#tech #my-site

as part of my ongoing project to set up a home lab, i’ve started self-hosting yamtrack, a media tracker that can track all sorts of media – video games, books, movies, manga, etc. up until now i’ve been using various online services like anilist for anime/manga and infinitebacklog for video games for basic tracking, plus offline tools like obsidian and anytype for noting down my personal thoughts and organizing things i’ve watched by language. it’s never really bothered me, using multiple services for all the types of media i like, but after finally getting everything imported, i gotta say… it’s pretty nice being able to just see everything i’m watching/reading/playing at the moment, all in one place, and being able to create lists including all forms of media. also, after the ordeal of actually getting everything imported, including from services that don’t have an official export option (looking at you, mydramalist) i can safely say i’m never going back to a service that doesn’t offer simple csv exports lol. not even json – i love json, but i shouldn’t have to brush up on my jq skills just to see the start and end date of when i watched something. thankfully yamtrack does, so once i was done i exported my data just to see what the csv file looked like… and after filtering out the things that were marked to watch/read/play, i still had nearly 600 entries 😐 i really don’t feel like i consume THAT much media, especially since the last two or three years a lot of the time i used to spend watching anime or gaming i now spend watching thai and japanese youtube (which i only track as time spent watching, not number of videos watched) or sometimes tv/movies/games (which takes much longer to finish than in english)… guess i was wrong 😅 and here i was thinking that i was going to use it to build a page on my site with my complete media log, but once i actually scripted a bare bones version of the page, it was just… too much. once i really thought about it, i realized i get more value out of reading smaller, curated lists of media people love or want to share than just a huge list of everything they’ve ever laid their eyes upon, especially if there’s no reviews or ratings system. so i think i’ll rework the code a bit and post just the subset of my favorite things… eventually….

a lot of the services i’ve set up in my home lab use third party APIs to fetch metadata about media on the server, including yamtrack. for books, it uses hardcover, so i’ve been trying to migrate my own data on storygraph to hardcover too to find what’s missing in hardcover’s database, filling out the missing data myself and hopefully making the data yamtrack pulls more accurate. but i swear, it’s like hardcover actively doesn’t want me to help make its database more accurate. more than half the time i try to add a new edition to an already existing book, when i submit the isbn of the alternate edition it says it’s auto-fetching the information for it, just to go back to the original edition -_- does hardcover have beef with people who listen to audiobooks or something…

even though i’ve been trying to move as much of my data as i can to my new server, i’m realizing now just how hard it is to break free from Big Tech’s grasp. for example, even though i’ve mainly relied on sites like storygraph and worldcat to get book info, amazon-owned goodreads is still the only place i’ve been able to reliably get edition information on thai translations of books. so even if i keep all my book notes in plain markdown on my computer or try to contribute to an independent database like hardcover, i’ll inevitably rely on some tech giant’s database or infrastructure. now i’m not a FOSS+decentralized-or-bust kinda girlie – while i’m probably more privacy conscious than most people (based off the confused looks i get from most people irl when i mention trying to avoid using services by companies like google or meta when i can), i’m not above using these services when i feel like i get more out of it than i risk or when i have no other alternative. but sometimes it makes me wonder what’s the point in spending so much time and mental bandwidth on trying to minimize my reliance on them if i already know i’m not going to go 100% of the way in removing them from my life. then there are times i feel like i’ve learned so much in the process of prioritizing software not owned by the Big Five the last 2-3 years – from learning basic programming to more easily organize and process local text and media files, to now trying to run my own server teaching me about networking and system administration – that, even if its futile and google already knows my social security number or someone at apple’s already made deepfake nudes of me or whatever, the effort i’ve made hasn’t been entirely for nothing.

i added a new page on #my-site last week for my favorite video game locations. i've had this idea for awhile now, and i knew i wanted animated backgrounds, but i didn't know how to best implement them. after fiddling with turning my recordings into gif, webp, and mp4 files i learned about the avif format and omg the file size... it's like a tenth of the size of the gif versions i made and the quality is way better too. i was able to make the gif on my languages page 5mb (it was originally... 50mb....... 🙈) and the haruka gif on my resources site homepage only 552 kb! makes me wonder what the hell i was doing spending all that time on tumblr making gifsets while fighting for my life to make each gif less than 10mb 😭 i still have more places i want to add, but i need to find more free cam/no hud mods for some games.

i also overhauled my resources site to be generated with 11ty. it's now completely navigable without javascript enabled, even on mobile 🥳 i'm gonna keep tweaking my personal site to be functional without javascript. i saw some examples of pure css sidebar toggle on codepen that i want to try to implement. i doubt anyone who might view this site on mobile would also not have javascript enabled (tbh it's hard to think that anyone views this site at all), but i think it's a fun challenge. i feel like i learn more about what's possible with just html and css, and it gives me new ideas for my site.

i made a bunch of changes to #my-site this past week, mainly stuff behind the scenes. most notably i added an atom feed for my microblog. i checked it with the browser extension feedbro and it looks okay i think. i also made one for updates to the rest of my site, which you can access here. i enabled rss autodiscovery on the main pages as well to make adding it to your rss reader a bit easier. in addition to moving to neocities, as part of my ongoing effort to curb my #social-media use i'm also trying to read more blogs, essays, and other long-form written media by subscribing to substacks and rss feeds, and i figured i may as well make it possible for others to do the same with my site. (as a side note - i recently discovered the read it later app omnivore and it's been really helpful in actually finishing reading articles i find interesting. i know other read it later apps have existed for a long time, idk why i've been sleeping on them until now.) i restructured a bunch of stuff in the code too, mostly switching some divs for semantic tags and converting all jquery to vanilla javascript, so if something looks broken try hard refreshing (ctrl+f5). turns out theres so much more to web development than the random bits i hacked together from tutorials and stack overflow answers making tumblr themes in 2015... i'm not sure if my site is actually any more accessible or loads faster and i still need to test things with an actual screen reader, but i think my code is easier to read at least.

i think i've had some sort of ear infection for the last couple days. i was gonna go to the doctor today if i wasn't feeling any better when i woke up, but it doesn't hurt anymore so i didn't end up going. i don't know what's up with my body this past year, i usually only get sick once or twice a year but i got sick like three times this winter alone -_- i took advantage of all the time i spent this weekend lying on my side to catch up on some manga (i may or may not have been like 6 months behind on chainsaw man 🙈).

i also #played some piofiore and managed to finish yang's and dante's routes. now yang's route definitely lived up to its notoriety in the otome community but i found his best ending a little underwhelming. i usually do bad and good endings before the best ending and i found those really thrilling (i was even a little moved at the good ending ngl... may have even shed a tear or two) but the best ending... idk. i didn't dislike it but it wasn't as fun as his other endings or nicola's best ending. i found dante a little boring too. i think he kinda got shafted by having to carry the plot of the game on his route lol. i was grateful to finally learn about the key maiden stuff but honestly i spent the first half wishing i could've romanced leo instead. i'm halfway done with this game now and am looking forward to gilbert's route especially. orlock isn't really the type of LI i usually like but i'm also really intrigued by his role in the plot especially after seeing... yang's? best ending epilogue and dante's best ending.