Forward March
2021-02-14 11:59
by zlg

Work has done very well for me. I received a promotion to supervisor, and just
finished my first week in management. It's... weird. I don't have all the social
skills for this, and I know it. Still, I've been patient with the workload, the
attitudes of my somewhat surprised team, and the sheer amount of learning I'm
doing. For once, I feel like my effort paid off a little bit. I expressed some
cold feet after receiving the news, and my boss flat-out told me, "Look, I get
being self-conscious, but you can't be this way as a supe. I chose you because
you show up every day, do your work, don't create problems, and stay on top of
things when I'm not here. You've earned it. Whatever it is you need to do, find
that confidence -- that inner gangster -- and bring that shit to work." It's
stuck with me. Not sure I can be a gangster per se, but I get the gist.

The girlfriend picked up Taiko no Tatsujin games on the Switch, so we messed
around with them for a while. Steam has the Lunar New Year Sale going on, and I
picked up Sekiro: Shadows Die Twice. Bruh. If you have any love whatsoever for
ancient Japanese motif and high-risk, high-reward gameplay, play it! I'm only
just past the first level and I'm already invested in the story. Controls are
smooth and responsive, and even though my machine technically isn't good enough
in specs (Phenom 2x6 1090T Black) and sounded like it was going to take off, it
held up on Medium settings on 2010 hardware + an RX580. Not bad!

VGStash-web[1] received a refactor. It now only fetches and stores the JSON data
a single time -- on load -- and uses functions as filters for the same set of
data. VGStash pre-sorts when it exports its data, so it's a handy way to not
need to sort *and* filter the list of games. The UI is also more informative,
with tooltips and better visual indication of which filter you're viewing. I
still need to mark "All" as the active filter onload, but otherwise the UI
should be consistent now.

PICO-8 dev hasn't been as successful as I'd like. I'm not happy with the 
structure of my textbox, so I've been revisiting it and re-implementing until I 
get a design that's flexible enough to be reusable in a lot of different games. 
The problem with this approach is that it takes up a lot of tokens, but the 
end-result will be SO damn usable... I'm beginning to understand how APIs get 
made, though. You make up an interface, try it out, see how painful it is, 
revisit the API, rinse, repeat. Perhaps after a while, one begins to understand 
what makes a good API and what doesn't. Learning through trial and error may 
suck, but it's a concrete way to understand flaws in one's understanding of the 
problem. I try to see it as refining my understanding of how to do a textbox. I 
have tons of experience in gaming, so I know how I want it to behave. It's 
making that happen, in a reusable and flexible way, that's difficult.

At the same time, I'm getting an appetite for a less restrictive platform to
do casual gamedev on. TIC-80 is similar with weaker limitations (and is free
software!), LOVE2D uses plain Lua and so is probably more expressive, I could go
grind on SDL2 and C what happens, too... I don't really know, but I think now
that I'm understanding the fundamentals of game development and simulations, I
like this area. It's harder to debug since it's powered by loops and coroutines,
or events in later engines, but it's so satisfying the first time you get a
character moving and interacting with your imaginary world. Each new feature
or effect shows visual feedback of progress in the form of something fun. It's
frustrating when you work on a bug for a whole day and still don't get back to
interactivity... but it's part of learning it.

It's early morning and I had a chance to write. Happy Valentine's Day, I guess. 
:P

[1]: https://zlg.space/misc/vgstash.html