I never thought I'd get to see if a font could be fair use due to the parody exemption - but I'd suspect there would be a good chance that this is the case.
In the US, typefaces themselves are not subject to copyright law (though that doesn’t mean ‘font piracy’ is legal - the ‘computer code’ backing a font is still copyrightable - but you could presumably extract the outlines and replicate a font exactly without running afoul of U.S. copyright law.) (disclaimer: IANAL)
A rasterized (imaged) font is not copyright-able. However the vector control points are considered a creative work and therefore subject to copyright. This was established in Adobe v. Southern Software:
What a bizarre ruling -- like saying that a parabola is not copyrightable, but three points defining a parabola are (because one could express the same parabola using many combinations of three points).
Surely, then, one could write a program to generate a copyright-free copy of a font by selecting different control points that yield the same curves. Or would the computer be the copyright-holder, as it selected those points?
> Surely, then, one could write a program to generate a copyright-free copy of a font by selecting different control points that yield the same curves. Or would the computer be the copyright-holder, as it selected those points?
The new three points would be derived from the old ones; hence it's still a derived work, and still falls under the old copyright. Your bits have the wrong color.[1]
Only if it's directly derived from the TTF file itself.
But if you rasterized the font (and as a specific representation of the front, it no longer has a copyright), and then had a computer generate curves from that, you'd be fine.
But from my understanding the hard part is creating the file that does a good job at every different font size. Creating a font for just 1 size is relatively easy.
So your computer would be doing the easy part, not the valuable part.
I have a hard time imagining a judge taking a sympathetic ear to this defense. Do you happen to know if there's relevant existing case law on this subject that you could point me to?
I don't have case law, but there are tons of fonts that explicitly say they are based on a copyrighted one:
* libre baskerville
* Liberation Mono is a sans-serif font metrically equivalent to Courier New
* Liberation Serif is metrically equivalent to Times New Roman.
"Metrically equivalent" just means they take up the same amount of space when typeset (i.e. if you take a document in Times New Roman and change the font to Liberation Serif, nothing should move around). As you can see in the case of Courier New v. Liberation Mono, the two fonts can be very different visually while still being metrically equivalent.
If you were to regenerate those "parabola points" by looking at non-copyrighted material, you are in the clear, even if you end up at the same values.
So to public domain a font, all you do is display it on one screen, then recreate it on another. Because the appearance of font itself is not copyrighted.
A TTC is a TrueType Collection. Multiple fonts in one file. Instead of having separate files for every size, weight, etc. they’re combined into one. The major advantage is the ability to share glyphs between fonts instead of duplicating information in each file.
As it happens, I looked at the link to see if it needed with '.ttf' and had it done so I would have skipped it. But as it did not I thought "hmm, wonder what a tf is."
Right, but the author made his intentions and wishes clear, and I feel like publicly drawing attention to him in this manner disrespects those intentions and wishes.
This is the Internet and it's bound to happen. If not me, someone else would have, and probably already, pointed it out.
I'm also not sure what you are referring to with "the author made his intentions and wishes clear", but think my pointing out of the git history has nothing to do with it.
Thinking my comment is in any way going to influence the fate of this repository seems a bit far fetched.
I don't interpret it this way. All the author has said is that he didn't want the hassle of fighting the take-down, not that he particularly wished for the take-down to proceed or that he thought it was correct (legally or otherwise).
Depending on the wording of the take-down notice, not having removed the history as well as the latest easily accessible versions could have been an error on his part that Monotype may now ask him to rectify (hopefully nicely and politely, if they do), or as he is active on this thread he may do anyway.
The tone of his comment above doesn't suggest he is bothered by people drawing attention to the matter.
I don't blame tholman for complying with the takedown, but my layman's understanding is that Monotype had no legal grounds for their complaint because parody is protected -- they were just using legal language to intimidate him into complying. Does that jibe with others' understanding of the law?
You are correct, and I reached out to the author about fighting their bogus claims. He wasn’t interested, but this was a while ago. Perhaps if this new parody font fights them, he’d be willing to do the same for smelvetica
IP law with fonts is kind of unintuitive. From what I’ve read, the design of a font can’t be copyrighted, but the implementation can. So you can make a font that looks identical to Helvetica and be okay, but if you copy and paste anything from the original, you’re on shaky ground. Kind of like how recipes can’t be copyrighted, but the wording and formatting of a recipe can. IANAL of course, but that’s what I’ve read in the past.
Parody doesn't apply to software. You might have noticed Monotype is always using the phrase "font software" to underline the fact that fonts are considered software, not a design, and thereby enjoy stronger protection than non-software copyrighted works.
Honestly, as a very clear parody, I don't see Monotype having a right to claim there are 'copyright issues at hand here'. In my eyes, this falls completely under fair use.
