A little less than a year ago I wrote about four things I wanted to accomplish in 2017. I thought that I would go back and take a look at those goals and then update it for 2018.
When I wrote this, I was still working with the U.S. Digital Service and spent a lot of time away from writing software. When I was involved with software, I was working with older systems or quick prototypes. For this goal, I wanted specifically to focus on a side project. I made some progress throughout the year, but ultimately left the US Digital Service and ended up writing software full-time again. Since that move, I have largely stopped working on Biblio, though I am working with more modern technology in my day job. So I give this a partial success.
I wanted to write at least once a month. I stopped writing after April.
I was a lot more diligent about this in 2017 than in previous years. I would say that I did pretty well in this regard – I’m much closer to being able to be able to at least get by in Spanish, if not fully able to have a conversation. I want to keep working on this one this year.
This one was sort of a gimme – I ended 2016 at 960, and continued to run the Wendler 5x3x1 program through 2017. I ended 2017 at 1025 total, though I had to deload about halfway through the year after taking some time off.
Other than the goals, 2017 was a pretty eventful year! I found a new job, moved back to New York, and got married (here’s a wonderful picture)! With all of this, here’s what I’m looking forward to trying in 2018:
After moving to New York, I started studying the piano in a somewhat-serious fashion. I want to record and release some music this year.
I’ve been working on some Philip Glass works and want to play one of them live. This recording of Glassworks: Opening is one of my favorites. The second Piano Étude from the same album is also wonderful.
Part of inspiration of Biblio was to track the books that I had read a bit better, but I wasn’t able to do that so well, so I’m going to abandon it for Goodreads. Eventually, I might try to figure out a way to host that information on this site directly, maybe through some sort of Jekyll yaml-header magic. Goodreads would obviously be easier to manage, but I wonder that if the increased labor would make me more likely to do it over the long haul.
One of my biggest challenges here is figuring out a way to be helpful in an ongoing way. Over the years, I have contributed to some repositories in ways that I can, often small changes to improve or clean up documentation, fix broken links, etc. I recently contributed a small patch to the code code that powers a ProPublica tool. This was a nice small change where I could be confident that I was being helpful and came with a nice side-effect of being able to look at something totally new to me (Rust!). My patch didn’t touch any of the executing code, so I had confidence that I wasn’t going to cause any regressions. Looking for similar places to be helpful is somewhat challenging. It’s difficult to know where one-off improvements are both welcome and achievable without huge background domain knowledge.
This is by no means a criticism of maintainers of these open source projects – authoring and maintaining a popular open source library sounds like an extremely daunting task. I have read quite a lot of convincing writing on the subject. My goal in contributing to a project is to make small improvements and fixes without being a burden on the maintainers.
It’s been awhile since I’ve refreshed any of the styles on the blog. I’d like to clean it up a bit more and add some features, like generate anchor tag links for the headers, though some quick Googling seems to indicate that this is a feature of Github-flavored markdown that I could enable. I also know a bit more about page speed and accessibility than I used to – I wrote a lot of the core styles and js way back before I really know how to do anything, and tried to clean it up a few years later. I think overall, I could probably delete a lot of stuff I don’t need. I’m loading a bunch of chart styles and a webfont, and those could probably be done away with and managed better using
_includes. I could also make the page a bit more readable by increasing the line height and font size.