Copyrighting font "source code"

Cory Maylett

This subject has come up before, but a quick search indicates that it's been a while.

I built a font for a client who hired an attorney to look into copyrighting the typeface, which, of course, isn't doable in the United States. My client and his attorney are unfamiliar with some of the terminology related to fonts, so they asked me to get involved.

To make a long story short, it's possible to register a copyright on computer source code in the U.S. if the code meets the following criteria from the copyright office. This makes it possible to theoretically copyright the code in the font file even though the typeface isn't copyrighted.

However, the U.S. federal copyright office says they will only register the copyright on source code, which they define as follows:

Source code: Source code is a set of statements and instructions written by a human being using a particular programming language, such as Java, LISP, LOGO, PASCAL, Programming Inquiry Learning or Teaching, Programming in Logic, Assembly Language, or other programming languages. Typically, these statements are comprehensible to a person who is familiar with the relevant programming language, but they are not comprehensible to a computer or other electronic device. In order to convey these statements and instructions to a machine, the source code must be converted into object code.

With the copyright office's definition in mind, they won't accept OTF or TTF files because they're not source code files. This TypeDrawers thread ( https://typedrawers.com/discussion/743/us-copyright-application-filing-for-font-software ) indicates some of you have had success by submitting .ttx (XML) files, so I supplied my client with a .ttx file. The attorney says the copyright office rejected the .ttx file's XML as not being source code. Presumably, they would say the same about a UFO file.

In addition, the definition of source code specifies that the code must be written by "a human being," which rules out any code written by the font creation software. Arguably, values entered by hand and simply transcribed into the source code by the program could be copyrightable, such as hand-entered sidebearing and kerning pair values. However, as I mentioned, they're rejecting XML files that show those values because XML is marked-up source code, not the original source code.

The attorney thinks they might accept the original Glyphs or Fontlab files as containing source code. But I know of no way to peek inside those files to see what might be considered source code.

The arguments for squeezing through the U.S. copyright office loopholes seem to be getting thin and convoluted. Even so, my client wants to try.

Can anyone shed any additional light on this subject, and does anyone know if it's possible to extract what might be considered raw source code from a Glyphs or Fontlab file?

John Hudson

Oh, the whole US intellectual law on software as literary works is stupid. It is a decision the US Copyright Office made decades ago, when instead of trying to understand the new creative technology and its particular implications they decided to apply an existing paradigm of copyright law—literature—to which they are now desparately clinging as it becomes ever more irrelevant for many kinds of software, not just fonts.

The attorney thinks they might accept the original Glyphs or Fontlab files as containing source code. But I know of no way to peek inside those files to see what might be considered source code.

A .glyphs file is a plain text format, so you can open it in any text editor and see contents.

A .vfc file is a binary format, but FontLab 7 can also write a .vfj which is a JSON format source.

JSON is, of course, an example of a notation that is intentionally easy both for humans to read and write and for machines to parse and generate, so in itself illustrates the false assumptions of the USCO’s distinction between source code and object code.

John Hudson

PS. It may be worthwhile informing your client’s lawyers about US Design Patent protection for typeface designs. This is limited protection of the design itself for 15 years, and like all patents comes with an obligation to defend, i.e. if someone copies the design the client will be obliged to take steps to counter the infringement or lose the patent protection. I have never registered a US Design Patent myself, but some of my clients have for typefaces that I have designed for them.

Cory Maylett

John Hudson said:

A .glyphs file is a plain text format, so you can open it in any text editor and see contents.

A .vfc file is a binary format, but FontLab 7 can also write a .vfj which is a JSON format source.

Thank you, John. I could swear I tried opening up the .glyphs file in a code editor and it didn't work. I just tried again using BBEdit, and you're right, of course; it worked. Whether or not the copyright office will accept it as meeting their definition of source code is another issue.

Copyrights worked fine in the pre-digital era. Today, though, the subject is often very squishy, full of loopholes, and awkward workarounds. Copyrighting digital code as a literary work is a good example of one of the weird contortions. The laws and regulations should be overhauled for the 21st Century, but I don't see that happening anytime soon.

