#I also feel this way about my code on github

[reifae]

Hi, Rei. I'm the one who asked you about TS4CC earlier and I'm glad you agreed with me to edit and use your textures. SIMS 4 updates the infants age stage, and I'd like to convert your Mouthbow and Teethbow to a version that works for infants. May I ask, do you agree?

I'm fine with any and all modifications of all my sims4 cc so long as you: 1. Credit me for textures if at all possible. 2. Have no plans to monetize the cc itself. (I'm not hounding you if you have a kofi or patreon just don't lock it behind that.) 3. Also please do not upload to Curseforge. I have fairly FOSS ideals to my CC that I think things can only be improved if we openly allow modification and collaboration. So go ahead and have fun converting or editing as you wish!

[linkons-most-wanted]

LaDS Fanfic & Lore Resource

I am documenting all of the Love and Deepspace lore over on GitHub!

View the WIP project here (Project contains many untagged spoilers!!)

I'm using a free mapping program called Obsidian that's basically a red string board in app form. Everything!! is connected!! The lore is so detailed and it's been so satisfying to go through.

So far, I have the main story up through Long Awaited Revelry, plus all anecdotes and the Beyond Cloudfall myth. I wanted to go ahead and share even though it's WIP since I've already made so many interesting connections.

For example, the format is still a bit rough but this Main Story Timeline gives an approx month/year for every event in the main story. Based on multiple sources, the main story almost certainly starts in July 2048!

More info below

GitHub/Obsidian are easy to set up if you have a basic coding background, but I know it's not the most accessible. If there's enough interest, I'll consider paying to host it through Obsidian so it's more of a normal web page, it's not too expensive! You can still click around and read the project in GitHub, but the document links won't work.

If you maintain a wiki, please feel free to use anything from this project! Mainly why I didn't contribute to a wiki is I wanted to add a lot of commentary along the way, as well as only really tracking what's important for fanfic/lore theories. Take any commentary with a grain of salt, as I'm consistently updating it as I bring new pieces in.

Since the storytelling of the game is pretty non-linear, my hope is also that the summaries of the main story and anecdotes help folks more easily understand the story. The game likes to be mysterious at first, but it often becomes quite clear once all the pieces are revealed.

One of my favorite things about the lore is that it follows romance conventions more than sci-fi conventions, so "because love" is the reason behind a lot of canon decisions 🥹 But there's still a lot of sci-fi to enjoy!

Anyway, I hope you'll forgive me for the WIP and I hope this reaches the other folks who are as obsessive about the lore as I am!

[baileylockheart]

Webmaster Webring!

Do YOU have a website of some sort? Any sort? Neocities, nekoweb, github? Hand-coded? Wordpress? A carrd? Maybe a fancy-looking rentry? Do you have a custom tumblr theme that you fancied up to be a little website all on its own?

well, if ANY of those apply, or if you're interested in getting them to, I'd like to invite you to the webmaster webring!!

this is a bit of a follow-up to my post about the many perfectly valid ways to make a website! to quote the about page of the webring:

So many webrings are out there, and it's wonderful getting to surf through them, but the vast majority of them specify that all sites must be hand-coded, and I feel like that simple restriction blocks out a lot more people than we'd like to think. Interesting and creative sites can be made with site-builders, and websites built on pre-made layouts are still websites!! Because of this, I wanted to make a webring that allows sites of all kinds :)

Many people overcomplicate what being a webmaster is, and I've seen a lot of people including owning a domain and hand-coding everything in their definitions. In truth, though, if you're a person who's got a website, you fit the bill!

I made this webring with the intention of being open to all kinds of sites, whether they're hand-coded, made with site builders, use static site generators, or are based on premade layouts. I want more people to make websites, so I won't turn someone away for making a website.

In addition to being a webring, I also have a page dedicated to resources meant to help people build their own sites! Whether you have a website or not, I'd like to invite you to take a look at the webring regardless. If you have a website of just about ANY kind, you can join, and if you don't, I have resources linked in hopes of helping you change that, if you're interested!

This post is less of me begging you to make a website in whatever way you can, and more just shilling for my webring, but the point still stands that I think everyone should have a website! If you agree, come take a look :D

[matthewzkrause]

How AI Broke My Coding Flow And How I'm Getting It Back

A little while ago, coding felt like a superpower. I could sit down, get in the zone, and hours would fly by while I pieced together ideas, logic, and lines of code into something that actually worked. It was frustrating sometimes, sure—but it was also deeply satisfying. Then AI came along, and… things got weird.

The Honeymoon Phase When I first tried tools like GitHub Copilot and ChatGPT, I was blown away. You mean I could just describe what I wanted and get code? Instantly? It felt like cheating—in the best way. I started moving faster, writing cleaner code, skipping the repetitive stuff. Honestly, it was exhilarating.

But slowly, something started to shift. I didn’t notice it at first. I just knew I wasn’t getting into that deep "coding flow" like I used to. I was moving fast, but the joy and the why behind what I was doing started to fade.

