lesleylai.info
I organize two online meetups: @gpvm.bsky.social and Programming Languages Virtual Meetup.
lesleylai.info/
mastodon.gamedev.place/@lesley
43 posts
467 followers
215 following
Regular Contributor
Active Commenter
comment in response to
post
It does make more sense for PCSS, though. "Multi-sampled soft shadows" don't convey enough information compared to "percentage-closer soft shadows"
comment in response to
post
Ahh 😅
comment in response to
post
I think the reason partly is that code that's simpler to read is often harder to write. For instance, I often start with an ugly first pass to achieve the desired behavior, then try to reflector the code. Making things simple is purposeful and, paradoxically, isn't simple itself.
comment in response to
post
I hand crafted some. Others are copied from github.com/nlsandler/wr...
comment in response to
post
I don't know that this hash tag exists 😅. Also, TIL that you can't follow hashtags here
comment in response to
post
This feature triggered my past trauma, so I disabled it 😬
comment in response to
post
I happen to have this piece of math memorized, but I wasted way too much time in part 1 to predict that part 2 would be a dynamic programming problem and over engineered.
comment in response to
post
I am probably genocidal by playing Paradox games
comment in response to
post
Somewhat related: some bloggers like to add irrelevant stock images to all of their blog posts, and now they're using horrible genAI images. Not every blog post needs an image.
comment in response to
post
When you want to do some coding and get sidetracked to fix weird issues like this 😬
comment in response to
post
As much as I dislike ads, I'd rather that they have a steady revenue stream.
Besides, there’s always Mastodon, which I prefer anyway.
comment in response to
post
I've met enough competent people with drastically different opinions and coding styles. Some people still prefer not having auto completion and even syntax highlighting, not to mention refactoring tools. The usability of tools won't attract those people.
comment in response to
post
These videos are still some of my favorites
youtu.be/FaOSCASqLsE?...
youtu.be/Brbrdnh74yA?...
comment in response to
post
IMO, using it as a tool is fine, but using it for learning is probably one of the activities that needs the most caveats.
On the other hand, treating it as a "mentor" is overly glorifying what it actually is. I've seen enough university students do that. The result is horrible.
comment in response to
post
Considering that even people significantly smarter than me say that they need to "understand the algorithm myself for hopefully the last time," I don't feel that bad. By the way, this blog post is the clearest explanation of Pratt parsing for me.
matklad.github.io/2020/04/13/s...
comment in response to
post
I tend to use indices exclusively for coding tree and graph-like things nowadays. Pointers are just too error-prone.
comment in response to
post
It is an array of uint8_t
comment in response to
post
It's part of a unit test for an arena allocator. I was verifying that the arena's pointer bump behaves correctly.
comment in response to
post
Yeah, sorry about that. There have been a few times when people in Europe expressed interest in the meetup, but it’s not possible to accommodate all time zones. We decided to accomodate North America and Asia.
comment in response to
post
bsky.social/about/blog/4...
comment in response to
post
Paper notebooks have many advantages, but they are also easy to lose and unsearchable in the future. I sometimes still use paper notes, but I always put my "permanent notes" into software.
comment in response to
post
I dabbled. Wrote a few interpreters before and had a dormant language project after figuring out I didn't have enough skill to write a compiler. I am currently writing a toy C compiler, and hopefully, I can learn more in this process to finally tackle my language project.
comment in response to
post
CSS also evolved a lot since when I learned it, and some advantages for preprocessing and CSS-in-JS like variables just no longer apply.
comment in response to
post
I don't think I will ever devote so much of my time to any single language community again
comment in response to
post
Just read the post. I was intentionally trying not to follow up these things too closely but this post just confirmed my suspicions, and also had tons of things I don't know and don't know whether I want to know. It is overwhelming.
comment in response to
post
It would probably feel more realistic to me if they used a fake number and periodically updated it in the backend. After so many scammy sites using this trick, my brain is already conditioned to assume that any number increasing in real-time as I browse is fake.