I mentioned a design patent to my client and his attorney. The attorney was already familiar with them, but they're considerably more time-consuming and expensive than registering a copyright. All things considered, it might be their best option.

Thank you again.

JoyceKetterer

@Cory Maylett this is something of a quagmire.  

@Thomas Phinney did a good article that explains the history of this fairly well.  Long story short, it's probably not worth the trouble to apply for a copyright in the US.  Here's the article.

If the lawyer wants to speak to someone so as to ask questions, I can probably connect them to the lawyer who was interviewed for this piece.  She's our former attorney, retired, so I'm not certain she'll be up to it.  But if not I'd be willing to speak with them and fill them in on the rest of the history as I remember it.  My memory isn't going to be as good as hers but it's not nothing.  If you need more just pm me.

Thomas Phinney

Feel free to DM me if you are interested in more detail or even private advice. The article was very space-limited....

This week sucks, though. I have a major font deadline on Thursday, and three active cases. You’ll get way more info from me next week. 

JoyceKetterer

@Cory Maylett speaking from  a non-legal, purely business strategy, perspective the only thing I think is worth the trouble is a trademark.  Last I checked, design patents are only for 15 years and you have to enforce every slight.  The resource drain of enforcing everything doesn't, for me, justify the short period of "protection".  Similarly, last I checked copyright requires you enforce in the jurisdiction where the violation happened.  That's exhausting just to contemplate.

A client like yours probably really only cares about the name of the font, which they likely have given some iteration of their own corporate name.  A registered Trademark will allow them to make dmca takedown requests of pirate copies, and to stop other companies from using the fonts.  Strategically, does anything else matter?

Cory Maylett

JoyceKetterer said:

A client like yours probably really only cares about the name of the font, which they likely have given some iteration of their own corporate name.

Thanks for all the great advice and the link to @Thomas Phinney's article. I'm forwarding the link to my client. Thank you, Thomas, as well.

Confidentiality concerns keep me from going into detail, but this client is a business that designs and manufactures various physical consumer products of an artistic nature. The various uses of their custom typeface on their products play a primary role in the products' design and saleability. The glyphs in the typeface are not part of a corporate logo or wordmark. Instead, their use is part of the products themselves.

They and their attorney can decide, but perhaps a design patent might be the most appropriate option to pursue in this case.

John Hudson

Confidentiality concerns keep me from going into detail, but this client is a business that designs and manufactures various physical consumer products of an artistic nature. The various uses of their custom typeface on their products play a primary role in the products' design and saleability.

In this case, the products themselves could perhaps be protected by trademark. So while protecting the custom typeface per se may be difficult, defending against someone producing similar or knockoff products should be easier.

John Butler

Heh, trademark every individual character.

Florian Pircher

®™

JoyceKetterer

@Cory Maylett I'm glad I could help.  I get that this is all very overwhelming for a client getting up to speed on the font world.

I would caution your client, and their lawyer, not to bring non-font assumptions to the font world.  I have a long list of seemingly reasonable requests I've gotten from very smart lawyers who don't understand fonts, requests I had to talk them out of.  Preliminarily, this feels like it falls into that category. Fonts are weird edge cases in law, software and design.  

To be blunt, does the client have the wherewithal to even know if someone releases a font that would infringe a design patent?  It's a high bar.  First you need to know a lot about font design.  Second you need to be paying close attention to everything that's released.  I can tell you that I don't think my team would be up to it.  We have the technical and design skill, but we spend too much time in our own sandbox and too little time watching the world go by.  And it's not like there are people you can hire to do it, those sorts of legal researchers can't recognise fonts to save their life. This is a question for a lawyer, but my understanding is that you really do have to enforce everything.  That sounds like a huge commitment to me, that I decided not to undertake.

jeremy tribby

I'm just spitballing here, but I wonder if USCO would be more amenable to "source code" that is, for example, a python-based build script, that just happens to have an entire .glyphs file inlined inside it?