Back to Today

Life Update

September 19 - October 18, 2025

TL;DR

I typically write this newsletter as the month goes by, but this month I forgot, so it has been entirely written in the past two days. I therefore apologize for any worsened quality or increased recency bias.

Rose: I've gotten really invested in work projects
Thorn: it gets dark too soon after I leave work
Bud: USWNT is playing in Hartford

Work

As I believe I explained in a previous update, two of our studies are occurring as part of a special studies because the participants need to learn things before they can representatively participate in our tasks and this also makes people more willing to dedicate nontrivial amounts of time to the study. But the point is, we have a class of ten people and I had to teach a class about using git because I "need to get used to it" and "people are a lot meaner when you're not at a historically women's college" or something along those lines. I suppose this is a fair point, but I was nonetheless really sweaty the whole time.

Warning: some unexplained terminology about git forthcoming. Skip if you wish.

My life has revolved around running the CI/CD pipeline for all of the development that has occurred on our GitLab extension. For backstory, development on this project has been occurring since the summer of 2024, although I wasn't on it until recently. For a while, the lab consensus (before my time; I don't really know the reasoning behind it) was that we don't need to pass the pipeline, which is created by the official GitLab developers and essentially copies the entire repository into a container on a server (ours is called Gritty) and then runs a series of tests, jobs, and sub-pipelines according to which particular files were changed to verify code quality before you merge any code into another branch. Then, the lab consensus changed and it was decided that we should be passing the pipeline before merging anything. So, I was tasked with cherry-picking every commit that was made from the past fourteen or so months and modifying them according to the cryptic messages spit out by the pipeline until it passed. The inconvenient thing about the pipeline is that it ranges in length from approximately 12 to 36 hours, and the codebase only exists on the computers physically in Ford 343, so I have to go into lab as soon as possible after a pipeline fails or passes, which has caused me camp out at Fruit St. until ungodly hours of the night (okay, like 11 PM, which is ungodly compared to my usual bedtime) so that I can be in proximity to Ford in order to maximize the amount of pipelining that occurs overnight. But it's all worth it for the euphoria I feel when I finally see the row of green checkmarks that means I get to merge another chunk of commits.

Then, just this week I discovered than in the intervening time between summer of 2024 when we originally branched off of the GitLab master branch and when we pulled from the GitLab master branch in July as the standard against which to compare pipeline, the features off of which we built our new functionality have been completely re-routed such that all of our changes, while they may not throw errors in the pipeline, have no actual effect on the software because those files are now just dead code that never get called in the new backend of GitLab.

To add insult to injury, this past Wednesday night, the ITS people somehow remotely wiped all the data off of the computer we had been using for development. They meant to wipe another computer that was supposed to be wiped, but typed in one of the numbers wrong and so they irretrievably erased the OS and everything that had ever been installed or saved on the computer. While it is true that we probably should have been pushing to remote more often, it was still devastating news to lose all the work that had been saved on that computer and it seems quite absurd that the ITS people just did that on a random Wednesday night.

In other news, I've been conducting one on one interviews with people for our goal modeling methodology study, which is a new type of research for me. It's not my favorite type of research but we have gotten some unexpected and useful insights.

Yesterday, our lab hosted the CS lunch talk series speaker. I'd met him at two conferences and was a little worried because he always aggressively asks questions after conference talks. It was a bit of a struggle to get him to stop giving opinions and actually listen to what I was trying to explain to him, but then he did eventually give some useful tips. He's also the guy who writes ey/em neopronouns when referring to third person singular in his papers, which I think is absolutley iconic. His lunch talk was about the technical debt created by using AI at various stages of the software development process and how we can measure the net time actually saved or wasted because of the increased debugging costs caused by the perceived saved time by vibe coding.

Knick the Knee

Knick is doing good! I've been running regularly, though have been plagued by increasing shin pain that I refuse to give the satisfaction of calling a shin splint. I'm sure I have weak tibs and calves after being sedentary for so long so hopefully as I continue strengthening I will be able to increase the mileage in a sustainable way.

I recently graduated from PT (hopefully forever? knock on wood) but it got pretty exciting in there before I got kicked out. I'm allowed to jump, run as fast as I am physically able (not very fast), stop as abruptly as I am physically able (not very abruptly), and change direction. I'm still not supposed to lift too heavy or do contact, but I basically can do things which feels amazing and makes me want to just frolic around.

Plus, the cartilage is only eleven days away from being the consistency of cheese, which is a milestone that I have been eagerly awaiting pretty much since January 29.

Evenings and Weekends

I spent a day visiting Riley and Sky in Cambridge. Riley made me some fancy coffee with her espresso machine and then we went to Boston Common, took a spin on the carousel, and ate Chipotle in the garden where the Make Way for Ducklings book takes place. While I was in the eastern part of the state, I also went to my cousin's hockey game in Needham (he's ten I'm pretty sure). They lost 1-0 but his team was a lot better than I expected them to be. Then I went out to dinner at Bertucci's with my family.

I've also continued to attend Smith soccer's home games. My parents seem to not know what to do with themselves on the weekends without games to watch, so they've come up to watch some as well and so I also had a nice walk along the Mill River with them and have gotten some free dinners. It was great to see a bunch of people who came from near and far on Senior Day and to celebrate four of my favorite people. After, we went out to Packard's and discovered that we are all terrible at pool.

Mountain Day is way less exciting as a staff member, but I got my donut and shirt in the morning and then had a fairly productive day at work and then ran up Mt. Holyoke and then went to my book club. All in all, it was a pretty nice day.

I seem to have been reading less this month based on the number of books completed, possibly due to the time spent at work waiting for the pipeline and also the lovely time spent at Fruit St., who have been gracious to welcome me into their home an sometimes even feed me. I've also been baking a bit more. I've made brownies, scones, and chocolate chip cookies recently.

This weekend, I'm in Boston mainly visiting for Katie's birthday, but last night Katie, Chelsea, Sky, Riley, and I went out for dinner and drinks. Today, Katie, Chelsea, Alex, and I hiked Cannon Mountain in New Hampshire. The hike was nice except at the summit where there were hoards of people who had taken the tram to the top. Then we went to an awesome ice cream shop called Super Secret Ice Cream with fun seasonal flavors.

Grad Applications

No comment.

Books Read This Month