Fun to think about, have personally gone down this rabbit hole before, but I'd be surprised if this was permitted by courts.
> The key point here is that the shape of the glyphs themselves, for example, non-trademarked text posted on advertisements or products with printed text, are not copyrightable. So long as you’re not stealing the creative work, like the advertisement, itself, the shape of each letter is in the public domain.
There is a jump from this assertion to:
> This got me thinking, as all things do, on whether I could simply scrape the internet for public, non-creative, non-trademarked use of fonts
The missing step here is that the glyphs for a webpage are not embedded in some sort of rendered image or physical medium. They're _rendered from the copyrighted font file_ to the browser.
The font file is delivered to the browser and that copyrighted file is used to render the glyphs.
This pipeline might work for creating a font that is "inspired by" the available glyphs rendered in the credits of your favorite film.
But downloading the entire copyrighted font file, generating a character sheet from it and clipping the individual characters from that sheet? I can't imagine how any sensible court system would not consider that a derivative work.
> But downloading the entire copyrighted font file, generating a character sheet from it and clipping the individual characters from that sheet
That's not what this project is doing. It's extracting the characters from already published pictures, without touching the original font files at all.
I believe he's downloading png files from Meta, converting them to SVG, then creating the glyphs. I'm not sure what the law is on this, but there's no need for the font files to ever be on your computer. E.g., https://render.myfonts.net/fonts/font_rend.php?id=a6719376b9...
Yeah this is similar to how facts are not copyrightable, but data is. You can't just extract rows out of a data file one-by-one, into a variable named "just_a_fact", put it back into a new file and pretend that you didn't derive it from the original.
Courts are not dumb. Trying to scale up a technicality to circumvent the law usually does not fly.
If this were a legal issue it can be gotten around the same way the IBM BIOS copyright was circumvented: with a clean room reimplementation. Team A renders the text with the copyrighted font files; Team B, otherwise not in contact with Team A, scrapes Team A's text images, vectorizes them, and imports them into FontForge.
Yeah, you're using the Internet as Team A in the author's example. You're looking for images of fonts. The people that made those images bought a license to use the font to make images, so it's fine if they share them with you.
None of that guarantees you'll be free from legal concerns, of course. Maybe the original image author didn't have a license. Maybe the licensor thinks they can change the terms of the contract at will. You'll probably be caught in the middle of all of this.
My take is that the biggest problem you'll find is naming your free fonts. Someone probably owns the word "Helvetica", so you'll have to call it something else. Someone looking for a font will then not find yours, and will buy a Helvetica license.
Back in the day (Windows 3.x era) I used to use cheap lookalike fonts from shareware compilation disks/CDs. So there was like a Palatino clone called "Palamino", a Cooper Black clone called "Cookie", an Optima clone called "Proxima", etc.
You can still find fonts like this on font sites all over the web.
The Atari ST used to come with a font named "Swiss". It was basically Helvetica and was named that because "Helvetica" is Latin for "Swiss" (as in "Confoederatio helvetica").
It may be considered a derivative work, but that doesn’t matter if it’s a derivative of only the portion of the original work (the glyph shapes) that cannot be copyrighted as a matter of law.
> The key point here is that the shape of the glyphs themselves, for example, non-trademarked text posted on advertisements or products with printed text, are not copyrightable. So long as you’re not stealing the creative work, like the advertisement, itself, the shape of each letter is in the public domain.
There is a jump from this assertion to:
> This got me thinking, as all things do, on whether I could simply scrape the internet for public, non-creative, non-trademarked use of fonts
The missing step here is that the glyphs for a webpage are not embedded in some sort of rendered image or physical medium. They're _rendered from the copyrighted font file_ to the browser.
The font file is delivered to the browser and that copyrighted file is used to render the glyphs.
This pipeline might work for creating a font that is "inspired by" the available glyphs rendered in the credits of your favorite film.
But downloading the entire copyrighted font file, generating a character sheet from it and clipping the individual characters from that sheet? I can't imagine how any sensible court system would not consider that a derivative work.