Losing the Thread At some point, I realized I was letting AI take the wheel a little too often. I’d let it write functions, name variables, even handle logic I didn’t fully understand. It worked… until it didn’t.

I started to lose confidence in my own instincts. When I didn’t have AI right there, I’d freeze. Even basic tasks felt harder than they should have. I wasn’t building things anymore—I was just assembling pieces someone (or something) else gave me.

The worst moment? A code review where I couldn’t explain my own logic—because I didn’t really write it. That hit hard. I had outsourced the thinking part of coding, and I missed it more than I expected.

Finding My Way Back That was the wake-up call. I realized I didn’t want to be someone who just glued together code suggestions. I wanted to be a developer again—someone who understood the why, not just the how.

So, I decided to do something about it.

Going AI-Free (Sometimes) I started with a personal project and promised myself: no AI. Just me, a blank editor, and Google if I got stuck. It was slow. I fumbled a lot. But you know what? That sense of satisfaction came rushing back. I remembered how it felt to solve something on my own—and it felt good.

Treating AI Like a Collaborator Now, I still use AI—but differently. I ask it questions, explore options, or get feedback. I don’t let it drive. I use it the way I’d use a senior dev on my team: as someone to learn from, not someone to do my job for me.

Relearning the Basics I went back to old-school stuff: data structures, algorithms, writing small functions from scratch. Not because I needed them for my job, but because I wanted to rebuild that muscle. Coding isn’t just about getting something that works—it’s about understanding how and why it works.

Talking Through My Code Again I started pairing more, doing code reviews, even talking out loud while I code. That helped me rebuild my ability to explain what I was doing—and catch myself when I didn’t fully get it.

What I Learned I don’t blame AI. It’s an incredible tool. But I learned (the hard way) that how you use it matters just as much as what it can do.

Coding is more than getting output—it’s about thinking, learning, solving, creating. I’d forgotten that for a while. But I’m back in the flow now, and it feels like coming home.

If you’ve been feeling the same—like your brain is getting a little too quiet during coding sessions—you’re not alone. Step away from the autocomplete. Get your hands dirty. Struggle a bit. You’ll be surprised how fast your mind remembers what it used to love.

[kitscodingblog]

Coding Update 6

I think its been a while since I've updated. I fell behind a little on my learning cause life has been really difficult lately.

Hope y'all had a good Thanksgiving and having a good start into the holiday season!! Yadda yadda more under the cut.

So I just finished Part 1 of my book. This mostly contained the introduction to Python, obviously, while learning a lot of the major functions of the program. I think it took me a bit to get into the swing of coding, especially cause it felt like I've had to rewire my own brain doing this haha.

The good news is I feel a LOT more comfortable with Python now. Not like "i can do anything!" yet but enough that it's actually super fun and I'm excited to work on projects!

The last part of the chapter taught me to use the "pytest" ability. I.E: writing test code so that I can make sure my programs are working properly and as intended. That part was really interesting, mostly because it was super duper busted at first for me.

That ended up being because where my "default folder" is set is like my main python hub, so i have to use the uh. What's it called? True access link? Where I write the entire string to the code's location.

Which also taught me that in the Terminal I have to use quotes for the location cause before I learned proper coding practices, I used spaces in some of the initial folders.

We're good now though.

The next part, Part II, is all about learning to build fully functional projects!! I'm so HYPED. There's four projects, of which it was like choose whatever you want! But I'm gonna start with an Alien Invasion remake. You know, the game where you're the little ship at the bottom shooting at aliens as they slowly decend on the screen. I should learn a lot from this one.

The other project I'm looking forward to is a simple online blog database. It'll have users create accounts, be accessible online, and you can make little journal posts! That should actually teach me a lot of stuff that I want to do.

There's another for data visualization, which I think I'll send to my cousin. He works in a lab at MIT and I know they use python for their programs. Maybe I can work my way into his work by doing that lmao.

Anyway, I'm really excited for all of this. It should teach me a ton of usable skills, and then i can add these projects to my portfolio to show off. Also I can spin off and make my own stuff.

Also also, if anyone wants to help me test my projects, feel free to let me know! I already know a few who are more than willing, but I'd appreciate any and all feedback as I go.

Oh! It also recommended learning version control, which I know almost nothing about. So I'll learn how to use GitHUB to store projects and recall old ones as I go if things break horribly. Which will be fun! Cause I know that for sure is going to be an important skill to have.

For a last fun fact, did you know places are like "requirements: typing 30 words per second." Do you know how fast I can type? At my peak I'm like over 110. I baseline at like 95. I don't know if that's actually fast but it makes me feel like the specialist little guy.

I hope you all have a good holiday season. Sorry no code in this post, I'm writing it so I can give you all an update, and I'm dog tired today. But but I promise to snip actual code for you as I go forward. And It'll be fun, especially cause this alien project will teach me about making VISUAL things!

