Font copyright law changes coming to U.S.

The USPTO is in the process of changing the the copyright protection of font software in the U.S.

Time to contact your senators and congress people.

Attached is a form letter for your use and a pdf of the letter the Martinez Group PLLC has sent to Senator Charles Schumer.

I am posting these with the permission of the Martinez Group
«1345

Comments

  • AbrahamLeeAbrahamLee Posts: 202
    Thanks for sharing this, James. How did you hear about it? This is very disturbing, but unfortunately seems like a very common misconception of almost any software tool nowadays:

    “Just point, click, and out comes a magically perfect [name any digital product]! See how easy it is? Anyone can do it, even little Timmy and your 150-year old grandmother!”

    What a shame that someone is so misguided they feel the need to make such a drastic change as this. They really don’t understand the implications of what this would do to folks who actually do this for a living.
  • Will this mean that any software created visually (UIBuilder in Xcode for example) will no longer be copyrightable as well?
  • Yikes. Thanks, James (and Frank M.).
  • It will be very interesting to find out how this rule change came about—and at who’s instigation. I hope the results of the FOIA request will be published here.

    American type designers are already at a disadvantage in regard to copyright protections and I believe that showing the contrast between U.S., EU, and Japanese regulations would be very useful in making the case to U.S. politicians. But even more important would be a clearer demonstration of what font design and production involves. Without it, the reader of the letters might still be left with the notion that it’s no more than “paint by numbers.” To get back to the regulations as they were could require going after something much more.

    I think there’s a larger, overriding logic to the argument of the authorship of each line of code than Frank presents: Regardless of of any automation of code lines generated by the font authoring applications, each line is the direct result of a creative and purposeful action taken by the designer, without which there is nothing. The application is, therefore, little more than a pencil, and to claim otherwise would be the equivalent of saying that no artwork can be original unless the artist makes her or his own pencils, brushes, or colors.

    For such letters to be effective, they must lay out a complete description of the work for which protection is sought. As a type designer, I understand what Frank is writing about, but if I imagine myself as an ignorant reader, it’s not easy to comprehend.
  • James MontalbanoJames Montalbano Posts: 861
    edited August 20
    This is the FOIA Request that the Martinez Group filed. They were contacted on Friday and asked to narrow the request.
  • Thomas PhinneyThomas Phinney Posts: 1,143
    I was corresponding with Frank Martinez (Martinez Group PLLC), and pointed out a couple of glitches in the original version of the letter to congress, so he shared an updated version to post here. See attached.
  • I’m sorry to say that Frank’s letter, though slightly improved, may miss its mark. You have to understand your audience. Members of the Senate and House are not people who type their own letters or design their own campaign literature. You have to make the case from the beginning, with the understanding that what’s obvious to us is not at all obvious to them. You have to make them smart if you want them to advocate for your position.

    • Here’s a possible beginning:  [from Frank, edited] “Recently, the U.S. Copyright Office has changed its 25+ year policy regarding the registration of typeface font software. This change of policy means that I have no method of legal protection for the software that I create and rely upon to make a living. As a type designer, [graphic designer, and visual artist] in the United States and your district, I am writing to you to appeal this decision to the Copyright Office on behalf of the American type industry.

    [New] "Every character you type or read comes from a font. These are not generated automatically, but are made by people—artisan type designers and code writers—who labor for hundreds of hours on each type design. There are well over 200,000 fonts for sale in the marketplace and Americans have been the leaders in this field. The scant copyright protection that had been granted by the Copyright Office for such work (far less than that granted by other countries) has just been entirely undermined by them. They have failed to explain why—or at who’s behest—this change has been made. The ruling betrays a fundamental ignorance of what fonts are and how they are made. Unless the ruling is reversed, an industry that generates well over $100 million in annual income and employs designers and technologists in your district may be destroyed, exposing it to theft that will leave it too unprofitable to be pursued. Theft of American designs and data will become rampant. It will also, albeit indirectly, undermine numerous university-level education programs. The industry will be forced to move to the EU and countries such as Japan, where this work enjoys a level of copyright protection greater than that ever granted in the United States.”

