A small insignificant dot at the end of a long arduous transitive closure.
TBD
A lot of my other fennel+love2d projects got too abstract to begin with As much as I want to define-syntax via fennel macros to build languages for various gamedev endeavours, I’m changing my philosophy to instead build modules that are as small as possible, with zero abstraction, and then learn from those to build a higher level languages
An SDF raymarching engine for VJing and a voxel raymarching engine At some point I need to split these engines up an finish them I tried to abstract out a universal graphics pipeline too early
technically complete, but i want to add bloom gaussian kernel generation rather than having a static kernel, as well as integrate it with my VJengine that will be made flexible enough to support all shader graphics pipelineing eventually
right now it’s just some assorted macros and machine learning algorithms… i want to explore macros that allow you to just write these algorithms with no specification of dt, and it automatically breaks the algorithm steps down frame by frame, with racket-like stepping and replay controls
evaluation results available, one day i’ll write luaJIT for WASM
WE WON OUT OF TWO DOZEN TEAMS!!!! WOOO~ Fauna~otchi is a game about smashing bugs, and watering a plant on a silly console. One of my first big Ws in years, and a great experience. I love the Lisp Game Jam and Fennel community! :D
First time successfully collaborating. Despite the challenges, I had fun and appreciated my partner’s unwavering enthusiasm.
Somehow finished a game and placed in 8th / 48 entries, while recovering from major surgery complications. I spent most of 2024 in bed in severe pain. Proud of the person who came out the other side of that year.
A partner criticized by my previous colleagues turned out to be the only reason I’m still alive today. While I won’t go into the gorey details of my CSA, this surgery made me feel like my own person for the first time in my life.
Reminded me how much I love teaching, especially when I’m free to spend as much time as I want doing so.
Webscraped the local farmers market distributor’s product site, and allowed users to request items at a certain amount and quantity, that would only be ordered if-and-only-if enough users “bought in” on the bulk deal, such that you get a reasonable amount of produce for a single person, at a very reasonable price.
I won’t name the student, but mentoring him through a successful extended abstract and watching him go off to be a successful grad student at another prestigious university, was probably the last good thing I did in academia.
Reminded me why I wanted to be a professor in the first place, though admittedly I sucked at not giving too much of myself to the office hours. I cared less about the grades and more if the students were learning and enjoying themselves. In retrospect, that makes me a pretty poor fit for academia.
Adds softall and maxsome quantification, soft constraints compiled to CNF with weights, and solved with maxSAT (namely z3) Tested the repo and it’s a bit brittle, but I can walk you through it if interested (email me)
My only first authorship, and despite the MTurk data used, it’s probably the best signifier of my academic work
Provenance, ask “why” and “why not” over solution models (FSE 2017 Best Paper and only one to pass artifact evaluation) Tested the repo and it should be fairly solid on anyone’s machine
This may seem like a strange section, but failures are merely learning experiences. Through these projects, I often learned more about myself and the world than a success could ever teach me.
I needed the hard cutoff to move on with my life. I spent a year on leave and not a single day felt like I wasn’t immediately about to return to my program. I wish I had the affordances and support to take a leave earlier and return in better shape. Unfortunately, I did not really have family, finances, or even friends to support such ideal circumstances. I will always regret not finding a better solution between such a rock and a hard place.
My desperate attempt of forging my own path in research with what little fellowship I had left. Thinking there was any hope to combine artifical intelligence models and reinforcement learning agents, with any/all formal methods techniques### without an advisor who had such a vision in mind### was a fools errand.
My post-doc colleague wrote a great paper out of this work, while I barely contributed anything to the project. I failed to have a backbone to admit how uncomfortable defense money made me. Maybe there would have be a positive, amicable outcome### if I stood up for myself and did not avoid conflict. I also failed to repair my lack of affordances and support until years later, which honestly were the largest detriments to my success and ability to speak up.
I may have published a pretty great paper off of such user studies, but in retrospect, I feel so uncomfortable with the folks we paid a living wage to. They were incredibly grateful because I think I was one of few researchers who did the math to make sure they got paid minimum wage by the standards of my country/state of origin, as opposed to handing them the bare minimum. If there was anything as uncomfortable as using an LLM prompt today, knowing the amount of water and watts you just flushed away… it was using this website to crowd source survey takers back then. It’s a shame that there was not an effort in academia to web together a bunch of undergrads getting paid and working under ethical conditions. I think it exists now, but this was the defacto for HCI research at the time.