In my sincere hope to make sure this blog extends beyond my grad school to first permanent job period, I will be setting aside some very explicit time to write upates here again. Some of my favorite bits of writing live here. I recall getting told once during graduate school that my exam answers “read like a blog post…” I think that was meant to get me to straighten up and write like a spatial scientist, but I actually (unsurprisingly) like the more confessional verbal vernacular way of talking about research.
We’re GLaD you’re here
Maybe this is why I’m still trying to get that podcast off the ground w/ Rachel & Dani. I truly believe our “science” lives in us, and we enact it as we go along very peculiar community norms about what “counts” as a contribution to a body of knowlege. There’s no reason that, a priori, podcasts and blogs can’t do that, except maybe for the data-dependent difficulty of archiving A/V rather than text. The issue, though, is that while science is not a popularity contest, but it’s run like one. And, at a point, I don’t think that I mix well with the fast pace and atomic structure of scientific publishing. Our scholarship should be understood as both particles of distinct “contributions” and a wave forming longer-term thought, thinking, and development.
One thing that has made it particularly hard for me to deal with the pace and structure of my academic workload is coming back from sabbatical. In particular, my school is very bad at managing “crunch”, the periods of intense but temporary over-work that characterises certain periods of our lives. I don’t know if this is as bad everywhere, but this has been a consistent issue for me at Bristol. Normally, I’m very good at just getting through it. But, this was unusually tough. My school was in a bind: I am central to the skills teaching in the department, and they couldn’t find replacements or cover for my time. So, they ended up just concentrating more than a years-worth of teaching into this spring term and, to top it off, most of the teaching was new material. I even won a university award to extend my sabbatical and support teaching replacement. While this was a nice recognition of my outstanding performance so far and an investment in my future, it didn’t, at any point, pay for the teaching replacement that it promised to, it just encouraged my school to shuffle the load around. No good deed goes unpunished; this (among a ton of other reasons) means I think I’ll pass when my sabbatical comes up again.
In spite of this, I’m still very interested in trying to contribute to useful things: I attended the 2023 Developer Summit for
scipy. Building things that people use is the biggest thing I miss the most about working at Nextdoor or CARTO: stuff I wrote actually got used by people in my teams, nearly as I wrote it. I implemented skater clustering in a week to support CARTO folks; I wrote all of cenpy (and much that never was let in) as a way to help Nextdoor work with census data. I miss that in academia, where I tend to implement techniques that only sometimes get eventually pulled into someones’ scholarship downstream. While “science” isn’t a popularity contest, academia is one.
I really enjoyed the SciPy summit. I started by focusing on the sparse arrays work. I use sparse arrays all the time in my own work, since they provide a useful compact representation of geographic networks. I spend a lot of time focused on fast ways to convert geographic data into topologies, so I am very up-to-date with how sparse matrices evolve in the scipy ecosystem. Unfortunately, though, my past attempts at contributing to scipy collided with my own committments, and I always find this hard to navigate. Setting aside time to work together on writing and merging things quickly (a “sprint” in developer terms) is the main way that I find I can be productive on these things. So, it was quite honestly transformative to be involved, and I’m very grateful to Jarrod and Stefan for inviting me to the summit. It was great to be able to work directly with other developers in the sparse matrix ecosystem, get contributions written and merged quickly, and then also work directly with
geopandas folks (Martin, who also came to the summit) to make sure these changes didn’t jeopardise our spatial analysis code downstream. I can understand that, in general, the “long wave” committments that these kinds of projects require will make it challenging to contribute in the long run as a working academic, but I still have very strong, very persistent aspirations to build things that are generally useful. I hope that I can continue to be involved in these kinds of developer communities, and no doubt that the summit will be a highlight of my spring term.
It was really nice to be back in an environment that felt like how CARTO and Nextdoor used to feel; build cool things that are useful for others. Having this at the tail end of my really difficult term… I’m still deciding how I feel about everything. In general, the outlook for the UK has not improved since I moved here, and I’ve had to take on additional consulting work in the evenings to make ends meet due to “Cozzie Livs”. I still enjoy being an academic, but I’m not sure that I enjoy this academia anymore.