HuggingFace Jam 2023 happened to coincide with this month’s project. While I’d always intended to have the plan in place by the third day so I can work on it, practicality beats purity.

The theme of the jam is “expand”, which I’m going to take to mean “expand your skills”. A long time ago I wrote “The Rage of Painting” and I’m still proud of what I’ve done there. I’d like to rebuild that game to run on mobile and web platforms, and to use more of the latest advancements in computer vision.

The jam is 48 hours long, but 16 of that will be sleeping for me, at least, then add 12 hours for everything else I have to do. 20 hours of time budget.

4 hours: get a screen in Godot with the ability to draw to it and a variety of colors.

4 hours: get a text UI up and have the tutor give the prompts. (No judging yet.)

4 hours: polish the assets and work off the rough edges. Playable build uploaded to itch.io.

4 hours: try and run uploads of the image to a service for ‘judging’.

That’s 16 hours. The remaining 4 I’m going to write off as buffer.

Let’s gooooo.

Hour 1:

My first big decision is how to actually draw the points to the screen. While I’ve historically used a TextureRect and set pixels based on user mouse coordinates, I’m wondering if I should perhaps try and use a shader here to be faster and more memory efficient. I’m going to spend 30 minutes trying to figure it out before I fall back to the dumb solution (since I have some afternoon obligations).

Hour 2-3:

Unfortunately, using a shader to do the drawing was a no-go. I can’t write to the texture buffer from the shader unless I use a compute shader, and the web build doesn’t support compute shaders. Fortunately, using Image and blitting a brush works fairly well. I’m trying not to overengineer the brushes by adding support for movement speed, brush dynamics, etc. Perhaps I’ll come back and add that.

Hour 4:

I did end up adding some interpolation to the brush to help with smoothness. I thought there might be an issue with the number of substeps, but it still performs okay. I did a build and uploaded it to itch.io to make sure everything was fine on the web and it works even better there.

Hours 5-48:

Things have developed, not necessarily for the better. A planned visit to scout a new apartment and short stay at a birthday party turned into a full day and a half adventure. My Sundays are spoken for already, leaving no time to complete most anything. This was perhaps a complete failure.

I found myself also unable to get motivated enough to carry on with the project through the rest of the month, between disenchantment, apartment hunting, and life in general. I’m going to call July a failure and start again.

This month’s project was something of a success, though I’m late on the write-up. The suggestion came from my partner Lisa, who proposed something like it for our friends’ group.

GooglyEyeBot sits in a server and, with a configurable random probability, replies to a message and applies Googly Eyes. It can also be set to only apply googly eyes when tagged, which I’ve made the default.

Things that went well:

  • First version was very fast to produce. Used MediaPipe with the expectation that I’d find time to retrain a face model when the time came.
  • Discord.py is rather nice, even if “the right way” to do something is a little impenetrable at times. I found myself looking for how messages would connect with each other. I think Discord itself is moving through some changes, so it makes sense.

Things that were tricky:

  • I’m still not entirely sure how to determine if the user sending a command is an administrator of the server (guild).
  • Never got around to making my own face or training set.

Repo: https://github.com/JosephCatrambone/googlyeyebot

Welcome back to the Bro Jogan experience.

Bro: Today we have our long awaited debate between Professor Albert Michaelson…

Albert: Thank you, Bro.

Bro: And Xalthra the Perpetually Vomiting.

Xalthra: *vomits*

Bro: Thanks for coming. So uh, Albert, you’re an expert on gravity.

Albert: That’s probably safe to say, yes. I’m the chair of the physics department at Purdue University and the head of the joint research council for cosmological and applied physics.

Bro: And Xalthra, welcome back to the show. I understand you have a new book on how gravity is a false concept. “Smash-and-Grab-ity: How big aviation wants you to pay more for less safety.”

Xalthra: *vomits*

Bro: We’ve been seeing stories about this all over the place. So Albert, why should we be paying more for less safety?

Albert: That feels like a mischaracterization of the position. I’m not proposing that we pay more for less safety, I’m suggesting that Xalthra’s proposal to completely deregulate the aviation industry is not based on any evidence. Or, really, I’m rebutting the idea that gravity isn’t real for domestic practical purposes, and I’m suggesting that we do have a good understanding of the behavior of gravity on the scale of classical dynamics.

Xalthra: *vomits*

Bro: I think Xalthra is making a good point there. Do you have a rebuttal?

Albert: I’m not sure quite where to begin. There was nothing really to rebut. I’m not sure there were even words in there.

Bro: But you are the expert in gravity? So what do you think about all the people who have fallen out of airplanes and survived?

Albert: I mean, there do exist people who have fallen out of airplanes and survived, but they constistute a small minority of individuals, and even if they did survive–

Xalthra: *vomits again, more violently*

Bro: Yeah, it sounds to me like falling distance and death are uncorrugated.

Albert: Uncorrelated?

Bro: Yeah, like fall more and you’re no more likely to die. I have a buddy who fell out of a fourth story window one night, landed on his neck, and he’s still alive, so why should we be giving the airlines all this money to keep us safe?

Albert: There’s a lot to unpack here. The plural of anecdote is not data. And I’m fairly sure falling and death may well be correlated. Even if they weren’t, you can still test gravity, which is the point. Giving money to the aviation industry doesn’t necessarily mean they’re keeping you safer. It’s not like your dollars go to airplane manufacturers to keep us safe. It’s the regulatory body that’s responsible for ensuring the airplane manuf–

Bro: Let’s go into that. Giving money to the aviation industry doesn’t mean they’re keeping you safer.

Albert: That’s not the best pull-quote.

Bro: I know all kinds of old people that will fall out of their beds and die. The stories are all over the internet, you can do the research.

Xalthra: *vomits*

Bro: Exactly, so really, can any of us know that gravity exists for a fact?

Albert: That’s more of a philosophical tangent. Epistemology is not my specialty and not the route I’m hoping to travel.

Bro: But you’re the one that said epidemiology should track all kinds of deaths?

Albert: Uh.Bro: We have the pull quote, “The CDC is a perfectly fine agency to measure gun fatalities. It’s not perfect, but they’re well equipped to monitor other causes of death, so it’s a good stopgap.”

Albert: Epidemiology is not epistomol–

Bro: So did you or did you not say that?

Albert: That’s not relevant.

Bro: Sounds relevant to me.

Xalthra: *vomits again*

Bro: I think we’re going to have to wrap this up. I’d like to thank our guests for coming on the show tonight.