Seasons Greasons Tumblr! -Kit

[asyncmeow]

how shinigami eyes works

first - the short version:

shinigami eyes is a browser extension that uses a fancy list called a bloom filter to identify people who can be considered transphobic or trans-supportive. a bloom filter is a kind of list that lets you check if someone is in the list, without being able to actually know the contents of the list. i go below into how it works, but the short version is Math™️, with a side effect of having false positives - thinking an item is in the list when it really isn't. whenever a social media username is detected (the exact method how depends on the website), it checks the username against that list, and applies some CSS to change the color of the text depending on what list the username is in.

the long version

shinigami eyes can be split into a few parts:

bloom filters

submissions

name highlighting

bloom filters

i'll start with the most complex part - bloom filters. the most common misconception about Shinigami Eyes is this: the filters are not updated in real-time. they are shipped with the extension which had last been updated since november 2022, according to the FF extension site. in other words: nothing marked since then can be seen by anyone other than who marked it. you can see that in the code here, where it loads the bloom filters from a data/[something].dat file included in the extension (but not in the github repo).

the following information about bloom filters is my summarized version of this page.

bloom filters are, in a slightly longer explanation than before, a way to know if an item is *not* in a list with 100% certainty, but there's a false positive rate that grows as more names are added to the list. a bloom filter of a single size is able to handle any number of items in the list, though. there's also the issue that you can't *delete* stuff from a bloom filter - you would need to regenerate it from scratch to do that.

now that the medum-sized explanation of what they are is done, let's go into how they work. a bloom filter is a set of n bits, initially all set to zero. to add items into the filter, you need a few hash functions, in this example i'll use h1, h2, and h3, with n=10. if I wanted to add the text asyncmeow to the list, i would do this: h1("asyncmeow") % n // n = 10, h1(...) % 10 = 9 h2("asyncmeow") % n // n = 10, h2(...) % 10 = 5 h3("asyncmeow") % n // n = 10, h3(...) % 10 = 8

after that, i have a list that looks like this (keep in mind that the list is zero-indexed):

[0, 0, 0, 0, 0, 1, 0, 0, 1, 1]

you can then check if something is in the filter by running the same hashing functions and checking if the result bits are set in the filter. you can access the bloom filters used for shinigami eyes by going to about:debugging#/runtime/this-firefox in Firefox and clicking "Inspect" on Shinigami Eyes, then running bloomFilters in the console. as this could possibly change in an update, i don't want to go into how they are set up, and i haven't dug enough into how their bloom filter code works well enough to say anyways.

submissions

submissions on shinigami eyes are encrypted (as in - encryption separate from HTTPS), then posted to https://shini-api.xyz/submit-vote. you can see the code for this here. when you right click someone to mark them, their name is stored in the local data of your browser in an overrides property.

name highlighting

name highlighting is done by checking them against the bloom filters and your local overrides. if a user is present in either bloom filter, or present in your local overrides, they are marked accordingly. not much to it from there.

that's really it, i think? feel free to ask if you have any questions! nya :3

[deardarlinggames]

Devblog 1

Hi! My name is Wendy, I'm a software developer, and the creator of Dear Darling Games. I'm going to be using Tumblr as a more relaxed and unpolished blog. For now, because I don't know how to format a blog post professionally quite yet. Later, I'll want a break from the structure of it, and have many things I feel I can chat about.

About me! I'm currently twenty two years old, and my goal is to be a solo game developer with a heavy focus on visual novel RPGs. I have no experience, and I'll be logging and citing my entire process as I learn to the best of my ability. Even figuring out the right questions to ask has been quite the task.

Tonight, I'm starting at square one. I'm using an IdeaPad laptop, and operate on a night-shift schedule due to work and life circumstances. This means most updates will be around four in the morning for me. I'm head of household in many ways, so sometimes I will have to step away for a day or two to get my affairs in order.

Here's all I'm learning, planning to learn as of now, and what I have so far.

Planning to learn: HTML, CSS, JavaScript, Godot Engine, Procreate, and Blender. 2D paper-doll animation, rigging, video editing, layering, how to use Alpha lock... Sound design - cello, violin, piano, flute, foley techniques, and general sound equipment. Navigate and create - a website, a put together GitHub profile, and my first game pair; a 2D Mouse themed VN RPG, and a 2D farming game inspired by Zombie Farm with significant changes to the storyline, main mechanics, and characters. In essence, a reworked fan remake, and it will be free to play. I will also be learning how to navigate matters of intellectual property, copyrighting, and more in that area. Finally, I'll be learning Bootstrap, Sass, and React and Redux to create Single Page Applications.

Learning now?: HTML, CSS, Godot Engine, Procreate. Foley techniques, and I've officially gotten down plucking scales on my cello. No luck with the bow yet... I'm refreshing my guitar skills, and saving for a keyboard. I'm utilizing RPG Maker to start familiarizing myself with very, very basic aspects of how to communicate with the computer. Plus, it gives fast results which help lulls in attention span for learning how to do it all myself. I am not planning on publishing a game with RPG maker for *professional* purposes to illustrate learning or ability, but it is very fun to use.

