So im beginning development into making this something really vast and then maybe make a game around this…
or interact in whatever way
the story is a little confusing without context.
Seven Minute Shadow starts at 17:53.
At 18:00, Kiyoshima will be bombed.
You are just a postman trying to finish one last route before it happens.
You cannot stop the bombing. 
You can only decide how much gets delivered before the seven minutes are over.
That is basically the whole point of the game.
Not saving the city.
Not changing history.
Just trying to finish something before there is no time left. and the end scene after the game over or you win scene is the same throughout. it is set in the future with the kiyoshima bombing archived, however many minigames you complete is the number of the letters that survived, and the letters that you can read.
guys. just to remind you before you rate this work. the best experience of the site is ona good pc with hardware acceleration *on in your browser. for mobile and lower end computers it is very basic. it will give out the same info but the design is very basic. and the design is the main thing abt it. it is very basic on phones etc. as seen below. ofc the effort has still been put in to make it look good with a video and other features.
I ended up reworking a lot of the game, and it’s definitely better than where it started.
It is not the exact idea I came into this with. The current version is still pretty crude, but it’s now a proper playable game with 4 minigames, win and game over scenes, and an actual ending.
This is still my first Godot game, so I’m going to keep it around as a benchmark for myself and see how much better I get over time.
There is a storyline to it too.
Do check it out guys and tell me how you liked it.
I think heith.me is done.
Everything I wanted to implement is in now. The full site, the effects, the different device cases, the lite mode, the mobile mode, the interactions, the performance fallbacks, all of it.
I’ve also finished the design process stuff I wanted to document, so the site isn’t just showing the final result. There is actually an explanation behind why I chose certain colours, fonts, layouts and effects.
I wanted to build the whole thing as one process instead of constantly adding random features forever.
So I’m stopping here.
I don’t really plan on committing changes later down the line unless something is actually broken or I absolutely need to change something. I wanted this to be a finished piece, not another project that keeps getting new features until the original idea is gone.
The only thing left is getting the domain.
I’m waiting for the website domain grant to go through so I can finally buy the domain I designed this site around.
So yeah.
here guys.
https://aamosm.github.io/heith.me/
..the attached image is the portfolio page. also a note i wanted to add.. : this website has taken much longer than what is timed. the actual pen paper designs. getting together failed projects in other folders. designing in figma etc. was not tracked. This happens a lot with the projects i do in missions but somehow it affected a chunk of this one too. also i just realised this project has been listed as hardware???? its software. and i cant change it now. ill have to get help from support
The site has changed quite a bit since the first version.
I’ve changed a lot of the actual interface and started pushing the visual direction much further.
I also added the custom cursor from xenobiota. I really liked how something that small could change the whole feel of the site, so I wanted to try it here too.
I’ve also been taking a lot of inspiration from Awwwards sites. One thing I’ve noticed is that the best sites aren’t just crazy because they have crazy effects. The really good ones make those effects work properly across devices.
That became a pretty big focus for me.
I added separate cases for different hardware.
If a device has hardware acceleration, it can use the full experience. If it doesn’t, the site falls back to a lighter version so the effects don’t completely destroy performance.
There’s also a separate lite mode and a mobile mode. I don’t want the site to only look good on the machine I built it on.
I’ve also been making the interactions a bit more obvious, especially around muting and unmuting.
There are still a lot of little things like this that I’m catching as I actually use the site instead of just looking at it.
I’m also working on something I want to add directly to the site: a breakdown of the design process.
I want to explain why certain colours, fonts, layouts and effects were chosen instead of just putting the finished site in front of someone and leaving them to guess.
The site is the portfolio piece, but the process behind making it should be part of it too.
I went into this with a pretty clear idea of what I wanted, I just didn’t expect the actual design to stay the same.
The main references were:
I liked how all of them had their own identity instead of looking like a normal portfolio template. That was the main thing I wanted here too.
Visually I was going for a Wes Anderson-ish kind of thing. Not copying it, more the controlled layouts, odd colour combinations and slightly theatrical feeling.
The idea behind heith.me is also kinda recursive. The site itself is one of my portfolio pieces, and I’m going to be breaking it down in my portfolio too.
I knew I wanted the opening to feel more like entering something than just seeing a hero section.
I went through multiple versions of the opening before landing on the torn-paper idea. The paper tear gave me a way to make the video feel like it was underneath the page instead of just sitting in a box.
I changed the tear quite a few times too, especially the shape and colours.
The video is an old render of mine, and one of my first animation attempts. I didn’t really want to make a completely new video just for the site, so I reused it and messed with the presentation instead.
I chopped it up, added the scanline treatment and made it more chaotic. It felt way more fitting than just putting the original render there normally.
I also originally had the text arranged diagonally.
It looked cool, but once everything was actually on the page it started fighting with the person in the centre, so I straightened it out.
A lot of the site has basically been this. I have an idea, build it, realise it doesn’t fit as well as I thought, change it.