    • Show what font making looks like in three or four illustrations, including the human-made outline and its resulting code. Explain that the code generated is entirely instigated by human actions. Show several diverse examples of different fonts. Avoid even common technical terms, such as XML.

    If we want to win, we have to make our case with absolute clarity.
  • AbrahamLeeAbrahamLee Posts: 202
    I’m sorry to say that Frank’s letter, though slightly improved, may miss its mark. You have to understand your audience. Members of the Senate and House are not people who type their own letters or design their own campaign literature. You have to make the case from the beginning, with the understanding that what’s obvious to us is not at all obvious to them. You have to make them smart if you want them to advocate for your position.

    If we want to win, we have to make our case with absolute clarity.
    *Dumbed down* absolute clarity. Politicians have a lot on their mind (I think) and not a lot of time to devote to any one issue.

    I’m still wondering what/who instigated this change?
  • Good thoughts, all.

    I totally rewrote Frank’s helpful template—shortened it to one page, translated it into plain English (e.g., used phrases like “we’re designers, not computer programmers” and “font piracy”), tailored it to each of my lawmakers’ interests/concerns, emphasized the intellectual property angle, underscored how many hours it takes me to make a font, highlighted some of my work and where it appears (e.g., Smithsonian). Who knows if it’ll help, but my senators and congresswoman have heard from me on the subject.

    (The process brought me right back to the old TypeRight days of 20-plus years ago.)

    Thanks again for the post.
  • Nick ShinnNick Shinn Posts: 1,314
    edited August 22
    “we’re designers, not computer programmers”

    I don’t think that’s a very convincing argument.
    The problem is that making a visible glyph shape is not authoring code per se, it is authoring a glyph shape, which is not copyrightable in the US. 

    However, I do author a bit of code (in OpenType substitution features such as figure alternates and pseudo-randomization) for almost all my fonts, and am about to publish some fonts with generative features created by custom-written code.

    So, in order to preserve the copyright of that part of a font which a type designer does actually write code for, I don’t see how that’s practically possible unless the whole font is protected.


  • Dave CrosslandDave Crossland Posts: 1,012
    Will this mean that any software created visually (UIBuilder in Xcode for example) will no longer be copyrightable as well?
    What does that have to do with the change in policy? From what I read in this thread and attachments, it's the ttx xml that isn't accepted for registration.

     I'm confused about that bring a good submission format in the first place. What is better about the versus a document that shows the authored outlines?
  • Mark SimonsonMark Simonson Posts: 987
    edited August 23
    The copyright office has accepted .ttx files as "source code" for the font until now when registering digital fonts for copyright. It now seems to be saying that if the font designer didn't literally write this file, typing it in line by line in a text editor to create the font, it won't be accepted. Arguably, the font designer has written this file indirectly using graphical tools (font editors), but they seem to now think this is unacceptable. There are graphical tools for creating other kinds of software, such as the UIBuilder in Xcode. Presumably, they haven't changed their criteria for copyright protection on software made with Xcode, so it seems inconsistent at the very least.
  • Mark SimonsonMark Simonson Posts: 987
    edited August 23
    As for why ttx rather than a native font editor file, I don't know, but they apparently have not ever accepted those, perhaps because they are not human-readable as text, while a ttx file is human-readable as text. (Come to think of it, a .glyphs file is a native source file and is a text file which can be read directly in a text editor.)
  • André G. IsaakAndré G. Isaak Posts: 239
    edited August 23
    The copyright office has accepted .ttx files as "source code" for the font until now when registering digital fonts for copyright. It now seems to be saying that if the font designer didn't literally write this file, typing it in line by line in a text editor to create the font, it won't be accepted.

    How exactly are the people at the copyright office supposed to know that you didn’t actually write the .ttx code directly? I realize no sane designer would actually do this, but I doubt the people in the copyright office have ever actually designed a font and thus likely don’t know what approaches might be used.

    And if one were to submit source code for (e.g.) a program written in C, how do they know that it doesn't consist of decompiled code from a program which you created in some sort of “non-programmatic” environment? And even if we assume one did write the source code, the actual object code will, except in very rare cases, be created by the compiler rather than by the programmer. So does this mean that copyright applies only to the source code and not to the executable distributed to end users?

