finally back (post grad apps)

It has been a while since my last post, mostly due to PhD applications and a big project going on at work, both of which have now pretty much ended which means we are getting back to the regularly scheduled weekly posts and boy do I have a lot in my drafts. This week, I just wanted to talk about random tid-bits here and there from the past two-ish months.

Something that is most recent in my mind is a conversation I had with some friends who are not american citizens. One of them has already gotten picked in the H1B visa lottery but the other is still trying and we were observing how it felt like there was some innate country of origin bias in the process; either each country has their own lottery system or somehow immigrants from certain countries have better chances than others. Probability can be pretty deceiving, however, because based on how you define “feeling that an immigrant from one country has better chances than an immigrant from another country”, this statement and each immigrant having completely equal and independent chances can both be true. The example I had thought about was take a case where I know two australians and ten singaporeans all applying for an H1B visa and realistically each one has a 5% chance of getting chosen. It’s far more likely that at least 50% of my australian friends (aka one person) gets chosen than at least 50% of my singaporean friends getting chosen (aka five people) since you only need at least one australian to get selected but five singaporeans. If this percent of total population is the way you think about the probability of an australian getting an H1B vs a singaporean getting and H1B then even though the actual chance is the same because this country based definition depends on the immigrant population sizes from each country. You can also get pretty specific with the actual probabilities but I don’t feel like latexing the work for that all out right now other than saying if there are two countries with immigrant populations \(x, y\) where \(x< y\) trying to get an H1B visa and the probability of any one person getting the visa in the lottery is \(p < 0.5\). The chance of country X and country Y getting \(50\%\) of their population winning the lottery is \(\binom{x}{0.5x}p^{0.5x}(1-p)^{(1-0.5)x}\) and \(\binom{y}{0.5y}p^{0.5y}(1-p)^{(1-0.5)y}\) respectively (for the uninitiated, the \(\binom{x}{y}\) just means all the ways you can make groups of size \(y\) out of a total number of \(x\) many things). I’m being a little imprecise here for the sake of interpretability but two big caveats: this is all of course assuming that these country’s population sizes don’t have that much of an effect on the individual probability of winning the lottery, and I’m also assuming that both \(0.5x\) and \(0.5y\) are whole numbers. Long story short, the objective probability versus the way humans look at the world are quite different.

In other news, I’m finally restarting a few projects that have been on hiatus for a while. Two of them are some video game ideas that I’ve had. The first, will take a bit of time and is a small dungeon crawl, but the second is a recreation of a small mobile game called “Landing Confirmed”. The concept behind this game is very simple, you have to navigate a rocket around a series of obstacles but the main draw is that the rocket is very fragile so colliding with any obstacle results in having to restart the level. The means by which you control the movement of the rocket, however, is the main draw for me. It’s a bit difficult to get used to but once you get a hang of it, the movement is so intuitive and leaves so much room for skill expression that gives the game a lot of potential. These game concepts and the feeling of mastering the mechanics are hard to explain to non-gamers but all you need to know is that the game is really satisfying to play and master. You essentially control the rocket by clicking on either the left or right side of the screen to control the left and right boosters respectively. The boosters contribute to both rotational and positional acceleration the same way an actual rocket would work. There is no nonsense letting you move horizontally on command, you have to first angle the rocket properly and then push with both boosters (except there is some more nuance in that obviously if you angle a rocket completely horizontally, or generally the more horizontal the axis of movement, the more speed you need to keep the rocket up right to avoid falling, but if you are trying to navigate through a set of obstacles then you need to manage your angles of travel and speed very accurately to avoid crashing).

I’m not sure how far I’ll take my version of the game. I was thinking about some online multiplayer type shebang but we’ll see how I feel as I make progress. I do know that at the very least I want to create a basic version with one level and a rocket and match the feeling of the physics in the game. Regardless, you can find the current stage it is in here.

There was a lot more that I wanted to talk about but not gonna lie I am completely blanking on everything. So in the spirit of not procrastinating I’m just going to shoot this one off. Next week is a food post (which I have been told are some of my more popular ones)!