The colours went through a lot of versions too. I have not even made a proper repo for this on github. or hosted it yet. The plan is to get the website to how i want it to look and finish it off in one commit. and ofc future changes will be committed, but i want something to show first. The entire thing is just pure html, css, and js. no frameworks.
This update was mostly about making Obscure nicer to actually use.
I added a custom search engine dropdown that matches the current theme, along with controls for the background box and text length. The dropdown also got a few fixes after I found some problems with it not closing properly and text becoming hard to read.
The favicon now changes with the selected theme, so even the browser tab matches the page.
I also added local caching for the APOD response, so the same day’s image doesn’t need to be fetched again every time the page is opened.
I am probably going to work on my other projects unless I need to finish off tasks here. One glaringly obvious update I need to do is to update the weather API. Also the logging I made in the console straight up doxxes the api keys.
This update was mostly about making Obscure feel more personal.
The appearance settings are now fully wired up. Things like custom accent colours, blur, greeting, clock and theme switching all work properly instead of just being placeholders.
A lot of the work went into fixing issues that only showed up after actually using the page.
I fixed theme colours, improved native control contrast, stopped APOD from randomly changing when switching image quality, fixed stale descriptions after failed requests and handled video APOD entries properly by falling back to their thumbnail.
There were also a few small tweaks to make theme switching behave more naturally, like clearing custom accents when selecting a new theme.
The new tab page now has a working search interface with support for multiple search engines. I’ll be adding more over time, especially community-made ones once that part of the project is ready.
I also added NASA’s Astronomy Picture of the Day to the homepage.
At first I used the high-resolution images, but they took noticeably longer to load. I switched to the standard image instead, which makes the page feel much more responsive without losing much quality.
There were a few setup issues along the way too.
I fixed a GitHub Actions dependency problem and cleaned up a few things so deployments are a bit smoother.
I started by following the guide, but as usual I ended up taking it in my own direction.
For now, the project is a custom new tab page, but the long-term goal is much bigger. I want Obscure to support search engines created by other people, whether that’s YouTubers, fellow Stardance hackers or students.
The idea came from FaceDev’s video where he built his own internet protocol. It wasn’t meant to replace the web, but to create a much smaller network where people could host their own websites. I want to bring a similar idea to Obscure by making it easy to discover and search through community-made search engines.
This project also gave me a chance to work with GitHub’s deployment workflow instead of the hosting setup I normally use.
While setting everything up, I used a placeholder value for the API while wiring things together. Before adding the real key, I realised I’d forgotten to ignore the .env file. I fixed the .gitignore immediately, verified nothing sensitive had been committed, and only then added the actual credentials.
It was a small mistake, but a good reminder to check the repository before introducing any secrets.
The Alien Registry is finally here.
VoyagerOS can now browse user-submitted species as if they were discoveries made by Voyager-9. Alongside the 4 billion procedurally generated species from This Alien Does Not Exist, anyone can now contribute their own interpretations to the registry.
I want to start building a community registry.
I’d love to see how different people imagine the same species and watch the registry grow over time, and try to forgive my bad drawing.
Guide for submitting your .alr : https://github.com/aamosm/VoyagerOS/blob/main/alr/README.md
Project README:
https://github.com/aamosm/VoyagerOS/blob/main/README.md
The biggest addition in this update is the Alien Registry.
VoyagerOS can now browse, search and view procedurally generated alien species from This Alien Does Not Exist as if they were real discoveries stored aboard Voyager-9. Each entry includes its own observations, classification and registry information while remaining tied to its original seed.
The Paint app also allows exporting a custom file (basically a zip with a diff extension) as custom Aliens. I want them to be able to be uploaded to the repo and work like the dinosaur repo of hackclub.
This is the first step towards connecting the two projects into the same universe.
Instead of existing separately, VoyagerOS now has a purpose beyond being an operating system. It acts as the place where alien records can be documented, searched and explored.
Also made themes better.
The registry is only the beginning.
The next step is expanding it with better search, filtering, observations and more tools for exploring discovered species. also allowing user inputted aliens alongside the already 4.something billion ones.
This update focused on making VoyagerOS feel like it has a history instead of just an interface.
I introduced the Compact, a shared registry that keeps track of delegates, records, archived documents and historical events. Rather than placing lore in random text files, everything now lives inside the virtual filesystem and can be explored through normal terminal commands.
One problem I ran into was keeping certain files up to date.
Things like registry records and submission logs shouldn’t be static, so I changed the file viewer and terminal to generate them when they’re opened. Commands like read, grep, stat and the file viewer all use the same underlying data, keeping everything consistent regardless of how it’s accessed.
The shell also gained a few new commands including whois, delegates, charter, access and submit.
Instead of being standalone features, they’re connected directly to the registry and filesystem, making the terminal feel more like a real administration tool rather than just a command prompt.
One thing I’ve noticed while building VoyagerOS is that adding content is becoming just as important as adding features.
The very next thing I want to do is add my project “This Alien Does Not Exist” which has procedurally generated aliens, to make a registry. It could be like, aliens recorded till now.