    Since copyright applies to the end product rather than the workflow itself, it seems rather perverse to apply criteria based on how the product was created rather than what the product actually is.
  • Kent LewKent Lew Posts: 800
    How exactly are the people at the copyright office supposed to know that you didn’t actually write the .ttx code directly?

    That is the issue. As I understand it, the Copyright Office is now requiring a sworn statement that the author did in fact actually write all of the submitted .ttx XML code directly.

    Anyone would be free to swear to such a statement whether true or not, but I don’t recommend it.

  • Kent LewKent Lew Posts: 800
    edited August 23
    (Come to think of it, a .glyphs file is a native source file and is a text file which can be read directly in a text editor.)
    A UFO is also a source file and a package of .plist and XML files which are pretty easily read in a text editor.
    (And, frankly, I find UFO a lot more human-readable than a .glyphs file. Not that it really matters. ;-)
  • So does this mean that copyright applies only to the source code and not to the executable distributed to end users? Since copyright applies to the end product rather than the workflow itself, it seems rather perverse to apply criteria based on how the product was created rather than what the product actually is.

    André, you’ve just hit upon the central problem. The answer to your question is “no.” In essence, the copyright protection for fonts that was, until now, offered by the U.S. Copyright Office was wedged into a legal loophole, and, as such, offers no direct protection to the actual product. American type designers bought into it, even though it has always been vulnerable; they believed it was all they could get.

    “Useful articles,” which is the category into which type fonts are classified, is explicitly denied copyright in the oldest iterations of U.S. copyright law and it has been held as sacrosanct. But that consideration came from a much simpler time, from the era of metal type, and it was upheld later in the Copyright Act of 1976 and the 1978 case Eltra Corp. v. Ringer. But things have changed greatly since the advent of OpenType. I would suggest that, in the least, it may now be possible to obtain copyright for certain aspects of a font and not others. This sort of mixed status has been described by the Copyright Office itself, in examples such as a floral design on silver flatware being copyrightable, while the underlying design of the flatware would not be. 

    As I mentioned previously, the key part of a new approach would have to entail very clear and understandable descriptions of type design and manufacturing in all their aspects. In addition, it would be useful and possibly persuasive to compare aspects of type design with other kinds of creative work that that do enjoy full copyright protection. Some categories of copyrightable work worth looking at to strengthen the argument are cartographic works (why they not considered “useful articles” I’ll never know), collages, technical and architectural drawings, and “computer artwork” of certain kinds. I believe that doing so would stand a chance of changing the copyright aspect of fonts in a major way.

    I would be happy to assist with such an effort.
  • While I agree with you, Joyce, I do so only to an extent. Contract law applies only to legitimately purchased fonts, as only the purchasers have entered into an agreement; it does little or nothing to protect against piracy.The absence of any kind of copyright registration removes the possibility of proof in many kinds of disputes or in the enforcement of violations. What once had legal status is now moved into the realm of the merely ethical (or unethical, as the case may be). One wonders how the cases against unauthorized commercial users would have been settled without registration of the data.

    It’s a big deal.
  • JoyceKettererJoyceKetterer Posts: 211
    edited August 23
    Scott  -

    I've managed more than 30 license enforcements- most very large. 

    We never once uttered the word copyright .  

    In my experience the sort of infrigers you refer to are much more likely to have too little licensing rather than none at all.  We've gotten legal fee reimbursement every time because it's in our contract.  I'm not  a lawyer but  would argue, based on observing the high profile cases you meantion, that it was much easier than it would have been had we taken them to court under copyright law.  

    But that is not realivant.  What matters is that we can argue that the loss of copyright protection would be very bad without having to resort to exaggeration.  
  • Thomas PhinneyThomas Phinney Posts: 1,143
    edited August 23
    Frank Martinez has been kind enough to share some correspondence between himself and the Copyright Office. I think this gives a better context to what the Office is now saying, in two refusals. I would summarize it as: (1) anything submitted in the form of XML code will never be considered a computer program by the Copyright Office; and (2) they claim that this code is “merely assigned coordinates” which then require some other program to render the typeface.

