I'll be transparent: I mainly want to know if this, the whole site, counts as slop at this point.[^2] It should, I guess, but I don't think it does. Maybe Likeface pushes it back over the fence in one direction or the other?
I backed into this font creation tool -- though I've been obsessed with making fonts for 30 years -- the mechanics of which have become ... pretty elaborate. It started as a skill with token-hungry subagents arguing over individual glyphs they'd built as SVGs; now it's largely about a few efficient Gemini image API calls and a lot of chopping and vector conversion.
The outputs cost money to generate, so I set up a simple credits system: the Generational Electronic Token (GET).[^3]
You get 1 GET every day you sign in (bookmark it!) -- click the Queue button or try to do something that costs me money and you'll find the Sign In With Google prompt. No subscription model or anything.
I believe the quality will get even better over time[^4] because I have a pretty decent training loop set up; you can kind of make out a progression via the reverse chronological sort on the accompanying typeface marketplace, http://glyfs.store. You can sell the fonts you make there and get paid in tokens if people buy them. I need to figure out classifiers, especially since a lot of the test fonts (the only fonts listed there as of now) have misleading names.
In the end, the site and this module likely appear to be "co-coded," and they probably ought to.
But I dunno, it still feels pretty "me" (and I am chock full of humanity). It also doesn't feel like I am pretending to be some web wizard or business genius, I guess that helps (?).
I suppose I believe it's complicated, useful (arguably), and commercially merciful enough to not come across as slop.
If I am mistaken, let me hear it. I know you will.[^5]
[^1]: An honest question; I am pretty sure it is not unless you count the smidge of gen AI involved. It neither seems to be anywhere near displacing commercial foundry workers or independent fonticians nor replacing paid typefaces nor doing anything legally risky. Maybe trying to copy a foundry-owned font from a phone photo of a book cover, which it will not actively try to reject? I mean, I GUESS that's evil, on the Internet, in 2026. Still doesn't feel "evil" as a thing for a product to offer IMO (maybe the opposite)!
[^2]: 0i0.app started as a thing I built to learn about all the stuff I could make with just GCP + a frontend. I came back to it, made it a little prettier, cut some of the fluffier stuff (still plenty of fluff in there, don't worry). Full Picture Story is worth a look, and free (to users, it's like 10 API calls, it's fine). Writes you a story based on the hero's journey and a very unusual list of archetypes and themes I entered 6 months ago.
[^3]: The other modules make API calls too, so I made most of them cost 1 GET too. They are worth checking out; Redoer in particular can do a ton of cool stuff, and it feels like a good value since each run, even with multiple options selected, only costs 1 GET. Revewer exports a SWIF file you can use to make a real rug (allegedly; unsure how exactly I can debunk this claim.)
[^4]: This also means tokens are unlikely to ever get cheaper, and they never expire ...
[^5]: Feedback about pricing structure, however constructive it may be, also welcomed. I kind of feel like with the free credits, it's kind of a good deal for users and a bad deal for me, and I'm cool with that for now. Obviously, it would be amazing if this became a meme font smash hit, but that just seems like an impossible dream. Also, it is crazy how so many people post sites that get decried as slop and NONE of them say "sorry if this is slop" or ask "is this slop?" Maybe I've seen one do the former. I hope I'm not about to find out why!
Is it evil in some way?[^1]
I'll be transparent: I mainly want to know if this, the whole site, counts as slop at this point.[^2] It should, I guess, but I don't think it does. Maybe Likeface pushes it back over the fence in one direction or the other?
I backed into this font creation tool -- though I've been obsessed with making fonts for 30 years -- the mechanics of which have become ... pretty elaborate. It started as a skill with token-hungry subagents arguing over individual glyphs they'd built as SVGs; now it's largely about a few efficient Gemini image API calls and a lot of chopping and vector conversion.
The outputs cost money to generate, so I set up a simple credits system: the Generational Electronic Token (GET).[^3]
You get 1 GET every day you sign in (bookmark it!) -- click the Queue button or try to do something that costs me money and you'll find the Sign In With Google prompt. No subscription model or anything.
I believe the quality will get even better over time[^4] because I have a pretty decent training loop set up; you can kind of make out a progression via the reverse chronological sort on the accompanying typeface marketplace, http://glyfs.store. You can sell the fonts you make there and get paid in tokens if people buy them. I need to figure out classifiers, especially since a lot of the test fonts (the only fonts listed there as of now) have misleading names.
In the end, the site and this module likely appear to be "co-coded," and they probably ought to.
But I dunno, it still feels pretty "me" (and I am chock full of humanity). It also doesn't feel like I am pretending to be some web wizard or business genius, I guess that helps (?).
I suppose I believe it's complicated, useful (arguably), and commercially merciful enough to not come across as slop.
If I am mistaken, let me hear it. I know you will.[^5]
{footnotes in subsequent reply}
[^1]: An honest question; I am pretty sure it is not unless you count the smidge of gen AI involved. It neither seems to be anywhere near displacing commercial foundry workers or independent fonticians nor replacing paid typefaces nor doing anything legally risky. Maybe trying to copy a foundry-owned font from a phone photo of a book cover, which it will not actively try to reject? I mean, I GUESS that's evil, on the Internet, in 2026. Still doesn't feel "evil" as a thing for a product to offer IMO (maybe the opposite)!
[^2]: 0i0.app started as a thing I built to learn about all the stuff I could make with just GCP + a frontend. I came back to it, made it a little prettier, cut some of the fluffier stuff (still plenty of fluff in there, don't worry). Full Picture Story is worth a look, and free (to users, it's like 10 API calls, it's fine). Writes you a story based on the hero's journey and a very unusual list of archetypes and themes I entered 6 months ago.
[^3]: The other modules make API calls too, so I made most of them cost 1 GET too. They are worth checking out; Redoer in particular can do a ton of cool stuff, and it feels like a good value since each run, even with multiple options selected, only costs 1 GET. Revewer exports a SWIF file you can use to make a real rug (allegedly; unsure how exactly I can debunk this claim.)
[^4]: This also means tokens are unlikely to ever get cheaper, and they never expire ...
[^5]: Feedback about pricing structure, however constructive it may be, also welcomed. I kind of feel like with the free credits, it's kind of a good deal for users and a bad deal for me, and I'm cool with that for now. Obviously, it would be amazing if this became a meme font smash hit, but that just seems like an impossible dream. Also, it is crazy how so many people post sites that get decried as slop and NONE of them say "sorry if this is slop" or ask "is this slop?" Maybe I've seen one do the former. I hope I'm not about to find out why!