i spend a lot of time on random side projects. sometimes they evolve into things that are almost ready to share with the world (though i have yet to "finish" or "release" a project), like Pomelo, or MelonFit (my fitness tracker, named after my dog), or MelonHabits (two guesses what that does).
other times, they're just little scripts or toys for myself - like the smb rescanner, or a script i wrote recently to help manage all the services i run on my home server.
but sometimes, they just don't work out.
i really like rss. i follow a bunch of feeds - blogs, magazines, news, whatever. i love checking NetNewsWire in the morning and skimming through new album reviews, essays, art, comics...ever since i uninstalled Instagram, it's the main place i scroll.
rss readers almost all sort stories chronologically. lots of people hail this as one of the best parts of rss - no algorithms, just a chronological feed. that's fine, but personally, i'd rather have a good variety of stories to scroll through (feeeed does a great job of this, but there's no desktop client, and using the iPad client on desktop is not great, and it doesn't integrate with miniflux).
so i figured, what if i built a service that would connect to my miniflux instance, and look just like a miniflux server, except that when you went to pull the entries, I could sort them however I wanted? that way, i could keep all my feed subscriptions in miniflux, and i could use whatever rss reader i wanted and see my feed sorted however i wanted.
and honestly, it basically worked - it connected to miniflux, remixed the feeds into one feed that i could sort however i wanted - but the old Google Reader API that NNN uses to connect to miniflux is so convoluted, and rss readers cache so aggressively (for good reason) that testing and iterating was no fun, and eventually i got to a point where it was not quite close enough to working to be satisfactory (i couldn't get all the items reliably, i couldn't mark things read reliably, i couldn't format stuff the way i wanted...) and i wasn't having any fun working on it anymore. so i shut it down, and went back to scrolling :)
making the call to scrap the project (well, not literally - it's still on my drive. just in case i pick it up again.) sucked. i spent like 6 hours on it over the past couple days, and calling it quits made that feel like wasted time - especially because i don't feel like i learned a ton from working on it - it was mostly just frustrating. in some ways, i wish i had called it quits earlier - but i guess i already didn't do that.
this whole thing reminds me of something that happened when i was in highschool - i decided i wanted to find the perfect web browser. Chrome didn't exist yet. i had probably been using Firefox for a long time (in 8th grade my backpack was from the Firefox merch store), but i had recently gotten my first MacBook, and Safari was pretty cool too...and then I found Shiira (rest in peace, coolest logo ever), and then over the course of probably two weeks, i tried out every browser i could get my hands on - not out of curiosity, but out of this driving need to find the perfect browser, like a compulsion. eventually, weeks later, i kept using Firefox. i guess, what i can say about this rss side quest is that at least it didn't take weeks for me to come back around to
really exploring the classics lately.