    From the rejection, and my conversations with Frank, I assume that each submission consisted of the first and last 25 pages of a ttx XML dump of the font. This has been a common approach to copyright submissions for fonts for a long time now.

    I could go into why I think both of the objections are unreasonable, but I think showing the letters is more valuable:

    - - - - -

    August 08, 2018

    THE MARTINEZ GROUP PLLC

    Attn: Frank Martinez, Esq.

    55 Poplar Street Suite 1-D

    Brooklyn, NY 11201

    United States

    Correspondence ID: [redacted]

    RE: [redacted] OMPUTER PROGRAM

    Dear Frank Martinez, Esq.:

    We received your application to register a claim in "computer program" for "[redacted] COMPUTER PROGRAM."

    As explained in sec. 3 l 3.3(D), Compendium of U.S. Copyright Office Practices, Third Edition, the Copyright Office cannot register a claim to copyright in typeface or mere variations of typographic ornamentation or lettering. The Office may, however, register computer programs that generate a typeface provided that they contain a sufficient amount of literary authorship.

    In this case, however, the XML code you have submitted is not a computer program for copyright purposes.

    The copyright statute defines a computer program as "a set of statements or instructions to be used directly or indirectly in a computer to bring about a certain result." By contrast, XML code is a type of markup language. Markup languages are designed for the processing, definition, and presentation of text. They do not instruct the computer to perform any operational functions. As a result, while XML code may under some circumstances be copyrightable,  it cannot be registered as a computer program.

    Since our practices stipulate that the Office may register a computer program that generates a typeface, a work consisting of markup language which creates a typeface does not meet this requirement. We must, therefore, refuse registration. In accordance with our practices, we will keep the application, copy, and non-refundable fee. This letter is for your information only; no response is necessary.

    Sincerely,

    [signature]

    C. DiFolco

    Copyright Examiner

    U.S. Copyright Office

    Enclosures: Reply Sheet


    - - - - -


    March 05, 2018

    THE MARTINEZ GROUP PLLC

    Attn: Frank Martinez, Esq.

    55 Poplar Street Suite 1-D

    Brooklyn, NY 11201

    United States

    Conespondence ID: [redacted]

    RE: Computer Program for a [redacted] Typeface Font

    Dear Frank Martinez, Esq.:

    This work does not meet the deposit requirements for a computer program and the material deposited represents less than the required minimum amount of original authorship, registration must be refused. The copyright statute defines a computer program as “a set of statements or instructions to be used directly or indirectly in a computer to bring about a certain result.” Our regulations indicate that the deposit requirement for registering a computer program is the first and last 25 pages of source code. In these cases the registration is assumed to cover any images created by the instructions in the source code. Based on our examination of the application and deposit material, it is clear that this work is merely assigned coordinates to a particular letterform using a third party program to render typeface or typefont from those coordinates.

    In any case in which the Register of Copyrights determines that, in accordance with the provisions of title 17, the material deposited does not constitute copyrightable subject matter or that the claim is invalid for any other reason, the Copyright Office shall refuse registration.

    Since the material submitted does not meet the deposit requirements and does not contain the minimum amount of authorship required for registration, we must refuse registration. In accordance with our practices, we are closing our file without action and retaining the non-refundable filing fee.

    This letter is for your information only; no response is necessary.

    Sincerely,

    [signature]

    M. Graddy

    Copyright Registration Specialist

    Registration Program

    U.S. Copyright Office

  • By contrast, XML code is a type of markup language. Markup languages are designed for the processing, definition, and presentation of text.

    That seems like a rather odd way of describing XML.
  • Thomas PhinneyThomas Phinney Posts: 1,143
    Looks like they took that definition from Webopedia, here: https://www.webopedia.com/TERM/M/markup_language.html

    Here is a rather contrasting definition from TechTerms.com: https://techterms.com/definition/markup_language

    And the Wikipedia definition: https://en.wikipedia.org/wiki/Markup_language
  • John HudsonJohn Hudson Posts: 1,490
    Joyce, contract law provides protection against illegitimate use of legitimate fonts. Copyright law offers protection against people making illegitimate derivative fonts and selling or giving them away. That was a major problem in the 1980s and 1990s — entire companies existed whose business model consisted of copying other people's fonts and selling them in cheap bundles —, and only really became less of an issue after the summary judgement in the Adobe vs SSi case (1998), which affirmed that taking existing fonts and mechanically manipulating them constituted copyright infringement. [Design patents were also involved in that case, but not put to the test because SSi settled after the summary judgement was issued.]

