STATE OF THE FIXT - YEAR 4

so, as you may have noticed, the website looks a little different now. happy birthday, HYPERFIXT, the blog!

before i dive into all the juicy details behind the fresh new look, i want to get the usual housekeeping out of the way and reflect on the past year, as it pertains to me and my strange corner of the information superhighway. this time last year, i was talking a big game about trying to shake off some self-imposed notions of how i approached my writing, and did it pay off? eh, at least a little. this wasn't necessarily a banner year for HYPERFIXT, but i do think i'm coming out of it with at least a marginally healthier relationship to my craft. when i have written, i feel like i've put less pressure on myself to be comprehensive to the point of overindulgence. i've been willing to go down weird rabbit holes with less self-conscious remorse, or been timelier about my work, even when it meant having to rework sections of an article on a tight deadline. obviously, as always, i'd like to be able to bring more to the table - the dream is being able to do this consistently enough that i can at least sort of call it a job - but even when i haven't quite figured that out, i feel like i'm getting better at fighting that impulse to be my own worst critic.

aside from what you see here on this URL, i've also been up to all sorts of other stuff. over on Bluesky, i've continued trying to be at least a little more social than i was when i first started this place, and i've found ways to make microblogging work for me. i'm not going to pretend to understand where the popular concept of a 'media thread' came from, or try to claim with any authority that it's even that popular, but i know enough people who have one that i've worked it into my routine as well. it's a fun little way to journal about what i've been up to lately, and i think it's nice to be able to take note of things that are good, even if i can't turn around and weave them into something interesting to read about.

if i'm talking about stuff i wound up doing off-site over the course of the last year, it'd probably be negligent in some weird way to not acknowledge that, hey, me and some of my buddies thought of a dumb joke and it's now objectively the biggest footprint i've ever left on the internet. i don't have much interest in trying to leapfrog off the attention that side gig has drawn in for a few reasons; it'd be a betrayal of my commitment to the bit, i don't currently have much interest in HYPERFIXT blowing up to that extent, and most of all, it'd be wildly dishonest of me to take much credit for the project when it's long-time friend of the site gizmouse who's done the heavy lifting. she's the one who put together the pipeline that turns the titles into an accurate audiovisual title card experience, which is most of the reason the whole thing works in the first place, and she's contributed far more titles to keep up that daily schedule than i have. all i've done is chip in with a couple of goofs, and all i've gotten out of it is a legitimate physical sense of dizziness knowing that real people i really admire have seen and liked a joke i wrote.

taking another step back entirely, into the terrifying world known as 'offline', it's really been a year of getting my shit together. did you know that your 20s are a weird time? because your 20s are a weird time. i've had to straddle a lot of lines between taking the time to figure myself out and having to dive into the deep end and improvise as i go, but i'm grateful that i still don't have to rely on this blog financially yet, and i'm proud of myself for breaking down some particularly tough walls. 2026 is looking to be full of its own new types of weirdness to figure out, but i feel like there's tangible ways that 2025 left me more well-rounded and ready to face whatever comes my way.

now, normally this would be where i start taking questions, but, uh, fun story; Neosprings shut down like, two weeks after i announced i'd be using that for taking questions year-round. bummer! i also don't especially feel like collecting the questions manually from my friends, partially because i know i've got something else big to talk about. you're in my world now, suckers!

i have known for a hot minute that the blog needed a redesign. i mean, it came up last time i was looking back on another year of running this place, but the thought definitely predates that, and it super definitely only got stronger since then. as i write this, it's actually a little over two months from when this article and this redesign go live together, and i'm still actively working on all the code. writing this now is simultaneously so i can capture the ins-and-outs as i experience them, and to give myself something to do when i've been knee-deep in JavaScript for half a week and need to flex a different set of creative muscles. so, where better to start deep diving on the new and improved HYPERFIXT than by talking about why the old one needed the tune-up so badly?

i am not a web designer by trade. i'm very much a hobbyist. i've been using computers for as long as i can remember, and i'd picked up some of the basics here and there along the way, but HYPERFIXT is the first time i've ever had a place that's truly just for me, and more importantly, wholly by me. it was born out of conversations with my friends about how much we loathed the shape of the internet and social media, and how nice it'd be to have our own spaces, even if the thought of learning how to code seemed daunting. all this is to say that when i started this webpage, i didn't come in with a plan or any sort of experience, and basically everything was hodge-podged together as i went.