That is, UNLESS the exact glyphs were used, with only spacing altered - then I can see it being a fair point.
Yeah, like a lot of people mentioned here, I didn’t really fight the takedown. Didn’t really have a lot of interest wasting any time with people who’s sole job was to ruin the fun. And on the flipped side, found it even funnier that they felt it hurt their brand at all.
I'm a bit disappointed that it doesn't include an "rn" = "m" kerning mistake... my favorite and one that is already kind of an issue in Helvetica (and also can happen in the word "kern" itself, ie "kern" = "kem")
I one time saw a sign with intentionally bad kerning in the design department at work that said something like, "This sign makes designers very angry."
I was admiring the joke, when the the VP of design saw me smiling at it, and said, "Thank you! You have no idea how many people look at that sign and just don't get it."
Kerning is the amount of spacing between letters. Type with poorly applied kerning is a long standing joke as a way to annoy a designer. Font files can have the amount of kerning between letter pairs built in. This font is Helvetica with very poor kerning built in. It's "hell" for people who care about kerning.
63 comments
[ 3.0 ms ] story [ 140 ms ] threadhttps://en.wikipedia.org/wiki/Adobe_Systems,_Inc._v._Souther....
So even stripping out the hinting and OTF shaping "computer code" would still presumably leave you in violation of a font copyright.
Surely, then, one could write a program to generate a copyright-free copy of a font by selecting different control points that yield the same curves. Or would the computer be the copyright-holder, as it selected those points?
The new three points would be derived from the old ones; hence it's still a derived work, and still falls under the old copyright. Your bits have the wrong color.[1]
1: https://ansuz.sooke.bc.ca/entry/23
But if you rasterized the font (and as a specific representation of the front, it no longer has a copyright), and then had a computer generate curves from that, you'd be fine.
But from my understanding the hard part is creating the file that does a good job at every different font size. Creating a font for just 1 size is relatively easy.
So your computer would be doing the easy part, not the valuable part.
* libre baskerville * Liberation Mono is a sans-serif font metrically equivalent to Courier New * Liberation Serif is metrically equivalent to Times New Roman.
See: https://en.wikipedia.org/wiki/Liberation_fonts
Arial is based on Helvetica and we created specifically because of copyright on Helvetica.
If you were to regenerate those "parabola points" by looking at non-copyrighted material, you are in the clear, even if you end up at the same values.
So to public domain a font, all you do is display it on one screen, then recreate it on another. Because the appearance of font itself is not copyrighted.
That means you can take a vector font (copyrighted) and convert that to a rasterized font (non-copyrighted).
If you take that non-copyrighted rasterized font and produce a vector font from it, I presume that product falls under the original copyright?
Images are also copyrighted, but no one is adding DRM to them.
Even with .OTF, IE only is willing to use them if they have their "evil bit" set in a particular way.
Font companies are desperate to convince everyone they have more copyright than they actually do.
> If you take that non-copyrighted rasterized font and produce a vector font from it, I presume that product falls under the original copyright?
No, the new product (font file) is copyrighted to you.
I'm just sad that this doesn't kern r and n towards each other, to produce the classic misread "keming".
Or am I being Mandela'd.
[0] https://github.com/tholman/smelvetica/commit/83eadad7edf1878...
I'm also not sure what you are referring to with "the author made his intentions and wishes clear", but think my pointing out of the git history has nothing to do with it.
Thinking my comment is in any way going to influence the fate of this repository seems a bit far fetched.
I don't interpret it this way. All the author has said is that he didn't want the hassle of fighting the take-down, not that he particularly wished for the take-down to proceed or that he thought it was correct (legally or otherwise).
Depending on the wording of the take-down notice, not having removed the history as well as the latest easily accessible versions could have been an error on his part that Monotype may now ask him to rectify (hopefully nicely and politely, if they do), or as he is active on this thread he may do anyway.
The tone of his comment above doesn't suggest he is bothered by people drawing attention to the matter.
I don't blame tholman for complying with the takedown, but my layman's understanding is that Monotype had no legal grounds for their complaint because parody is protected -- they were just using legal language to intimidate him into complying. Does that jibe with others' understanding of the law?
That is, UNLESS the exact glyphs were used, with only spacing altered - then I can see it being a fair point.
Yeah, like a lot of people mentioned here, I didn’t really fight the takedown. Didn’t really have a lot of interest wasting any time with people who’s sole job was to ruin the fun. And on the flipped side, found it even funnier that they felt it hurt their brand at all.
Wondering if anyone is aware of any others.
(think this only works via proper google.com)
I was admiring the joke, when the the VP of design saw me smiling at it, and said, "Thank you! You have no idea how many people look at that sign and just don't get it."