    A cynic might also suggest that the advent of Usenet and then torrent sharing of original fonts put paid to the business model of font cloners: why pay for a bundle of often clumsily made derivative fonts when you can get the original ones for free?

    In reference to the present situation, I am wondering how USCO's new position with regard to code relates to the other aspect of the Adobe vs. SSi summary judgement: that the positioning of outline points in a vector format font constituted a creative act that could be subject to copyright protection?
  • John -

    I'm not making an argument that copyright protection is superfluous.  I know my history. Instead, I'm making a nuanced point that might not be coming through on this medium. 

    I'll spell it out.  

    Again, I'm not an attorney, so grain of salt: My understanding is that copyright only protects the font (I.e. Software) not the typeface  (design) -  As you point out the southern software case is unusual so we'll leave that aside.  As I understand it, that means that a from scratch copy of a font is never protected in the United States.  Copyright only protects us if the "derivative work" is a derivative of the software  - meaning it was lazily made by modifying the original.  This is also a contract violation if you want it to be (and of course you do if you're virtually anyone but Hrant). 

    The history is a big part of why it's worth trying to preserve copyright protection for fonts.  It also appeals to my sense of fairness.  Copyright status grants enforcement tools that we should have access to in any fair system.  

    That said, if your attorney is standing in front of a judge explaining the intracacies of font software technology it means a lot went wrong along the way. Copyright is the last scrap of defense not the first line.  

    By definition that means it is NOT our only protection.  


  • John HudsonJohn Hudson Posts: 1,490
    edited August 24
    As I understand it, that means that a from scratch copy of a font is never protected in the United States.  Copyright only protects us if the "derivative work" is a derivative of the software  - meaning it was lazily made by modifying the original.  
    If someone copies a typeface by making a new font from scratch, e.g. by scanning the original and tracing new outlines, that is indeed outside the font software copyright protection. It could, however, be a violation of a design patent, but unlike copyright a design patent must be registered (copyright registration is recommended, but not required), and also comes with an obligation to defend, i.e. you lose the patent if you don't take legal action against violators. Design patents are only valid for 15 years, so after that there is no protection in the US against from-scratch font copies.
  • John-

    Understood.  And that is not under threat at the moment, right?

    Joyce 
  • In reference to the present situation, I am wondering how USCO's new position with regard to code relates to the other aspect of the Adobe vs. SSi summary judgement: that the positioning of outline points in a vector format font constituted a creative act that could be subject to copyright protection?
    We looked at and referenced this case somewhat recently, and it seems to be for now a glaring contradiction to these copyright law changes. Very interesting, I'll love to see how this all plays out.

    It is sad to have received rejection notifications from the Copyright Office of numerous fonts that we'd previously registered successfully. If this is part of the new Idiocracy, then I guess it makes sense, although it's frustrating nonetheless.
  • “we’re designers, not computer programmers”

    I don’t think that’s a very convincing argument.
    The problem is that making a visible glyph shape is not authoring code per se, it is authoring a glyph shape, which is not copyrightable in the US. 

    However, I do author a bit of code (in OpenType substitution features such as figure alternates and pseudo-randomization) for almost all my fonts, and am about to publish some fonts with generative features created by custom-written code.

    So, in order to preserve the copyright of that part of a font which a type designer does actually write code for, I don’t see how that’s practically possible unless the whole font is protected.


    Thanks, Nick, and I agree—the phrase was mainly one attempt to appeal to my layperson lawmakers. (I happen actually to be a minor coder also.) But having been party to a couple of IP actions—successful, thanks to Frank—I remain hopeful that a group effort might ameliorate our wacky U.S. laws where type protections are concerned.
Sign In or Register to comment.