now, i'd never disparage the noble art of hodge-podging. i really think more people ought to do at least some version of what i did; having your own website is work, but it can be so rewarding, and it seems like every year, having your own self-made corner of the internet feels smarter and smarter. i'll never turn my back on those core Web 1.0 sensibilities that drove me to make this place, no matter how much fancier the outward changes might seem.

what i personally struggled with, and what i'd share as a sort of cautionary tale to anyone else interested in the prospect of doing this, is that i didn't hodge-podge the right way. a lot of the site's visual assets were first drafts thrown together thoughtlessly, and a lot of the core features were developed in isolation, one at a time as they became necessary. i could create a website, sure, but i wasn't internalizing how features worked once i found a quick and dirty way to slap them onto the existing framework. i had the code, but i couldn't always pinpoint why any particular function worked the way it did, and that lack of deeper understanding behind why things did or didn't work only compounded and made it harder to have confidence in doing new things with the site.

so, as soon as AGDQ had wrapped up, i decided this was gonna be it. new stylesheet, new code, and most importantly, make sure i properly know what the hell i'm doing, so that hopefully, should i ever want to mix things up a little, i have a better understanding of the core components that i'll be rooting around through. all those little features i'd brainstormed over the years that'd drifted off into 'eh, eventually' territory? well, now's the time to crack those. i promised myself that no matter what, i need to have at least the basics of an improved user experience ready by the site's anniversary, which works out to be about a three-month runway from when i set myself to work on this.

being the dork that i am, with a love of presentational wrappers and categorization, i decided to give this whole project a name. the new design of HYPERFIXT that you're browsing right now is called 'Liquid Website', and i'd love to walk you through some of its biggest changes.

NEW COAT OF POLISH

i'd saying picking a new font was step one, but no, it's more like step zero. maybe even digging deep and hitting step negative one, if you account for the fact that i threw together messy little buttons to deal with this issue years ago.

i am a sucker for a stylish font, and one of the most commonly cited free-to-use font repositories on the internet right now is Google Fonts. it was a real 'kid in a candy store' moment for me, and, as was the case for a lot of things on this website, i committed wholeheartedly to the first cool thing that caught my eye. good ol' Audiowide isn't a bad font by any means, but it's definitely not built to be the main typeface you're looking at, especially for the length of posts i'm writing. there's a time and a place to break out the fancy stuff.

across all my friends and everyone who's ever taken a chance on reading one of my essays, the number one piece of feedback i've always heard is that the font was a real hurdle. adding in some buttons to automatically override everything and switch the page over to a sensible default sans-serif worked okay enough, but it was also a band-aid on something that really needed a cast to mend it into proper shape. very luckily, when the very first bits of what would become a full-scale redesign were starting to bounce around in my head, the fine folks over at Chequered Ink were running a wildly generous offer, selling their 1000+ font library for only nineteen dollars. you actually might've seen the first exploration i was doing with these fonts if you caught last year's retrospective; i did a lot of testing, but i was immediately drawn to Freudian Slit for logos and other big eye-catching display elements, with its bold blocky characters given subtle tweaks that almost read as curvature until you look closer and see how sharp the angles are.

like Audiowide before it, though, it's not built for comfort while reading. i tested a couple of Chequered Ink's fonts and went back-and-forth on what would be the new typeface of choice for the actual body of my writing, and with a bit of input from some friends, i wound up picking Brain Wants. it's stylish enough to still give the site some personality, but hopefully, it proves to be a lot more readable when you're trying to parse through my work, especially with some more proper calibration applied to the way the page renders and spaces out text.

the other big new visual element implemented as part of Liquid Website is the new background. the original HYPERFIXT background was, quite literally, just an image of some bubbles run through a bunch of filters; its actual filename on this very website is still 'TestBackground2', which i think paints a pretty clear picture of me messing around with some settings and calling it a day very quickly. for the redesign, i wanted to stay true to what i liked about my site's aesthetics, but also hone in on everything with a bit more intentionality. i also wanted to keep in mind that, hey, it's probably good to have a high-resolution source file on my end that i can then downsample as needed for different use cases.