What I've got: Full storyline and pathway branching for decisions, voice bits and character voices [done by me], snacks, water, and a dedicated workspace. Character sketches on paper, most of the dialogue, a working title for both, and most of the battle and other systems planned. It feels like all that's left to do is code and Learn How to Draw Digitally, but that's sure a lot when you have to break it down into all the little steps and refocus those into groups and whatnot. I also have a GitHub profile and joined their Developer Program, I have this tumblr, a Jira account to break down tasks and to-dos in a more manageable way, and my Neocities website has officially been set up to the point of Having It.

I'll release a pinned post with my production announcements later.

fin: 4:04AM

[allynoire]

✦ About

Hi! I'm Allyn (they/them)

I study computer science, and enjoy drawing every now and then. I'll mainly post my art here, but I might expand this, to include other things I make as well.

I'm non-binary and working my way towards some form of androgyny. I don't know whether you need to know this, but I like to share, I guess :)

I love playing TTRPGs (especially Cairn) and also do some improv.

I'm interested in making new friends, so if you wanna code or do something together, feel free to hit me up!

✦ Links

I'm not active on any social media right now, but if you like, you can check out my code on github or play my crapy games on itch :)

pronouns

github

itch

[bloobluebloo]

Hello!

This might be a weird question, but since you work in IT, do you use AI tools like ChatGPT or Claude a lot, or not at all? I’ve been learning programming for a few months, and honestly, it’s super hard. I’m definitely not a genius, so I use AI a lot to help me figure out what I’m doing and generate code.

The problem is that other students kind of judge and look down on people who use these tools, and it’s making me feel bad about it. Should I stop using AI altogether? I just don’t know how to manage without help or researching all the time.

If you have any tips, they'd really help me out 🙏

Thanks for reading this!

Hey anon! Well, the thing is that the IT industry in its entirety is pushing for AI integration as a whole into their products, so industry-wise it has become sort of inevitable. That being said, because we are still early into the adoption of AI I personally don't use it as I don't have much of a need for it in my current projects. However, Github Copilot is a tool that a lot of my colleagues like to use to assist with their code, and IDEs like IntelliJ have also begun to integrate AI coding assistance into their software. Some of my colleagues do use ChatGPT to ask very obscure and intricate questions about some topics, less to do with getting a direct answer and moreso to get a general idea of what they should be looking at which will segway into my next point. So code generation. The thing is, before the advent of ChatGPT, there already existed plenty of tools that generate boilerplate templates for code. As a software engineer, you don't want to be wasting time reinventing the wheel, so we are already accustomed to using tools to generate code. Where your work actually comes in is writing the logic that is very specific to the way that your project functions. The way I see ChatGPT is that it's a bit smarter than the general libraries and APIs we already use to generate code, but it still doesn't take the entire scope of your project into consideration. The point I am getting at here is that I don't necessarily think there is a problem in generating code, whether you are using AI or anything else, but the problem is do you understand what the code is doing, why it works, and how it will affect your project? Can you take what ChatGPT gives you and actually optimize it to the specifics of your project, or do you just inject it, see that it works, and go on your merry way without another thought as to why it worked? So, I would say, as a student, I would suggest trying not to use ChatGPT to generate code, because it defeats the purpose of learning code. Software engineering as a whole is tough! It is actually the nature of the beast that, at times, you will spend hours trying to solve a specific problem, and often times the solution at the end is to add one line in one very specific place which can feel anticlimactic after so much effort. However, what you get from all those hours of debugging, researching, and asking questions is a wealth of knowledge that you can add to your toolbox, and that is what is most important as a software developer. The IT landscape is rapidly changing; you might be expected to pick up a different programming language and different framework within weeks, you might suddenly be saddled with a new project you've never seen in your life, or you might suddenly have something new like AI thrown at you where you suddenly have to take it into consideration in your current work. You can only keep up with this sort of environment if you have a good understanding of programming fundamentals. So, try not to lean too much on things like ChatGPT because it will get you through today, but it will hurt you down the line (like in tech interviews, for example).

[lotusprotocol]

long ramble about how i optimized my godot game for size (and how you can too!)

if i export the executable for dreamcatcher with the currently used scenes and the default export template, then the file size is 76.2mb. since it’s still early in development, i’m expecting this to get a lot bigger as i keep developing the game, and it will likely be at least 10x its current size.

i think a good amount of people reading this are familiar with the western game dev stereotype of having an executable that’s tens, if not hundreds of gigabytes in size. i’d like to avoid this as much as possible, since i want my game to be accessible to people who don’t have tons of storage to spare. also, i generally enjoy optimizing things to their limits, which is why i do tool assisted speedruns often (which will also be built in to dreamcatcher).

anyways, back to optimizing for size. a while back, i was doing some research on the topic and found this article on the godot docs: https://docs.godotengine.org/en/stable/contributing/development/compiling/optimizing_for_size.html

if it weren’t for this article, i would not have been able to optimize the file. also, if you’re a godot dev, i highly recommend looking into this!

