuck to my schedule. You’ll notice I have some random notes on the side of reminders of people to meet with, and I even drew out a network. However, I did end up taking a 1.5 hour break from lingering around and chatting, which I probably could have shortened and ended my day earlier.

Example hour plan for a day that went mostly as planned

Below is an example day that didn’t go as planned, where I started late and thus deviated from the schedule quite a bit. I also got the time wrong for when I was supposed to be teaching data cleaning (2pm not 12pm), so that changed things up too. While this didn’t go completely as planned, I was able to squeeze a lot of work in.

Example hour plan for a day that didn't go as planned

Publishing. It’s hard for me to call a project “done enough,” and yet this is exactly what I need to do to be able to publish. I have a couple conference abstract and presentation deadlines coming up in the next few months which will help me to focus on creating “publishable units.” Plus I have my advancement in a month, so that will be a big push to get manuscripts ready so I can say that certain projects are at least “submitted” to my committee.

Advancing to candidacy. In our program, the “Advancement Exam” or “Senate Exam” is when you show your committee what research you’ve accomplished so far, and your plan for your PhD. I’m very nervous for it, and have been doing a lot of background reading to be able to motivate why my projects are important, and why they make sense to be done together. I’m planning to do this in March, so wish me luck!

Fitness. I signed up for my first half-marathon, the La Jolla Half Marathon! I’ve been training hard for it, maybe a little too hard because I hurt my knee and am resting it now (no running for 1 week!).

Photography. I have a bunch of photos I’ve taken of nature, flowers, architecture, scenery, and LegoLand that I use as backgrounds, and I’m planning on sharing them as an album. As motivation, I’m also planning to apply to this year’s Ubuntu Wallpaper Contest.

Cello. I’m going to start scheduling time to play cello, even if its only fifteen minutes, it’s still an important investment.

comments powered by Disqus