the Liquid Website background is a composite of some elements i put together using Blender, and believe it or not, Blender is not the hard part of that! it's really just some basic sphere objects, some warped and blended together by basic automatic functions, placed inside a cube, with a plane pushed down like a net to form the grid pattern. some of the textures mind seem a little complicated, but thankfully, folks who work with Blender a whole lot more such as ninebaal taught me that it's actually fairly easy to procedurally generate material properties from noise using Blender's node systems. if i were going for something more specific, i might need to do a little more research, but the melty, wobbly, Ultra Q-style texture i got out of the process actually works perfectly for the vibe i'm going for.

the real tricky part came in compositing the image together. the bubbles required a different lighting pass than the rest of the elements, which was easy enough to render out, but i almost stumbled back into the exact same pitfall from 2022. i used gradient maps to bring the overall color tones back in-line with the actual hexidecimal values i use on this website after the light had affected them, and i'd created a version of the image i was really happy with... only to save it as a flattened and compressed .png before closing everything down for the night. realizing i needed to go back in and get a high-resolution image with layers i could adjust separately, i spent multiple days trying to recreate whatever effects i'd stumbled into, but nothing seemed to draw out the same peaks and valleys of hue and luminosity. ultimately, i decided the effects i were pursuing were probably a little noisy for a background anyways, and that if i kept throwing myself into a brick wall over this, i'd never sit down and actually focus on implementing better code.

i've made a few other changes - the wide, loosely Evangelion-inspired scrolling text at the top of each article has now been compressed into one component of a header that has more useful information on it, for example - but i hope nothing feels too unfamiliar. i think Liquid Website captures a lot of the things i liked about the original HYPERFIXT design, while also moving it closer to the types of things i tried to evoke but couldn't quite pull off. as i've said before, some of my biggest inspirations for how this website ought to look are the video game menus i grew up with, and now i have actual three-dimensional objects i rendered out to try and heighten that effect. everything is, hopefully, a little more sleek and readable. way back in the very first yearly retrospective, i said i wanted the aesthetic tone of this blog to be 'Web 1.5', mixing old-school simple formatting with a bit of modern polish, and i feel way closer to that stated intent than ever before.

UNDER THE MICROSCOPE

one reason i named this new redesign Liquid Website is because that's just sort of the aesthetic i was going for; the background's full of mysterious floating bubbles of Website Goo, and the logo's meant to evoke something like a test tube or an aquarium, its subtle glassy finish holding back its fluid contents. like any good name, though, you can read a little more into it! on an underlying level, this redesign uses a lot more modern HTML and CSS staples, with the specific goal in mind of flowing to fill whatever size container you're putting it in. almost like some kind of... Liquid Website.

one thing i put on the very rarely used about page when i first started throwing things together was meant to be something of a catch-all warning for when things broke; "HYPERFIXT was designed on a laptop with an unusual 2560 x 1600 screen size". four years later and i'm still on that same laptop, and i still don't know why Lenovo even makes them like these. it's a 'gamer' laptop! who's out here developing games and going "yeah, 16:10, that's the aspect ratio we want people to see our game in"?! are you supposed to run your games in 16:9 and then just do fuckin' whatever with the extra pixels?

point being, i'm working with a bit of a weird display. pair that with a very rudimentary understanding of CSS, using the most simplistic and barebones functions the standard has to offer for placing and spacing elements, and you had a blog that looked okay on my end, and could sort of squish itself into shape for other people's monitors, but often felt a little cluttered and messy. i don't think it ever outright broke down and became unreadable for anyone reading this blog on a desktop, but things were janky and less than ideal.

a lot has changed since i designed those elements, though. i hopped from Google Chrome to Firefox, which, at least as far as i'm aware, has a much more robust set of tools for web developers. i can test things at different resolutions, and when things break, i can get pointed straight to MDN, which is a much more reliable source of information than what i was using previously.

when it comes to making websites that scale well for whatever platform they find themselves on, things like Flexbox and Grid layouts are the modern standard. as much as i loathe the way big platforms have twisted these tools of web design for their own stupid and/or nefarious ends, they're standards for a reason, and they work really well. raw percentage-based rendering works alright, but using Flex and Grid gives me a lot more control over the spacing between things, or how much wiggle room they have to change scale based on the viewport. there'll always be edge cases and limitations, but i feel a lot more confident that the site should have consistent aesthetics and consistent usability.

while i'm on the subject of the underlying code that powers Liquid Website, i also want to say, as a personal point of pride and conviction, that all of this was done by real people. i toiled away in the Neocities editor, and when i didn't understand something, i asked for advice from friends who might, or referred to guides written by actual web developers. AI is the biggest crock of shit in the world. no matter how many times i felt like ripping my hair out when i didn't understand a bit of HTML formatting, no matter how many things i still have to learn, i'm not afraid to take pride in the fact that HYPERFIXT is not built on 'vibe coding', and that i made the website look the way i wanted it to with my own two hands.