recently, i decided to try out optimizing, so the first step was to download the godot source code (which you can find on the github repo), and download scons to build the source. here’s the links for those:

godot source code: https://github.com/godotengine/godot

scons: https://scons.org/pages/download.html

the second step was to run scons from the source code directory. this can be done by going to the command line, navigating to the directory, and running this command:

scons platform=[windows/linuxbsd/macos/web] target=template_release tools=no debug_symbols=no lto=full optimize=size [other options]

this will create an export template and put it in a new directory named bin. it takes about half an hour to build for the first time, so make sure you have something else to do while you wait, since it also uses a lot of your cpu. other options are dependent on the game you’re making, but some include disabling 3d, disabling advanced text, and disabling specific modules.

also, using the optimize=size option makes the build prioritize size over performance, which is generally done for web builds. if you’re making something that needs as much performance as possible, consider omitting this option from the command. however, in my case, the game is capped at 60fps, and i’ve already been optimizing for performance when making the game (which i might get into in another post)

the third step was to set up the export template. you’ll have to move the build into your godot export templates folder, and rename it accordingly. this video does a good job of explaining it: https://youtu.be/zKq25lXlsUE?si=AOoMVHdI6cIVG7FP&t=407

the video also explains other concepts that are mentioned in this post, so feel free to look around in it as well.

the fourth step was to export the project with the new template. go to project > export in your godot project, and make a new preset for the custom template. set the custom template path to the template you installed, select the resources you want to export if not all of them, click export project, and you should be good to go!

now, how is this whole wall of text worth it? well, i was able to cut down my file size from 76.2mb all the way to 51.6mb, which is a huge margin. and that’s before the fifth step: zipping the file.

the best way to do this is to download 7zip, which produces more optimized zip files without affecting other areas of performance. you can find it here: https://www.7-zip.org/download.html

once you have this downloaded, you can archive your executable, and you’re finally done! if you’re using the command line to do this, here’s the command format:

7z a -mx9 archivename.zip path/to/executable

zipping the file cut down the size further from 51.6mb to 28.1mb. overall, that’s almost 3x smaller than the original file, which is a really big improvement.

size comparison across all three files

if you made it all the way to the end of this post, thank you so much for reading! i hope this helps :>

[kaiclavier]

Would you still recommend Unity now that most of the storm has blown over? I was going learn it originally but then all that happened and now I dunno what engine to learn

There's a lot more I could say about the internal state of Unity, but I'd just consider what kind of games you want to make, and find an engine that matches. The reason I went with Unity in the first place is that it did these 3 things:

All coding can be done with a single non-proprietary language

Doesn't care about a hard divide between "2D" and "3D".

Can export to PC, Web, and consoles with no external tools.

I'm also a big fan of the object-based system Unity has going on! It lets me keep projects incredibly modular, and separate art from code when I need to. (eg: I can have a tween that scales a sprite to be the size of another sprite... and that other sprite I can just resize with editor tools.)

There's a bit more that's made me stay with Unity, but all of those are from the support of 3rd-party developers:

Good controller support (ReWired)

a good animation system (Mecanim, which is now Unity's internal animation system)

a convenient text renderer (Super Text Mesh... I spent 8 years on it! It's hard for me to drop!)

MIDI in *and* out support (DryWetMidi and keijiro's work)

Good tweening library (DOTween)

Good code editor (Rider)

I've been eyeing godot myself since it's similar in concept to Unity, but it doesn't have everything I need on my list, yet. It is quickly approaching though, and since the engine is open-source... I think it'll get there sooner than later.

So, I'd make a list of the kinds of features you want out of a game engine, and start comparing engines! For example, if you only want to make 2D games, maybe GameMaker or Monogame could be worth considering. Unity has a ton of 3rd party support from asset devs, which is what I feel makes it a strong contender for anything still. (I just wish we asset devs were treated with a bit more respect...!) I wouldn't worry about Unity going bankrupt or anything by the way, if that were to happen, either a) some dev would probably upload the source code on github the next day, or b) projects to convert Unity code to Godot code automatically will accelerate extremely quickly.

[unlimitedtrees]

so the game i made, 'UNITRES Dreams', is open source on github now

https://github.com/UnlimitedTrees/UNITRES-Dreams

hello. earlier i made my game 'UNITRES Dreams' open source on github. i want to talk about it. click the little 'read more' thingy to read da rest of this post. i had a lot of things i wanted to say about it and why i think open sourcing games is kind of awesome Lol !!!

also , By The Way, play da on newgrounds ehehehehe ~! or u can download it on itch.io too if you want

so. let me start off by saying that, although i was planning on releasing the source code for unitres for a long while, i was Not prepared to release it now. in fact, it kind of happened on accident.

in case you didnt know, earlier on twitter i was talking about how i wished there was a good way to release my archive of old builds for the game. i have a google drive full of old builds but i also i dont like using google drive. i sent a link to the drive in a reply to someones tweet, but there is one big thing i forgot: the game's project file is also in that drive.

anyways, instead of just deleting the reply or removing the file from the drive, i decided to just release the source code on github now, since i might as well. a part of me feels like i shouldnt have released it so suddenly... but also a part of me feels relieved.

you see, the unitres dreams source code is Bad. tons of messy code with almost no comments... a bunch of hackish solutions to tons of problems... everything is just incomprehensible unless youre me. considering ive been the only person working on the game for years, i understand most of the code and never needed to make comments (especially since making comments in construct 2 is Annoying).

my original plan for releasing the source code was to go back and clean everything up; improving tons of messy systems and adding comments explaining how everything worked. i also was planning on adding new content for a new update to the game aswell.. such as an options menu and maybe new levels and characters or something. i have Tried going back a few times to clean some little things up and work on a keybinding system... but i havent had the time to focus on it and do what ive wanted to do with it.

so, yeah. im kind of glad the source code is just Out now. it probably wouldve taken me Years just to do what i planned... and i think its best to just get it out now instead of just waiting years for something that might never happen, even if the source code right now is incredibly insane and unusable.

which leads me to something most people might be wondering: Why even release the source code in the first place? honestly, i doubt a lot of people will get use out of it, especially with the bad code and it being made in an old, dead engine that you cant even Buy anymore. but despite this... i think its good to release the source code for your games for a few reasons...

for one, its good for archival reasons. ive had a huge anxiety over losing all the source code for my games... its happened to me before and it really hurts. not only can you not use it in the future, but you lose a lot of work youve done, and for me ive lost a ton of games which were important to my journey as a game dev. nowadays, i try to hold onto all of my projects and have saved them to multiple computers and hard drives, but releasing them on stuff like github and google drive at least makes me feel a little better.. and hopefully with the unitres source code being public and the google drive for the old builds being out, it means that i wont have to worry about losing either of them.

another big reason is that, even if no one uses the source code for anything like mods or whatever, i think games having their source code out is good for learning game dev and understanding how games are made, even if the code is a big mess. even if i did get to go with my plan of cleaning up the source code... i feel like thatd be kind of dishonest, cus Most games Never have perfect code when theyre made. its good if you use the most messy code in existence just to get the game done... you can't always spend forever perfecting something, and at the end of the day the game eventually has to come out at some point. even big triple A projects will have weird code in some places... thats just a fact of life. i think having the source code out lets people see the kinds of crazy stuff that goes into a game, and maybe seeing my crazy code will influence others to do things a lot better in their own things. that's what i think, anyways.

and hey, even if My game never gets any sort of modding community, that doesnt mean Your game wont. there are a lot of communities out there who make some crazy things with mods.. and things such as open source games or decompilations can lead to some incredible stuff. with decompilations, youve got things such as mario 64 or the retro engine sonic games being ported to pretty much Everything, and then youve got open source games like Doom and friday night funkin which are being pretty much kept alive Because theyre open source and have a huge modding community. even if a game never gets as much interest as those.. it doesnt mean it never will, and having a game be open source can help a community grow so much more.

anyways umm yea. i think making games open source is awesome. there is a little part of me that feels weird about making unitres dreams specifically open source and the possibility that people might make fangames and mods out of it, as its something thats really personal to me (especially since the main character is named after me), but if people do make fangames and stuff with it i dont think ill mind, especially since a lot of my early work was fan games and unitres dreams takes a lot of inspiration from the media i love. all that i ask is that people include the original credits from the game in their thing, especially if you reuse any of the music from the original soundtrack.

i think, at the end of the day, my goal with this is to inspire more people to open source their games. a lot of good things can come out of it. that's all i have to say i think. play unitres dreams on newgrounds dot com ehehehe !!

[rossk]

Musings of an LLM Using Man

I know, the internet doesn’t need more words about AI, but not addressing my own usage here feels like an omission.

A good deal of the DC Tech Events code was written with Amazon Q. A few things led to this:

Being on the job market, I felt like I needed get a handle on this stuff, to at least have opinions formed by experience and not just stuff I read on the internet.

I managed to get my hands on $50 of AWS credit that could only be spent on Q.

So, I decided that DC Tech Events would be an experiment in working with an LLM coding assistant. I naturally tend to be a bit of an architecture astronaut. You could say Q exacerbated that, or at least didn’t temper that tendency at all. From another angle, it took me to the logical conclusion of my sketchiest ideas faster than I would have otherwise. To abuse the “astronaut” metaphor: Q got me to the moon (and the realization that life on the moon isn’t that pleasant) much sooner than I would have without it.

I had a CDK project deploying a defensible cloud architecture for the site, using S3, Cloudfront, Lambda, API Gateway, and DynamoDB. The first “maybe this sucks” moment came when I started working on tweaking the HTML and CSS, I didn’t have a good way to preview changes locally without a cdk deploy, which could take a couple of minutes.

That led to a container-centric refactor, that was able to run locally using docker compose. This is when I decided to share an early screenshot. It worked, but the complexity started making me feel nauseous.