HYPERFIXT IN YOUR POCKET

this right here is kind of the big one. my white whale of web design. it's a big part of why i had to tear up the whole flow of the blog right down to the bones, so that i could build it back up with this in mind. HYPERFIXT now has proper, intentional support for mobile devices and tablets.

you always could open this website on your phone. i mean, of course you could. it's a website. who's gonna stop you? but with such a busy font and the sidebars struggling to figure out what to do in a tall aspect ratio, it was an uphill battle.

on some level, i rationalized this to myself as a bit of Web 1.0 authenticity. if my website's meant to look and feel vintage, then yeah, it's gonna look a little janky on mobile, and that's just the way the cookie crumbles. really, though, it never did sit quite right with me. i'm all for old-school web design, but i also know that we now live in a world where the ways people access the internet are more diverse and detached from the desktop than ever before. i know this, because i'm part of that! even little ol' me, who still insists on running a blog in 2026, sometimes willingly opts to go lay down and scroll on my phone rather than sit at my full dual-monitor setup. i'm not trying to pull a heel-turn into the type of modern bullshit that drove me to make this blog out of spite, but i knew i could do better in a way that was additive to the HYPERFIXT package rather than transformative.

with my old layout, mobile support always felt daunting. i'm sure if i sat down and chipped away at it, some of the things i've done for the Liquid Website redesign could be applied to the old design, but it certainly wasn't built to do that, and as i've been saying, a big part of doing this was to teach myself better fundamentals and have hands-on experience remaking the components so that i'm confident i know how they work going forward.

thanks to Flexbox standards living up to their name and being, well, flexible, getting the website to render in a parsed-down, narrower format for phones and tablets was fairly easy to implement, albeit with a few hiccups. i am nothing if not a perfectionist, though, and there was a site feature i knew i had to figure out a way to include. it's something small, something that's almost definitely more trouble than it's worth, but i do consider it to be part of this blog's whole deal, weird as it may sound.

when you're viewing HYPERFIXT on your desktop, go ahead and hover your cursor over the images. there's a bit of text there in the tooltip! i like to hide little extra jokes or side tangents there, using HTML's "title" tags. it's very much inspired by the tradition of webcomics like xkcd, and believe it or not, sometimes it is literally the hardest part of the posts to write.

notice the key issue here when it comes to mobile support; the idea of "hovering your cursor" over stuff doesn't really exist. there's occasionally some way of parsing this type of content - i learned while coding my own personal workaround that, at some point since i last checked, iOS made that stuff display if you hold your thumb down on an image long enough - but MDN still cautions that it's not universal, and it seems like the best practice is to assume "title" text will be inaccessible on mobile. more importantly than official solutions or the lack thereof, though, i had a vision for how i wanted to pull this off my own way years ago, and i decided now was the time to give it a proper go.

i don't think it'd be accurate to say i learned JavaScript, because that would imply i now describe myself as a person who knows JavaScript, and i very much wouldn't. i understand a couple of basic things about JavaScript, and i was able to wrap my head around how to execute on some very specific ideas with the knowledge i had. there's still a couple of bits of syntax i don't quite understand, where all i really know is that the way MDN has laid out the base concepts works and i just have to have faith in it to keep working. it's something i'd still like to understand a little better, but also, at the same time, when i got this feature up and running, i did feel maybe an unhealthy amount of adrenaline rushing through my system as i took a victory lap around my home office.

now, whenever you tap on an image, it'll pull up an overlay that displays the image at a larger size and puts the "title" text underneath as a caption. it seems simple, and literally as i type this out, a couple of optimizations i could make are dawning on me, but i'm extraordinarily proud of this nonetheless. it's a feature i've wanted to do for ages, it's the type of thing where i knew very clearly what i wanted the visual end result to look like but had no knowledge of how to actually make it happen, and i taught myself how to make it real over the course of a couple of days.

silly as it may seem, stuff like this was a serious bottleneck to developing mobile support for HYPERFIXT. if the situation was already standing at 'you can read it on your phone, but it's a sub-optimal experience', then why would i want to settle for still leaving things out? sure, i never put anything crucial in the tooltips so that the actual body of my work stands on its own straightforward merits, but i'm here to write stuff, and those extra bits are still words i wrote. they're a part of the stuff i intentionally put into this site for people to enjoy, and to me, mobile support ought to mean that all that stuff has to work reasonably well whether you're at your desk or on your phone.

of course, there is a little bit of that bright-eyed naivety about the reality of the modern internet behind all this. i write big, long posts in an era where the more conventional choice would probably be a podcast or a video essay. the use cases for being able to comfortably read a 20,000 word essay on your phone are limited, at best. at the same time, though, if i'm going to commit to doing things my own weird way, that means i can choose to pursue things like this out of the hope that it'll improve someone's experience. i thought mobile support would be handy and that the site's better off having it, so i made it happen, and i feel really good about that. all i can do now is put it out into the world; worst-case scenario, i still did something i'm proud of, and best-case scenario, it makes your time reading this blog better in some way.

i think i've probably exhausted anything especially interesting about the new design, so if you're still reading along down here, i appreciate it. it's probably pretty clear that this is something i've put a hell of a lot of thought into! even though i'm still writing this from the past, where i'm hammering out the foundations and preparing for how to roll this new code out to old pages, i'm already feeling hopeful that doing all this will make HYPERFIXT a better blog.

speaking of the rollout, that's kind of one of the few catches here; since i'm changing things about how the actual individual bodies of pages display information, shifting the back catalog over is a manual process. i'm going to have to make edge case decisions about styling elements i only ever dabbled in once, and while i'm at it, i'll probably be fixing up typos here and there. ideally, everything should be switched over in short order, and i'm not planning on thoroughly redoing the contents of any page just yet, although the thought of doing some 'director's cuts' or revisits to old topics has certainly crossed my mind as a future endeavor.

i feel like whenever i've written these yearly retrospectives in the past, there's always been this sort of internalized bitterness that i can't help but see bleeding through when i go back and reread them. it's always very "this wasn't the big year where i made everything work". don't get me wrong, i still don't think this has been that, but i do think this is the most i've had a coherent and constructive vision for what's ahead of me.

as i worked on this redesign, i found that, as it so often does, the YouTube algorithm seemingly arbitrarily decided to push me towards some random hobby i don't even participate in, and i wound up enabling it, because i like learning weird new stuff. for whatever reason, this time, it was woodworking. first and foremost, watching someone plane a piece of wood down to a smooth finish really does something for my frantic neurons, but i think i also kind of wound up feeling a bit of - and i cannot stress this enough, extremely detached, twenty degrees removed, "i've never done woodworking in my goddamn life" - kinship? like, hell yeah. making stuff for yourself and developing a deep understanding and appreciation for even the mundane elements of your craft. i can vibe with that. i really hope i'm not stealing woodworker valor by saying that.

when i try to picture things about myself that i perceive as flaws, i think one of the big ones is that i'm very picky about the conditions i do things in. i'm very good at telling myself i have to wait for the perfect circumstances to make progress in my life, and it usually leads to things piling up until it feels impossible to chip away at the backlog of things that i could be doing to get closer to those perfect circumstances. i could stand to believe in myself and my agency to make my own life better a lot more.

maybe that's a little overly personal and drifting back into that weird self-deprecatory mode i want to avoid, but i bring this all up because i'd like to think that rebuilding the site is a reflection of me getting better at that. the flawed framework of my blog took on a sort of 'chicken and egg' quality, where i didn't know where to start fixing things and felt demoralized about both my writing and my medium. this time, though? i was able to be decisive, set a deadline, and learn what i needed to know to make the changes i wanted. it may not be perfect circumstances, sure, but i hope that not only have i made this blog better for my readers, but that i've made it better for myself as well, and that having a clean slate like this can help me do more cool things here.

whatever that all ends up looking like, i appreciate everyone who's been along for the ride so far, and i hope you'll be back soon. every little bit of support - reading, responding, lending a hand when i want to learn how to get something done - it all means the world to me.