This prompt was my hail mary:

Reimagine this whole project as a static site generator. There is a directory called _groups, with a yaml file describing each group. There is a directory called _single_events for events that don’t come from groups(also yaml). All “suggestions” and the review process will all happen via Github pull requests, so there is no need to provide UI or API’s enabling that. There is no longer a need for API’s or login or databases. Restructure the project to accomplish this as simply as possible.

The aggregator should work in two phases: one fetches ical files, and updates a local copy of the file only if it has updated (and supports conditional HTTP get via etag or last modified date). The other converts downloaded iCals and single event YAML into new YAML files:

upcoming.yaml : the remainder of the current month, and all events for the following month

per-month files (like july.yaml)

The flask app should be reconfigured to pull from these YAML files instead of dynamoDB.

Remove the current GithHub actions. Instead, when a change is made to main, the aggregator should run, freeze.py should run, and the built site should be deployed via github page

I don’t recall whether it worked on the first try, and it certainly wasn’t the end of the road (I eventually abandoned the per-month organization, for example), but it did the thing. I was impressed enough to save that prompt because it felt like a noteworthy moment.

I’d liken the whole experience to: banging software into shape by criticizing it. I like criticizing stuff! (I came into blogging during the new media douchebag era, after all). In the future, I think I prefer working this way, over not.

If I personally continue using this (and similar tech), am I contributing to making the world worse? The energy and environmental cost might be overstated, but it isn’t nothing. Is it akin to the other compromises I might make in a day, like driving my gasoline-powered car, grilling over charcoal, or zoning out in the shower? Much worse? Much less? I don’t know yet.

That isn’t the only lens where things look bleak, either: it’s the same tools and infrastructure that make the whiz-bang coding assistants work that lets search engines spit out fact-shaped, information-like blurbs that are only correct by coincidence. It’s shitty that with the right prompts, you can replicate an artists work, or apply their style to new subject matter, especially if that artist is still alive and working. I wonder if content generated by models trained on other model-generated work will be the grey goo fate of the web.

The title of this post was meant to be an X Files reference, but I wonder if cigarettes are in fact an apt metaphor: bad for you and the people around you, enjoyable (for some), and hard to quit.

[hardwarevent]

2025 Resolutions

Hello! It is January 3rd, 2025. Sorry I've missed the resolutions post from 2024. I'm very much aware that no one reads this blog and it's more for my own personal reference, but still. About twice as much stuff has happened since the last post in relation to the other resolutions posts as a result. Q2 and Q3 2023 was mostly wedding planning. The wedding went well! We had an incredible honeymoon in Bora Bora. While we were out there, my grandfather died, so I had to travel soon after arriving back in California to go back to my hometown. That was not a great time at all. Fast forward a year from that, and my wife and I spent our anniversary in Florence, Italy! This was the final leg of a trip to Scotland to visit my mom, who lives there after retiring. At the beginning of 2024 my wife and I moved into a townhouse that we rent not because we can't afford a house (although ehhhh), but because there are way more rentals open than actual houses to purchase.

Work has changed a lot. At the beginning of 2023, we started shifting a lot to AI (Wow! Amazing! What a surprise!). I got the previous project to a good place, then switched to making GitHub Copilot-likes and AI Assistants. That's where I've been for the past year and a half. Frankly, I'm sick of it. Luckily, I might be working on something that our users will actually use and find useful soon, maybe. Also, I'm just a regular senior software engineer now. They took away my direct reports. Well, "taking away" isn't totally accurate as much as my manager saying that I am a better engineer than manager, and me agreeing with him. Besides, that principal engineer in the previous post? Gone, kicked out. Another direct report left from burnout mid-2023. Another got kinda sorta pushed out in 2024. All of them I was happy to see go, for a variety of reasons. Just one person is left in terms of people I used to manage. Am I happier? For a while no, as I was working through these cloudy feelings of failure for a while. Then yes, when I realized I was indeed happier. Though right now, it kinda varies from project to project. There's this concept of a "staff engineer", where generally, you give up individual coding for more meetings, coordination of people, alignment, glue work, mentoring, etc. The idea behind it is that this is the "next level" beyond a senior engineer. Coding a lot of stuff really fast can only get you so far up the food chain. To keep climbing, you've got to increase your scope and influence. I've been reading books and focusing on that for career goals lately. Since the last post, the stock price has gone down a LOT. Then it went up some. Then it went down some. And so on. Right now it's up, and seems to be staying there, knock on wood. The tradeoff of the stock going down is that my wife works at the same company I do, and makes about as much and I do, and together we make a relative boatload of money. The stock is worth less, but I have more of it. Could I retire? Still probably not. But! It's enough to be a cushion I could theoretically coast on for a lifetime, if I'm smart about it and cut back a lot of things. The economy has gotten really shaky for tech workers. A lot of people have lost their jobs, and months have gone by without getting another one. Linkedin is full of these stories, and it's scary. As much as I'd want to explore a break, I feel really lucky to have what I have. Especially as the video-game-like trend continuing of things just getting harder (looking at you, incoming president).

I just finished reading a book called Tomorrow, and Tomorrow, and Tomorrow. it was about two very close friends who start a game company and make some games over the course of their lives, and go through various states of their relationship during those times. One thing it made me think about is time. Time moves fast, and as you get older, it feels like it moves faster. I think this was smartly illustrated by the book by more and more time passing, on average, between individual chapters. I think this happens with a lot of books and stories, and the passing of time and relationships changing over time is a common theme of life, but this story made me think about that in particular. Time is moving quickly. Like, the time it took to pass through 3 years of high school felt like forever, ages, eons. The time it took for me to get through 6 years of college felt enormous. In both of those time periods, I wasn't even remotely close to being the same person at the beginning than I was at the end. I grew by leaps and bounds. I changed dramatically and learned so much. But I look at me now, and the me from 3-4 years ago? Pretty similar. That scares me. Change scares me, but not changing scares me even more. Maybe I've changed more than I think though. It's weird to think that at 33 years old, I'm at a temporal twilight of my life. Surely not, right? I haven't even had kids yet! We might this year though, we will see.

2025 is the year of the Snake, they say. Wikipedia says it's a year of transformation, renewal, and spiritual growth. I'm paying attention to that transformation myself. I'd like to lose weight and build some muscle, I'd like to learn how to be a dad and do a good job with it, I'd like to make progress to reach the next level of my career. I'm working to renew my fighting spirit. I need to remember to be confident, even aggressive when I need to be. I no longer want to shy away from conflict. I want to practice keeping my focus on something, and having my mind wander less. I also want to be calmer, more resilient. I just want to be all-around better, and I want to do that by focusing on those individual issues. Times keep getting harder, and I want to rise to the occasion.

So, individual resolutions:

Get better at public speaking. I've been consistently bad at it. I want to be better.

Gain some muscle. Use those free weights I got. Actually use the Peloton I still have (and admittedly do use roughly twice a week).

Remember to be confident and focused. Keep presence of mind as much as I can.

Learn how to be a dad, or at least read some baby books.

Thanks for reading, future me. I bet you're on your way towards meeting those resolutions. Here's to you. I imagine you now in my mind's eye. I feel gratitude. I hope you feel like the past year wasn't a waste. I'll try to make sure it isn't.

[corvuscorona]

akuse! on github. honestly i only use it for older anime which i cant source copies of both online and offline. really curious about the sources they use bc i searched extensively for some anime online and couldnt find it for years and they just. HAVE it. they dont have any appimages just a linux guide on how to build it yourself with the terminal. there's not that many steps but i'm so scared of messing up

hi. I went and looked at this github project the second I saw this ask & immediately went "OH I SEE" because oh my god the technical writing skills on display in that readme file are . let's call the overall effect "confusing". let's describe this situation like "the dev could follow the trail of breadcrumbs they left in there & they figured that pretty much was good enough." many such cases unfortunately but I THINK I have decoded it now.

here is what you're gonna do:

1. where it says "code" on the project page, click on that & yoink the zip file. you do not need to learn git today, so don't worry about "cloning" it. just download it.

(that green fucker on the right.)

2. extract it. put the extracted folder wherever. I have a folder in my "home" directory called "Programz" where I put oddball stuff like this because it makes it easy to find later.

3. MAKE / LOG INTO AN ANILIST ACCOUNT, THEN go to the "developer settings" page they link in the readme file without telling you what it is (which redirects you unhelpfully to the homepage if you're not logged in. jesus christ)

4. I cannot see the form for this because I didn't try it myself, but you want to "create a new anilist API client." find whatever button does that. When you do that, there should be a "redirect URI" field. put "akuse://index,https://anilist.co/api/v2/oauth/pin" in it.

5. in the folder you extracted, go to the "src" subfolder and then the "modules" subfolder. make a new file in there called "clientData.js"

6. the readme explains this part okay. open clientData.js and put this stuff in it.

7. open your package manager. you are going to search for 3 different things, and if any of them aren't installed, you're going to install them and any dependencies the package manager says you should also install. search for: npm, libcrypt, and rpmbuild.

8. now I am going to tell you to open a terminal window. inside the modules folder*, right-click and there should be a menu item that says "open terminal here". do that. then, in the terminal, type "npm run dist:linux rpm" and hit enter. (*I THINK? if this doesnt work try the top level folder maybe. it won't explode if you do it wrong just try stuff)

Theoretically, this should generate some kind of Something that you can run like you would run akuse on your Windows machine! I didn't get a chance to step through this process, so it's possible I am fucking something up, but I hope these instructions are at least clearer than whatever's going on in that readme file. it's worth a shot!

If you try this and it breaks down along the way somewhere, you are so totally welcome to report back here & tell me about it if you feel like it. maybe I can help! maybe I can't help but having an outlet to go "hey what the fuck" at will be nice for you anyway. either way this is awesome enrichment for me thank u

(& also I would like to thank minecraft modders for being the worst ever at documentation, instructions, and all related topics, because had I not been tried in those fires I would not be able to read shit like this at all.)