Le Monde Journal vs. Spectral Lawsuit

Nadine ChahineNadine Chahine Posts: 52
edited February 2023 in Type Business
They're calling it the "font war" in France and this case will have big implications for our industry, particularly when it comes to the questions of original work and when one typeface is too close to another. Thought it interesting to share it here. Also big example of what the indemnity clause in all the custom type contracts means, and why business insurance is highly recommended.

A journalist who was at the public hearing tweeted about it:


Article (auto translation, fyi police = font):




«1

Comments

  • Interesting. Are these the fonts at issue?


  • PabloImpallariPabloImpallari Posts: 777
    edited February 2023
    I can see fleischman in both... more in S than in LMJ
    There must be something else going on among this 2 guys that we don't know...
    I suspect the winner will be the one with better lawyers and lobby power
  • I know that this is not the point, but I have always thought that Spectral is a bit overrated. 

    On topic: I wonder how this kind of argument could be made more objective. With lawyers and judges missing the typographic expertise to really see how similar or different two relatively plain typefaces are, I'd say we need a more objective measure. A kind of % similarity that goes beyond merely overlaying bezier nodes. Perhaps AI could have something interesting to say about this? @PabloImpallari
  • A kind of % similarity that goes beyond merely overlaying bezier nodes. Perhaps AI could have something interesting to say about this? @PabloImpallari
    I think it could. However, the Fleischman comment is on point. Many fonts will turn out to appear very similar, because they both copied something else. And we're in a very funny world if A and B both copying the old guys is legal, but A copying B copying the old guys is not.
  • Thomas PhinneyThomas Phinney Posts: 2,730
    edited February 2023
    Clarification: “police” is French for “font” but sometimes does not get translated correctly (“because it is already English,” I assume). Thus “police war” in the article is a mistranslation of “font war.”

    (UPDATE: oops, Nadine pointed that out already, missed it)
  • PabloImpallariPabloImpallari Posts: 777
    edited February 2023
    Perhaps AI could have something interesting to say about this? @PabloImpallari
    Its telling me that the small pointy serifs in LMJ are more related to calligraphy, as in Compugraphic's Athenaeum, Softmaker's C720-Deco, Chantelli Antiqua, Brother Industries's Maryland and so on.

    The more bulky seirfs in S are more related to typographic alphabets, similar to Clarendon if they get exaggerated to the extreme.

    The more consistent diagonal stress in S are more related to calligraphy too, since the shape is created by the flat nib (look at the stress in /c and /o and uppercase /O).

    While the more inconsistent stress treatment in LMS (Vertical stress in uppercase /O versus a more diagonal stress in lowercases /o versus full diagonal stress in /e) is more bounded to happen in type design where shapes are drawn instead of written. Whatever you want to call it a "mistake" or a "design decision" in order to increase letter shapes differentiation is up to you.

    The more you look into tiny details, the more different they are.
    The more you look in general, the more they both looks like times new roman.
  • PabloImpallariPabloImpallari Posts: 777
    edited February 2023
    Also big example of what the indemnity clause in all the custom type contracts means, and why business insurance is highly recommended.
    My Dad was a chief breast cancer surgeon, running the whole gynecological area of a huge hospital (he is now retired). There were a few problems once in a while at the hospital, and so my Mon always insisted to my dad "Please, buy malpractice insurance.. if something goes wrong we could lose everything we have". My dad always refused.

    I asked him once:
    - Dad, why you don't want to buy insurance?
    - Bad doctors needs insurance. I don't.

  • John SavardJohn Savard Posts: 1,088
    And we're in a very funny world if A and B both copying the old guys is legal, but A copying B copying the old guys is not.

    I cannot agree with that. We do indeed live in a world where that is exactly how the law works, and indeed, that is how the law should work.
    With an important caveat.
    The "old guys" are in the public domain. So of course "B copying the old guys" is legal.
    And so is "A copying the old guys".
    Now, how about "A copying B"? That's not necessarily illegal. It depends on what kind of copying is taking place. A can certainly be inspired by B when creating an original typeface that's somewhat similar to one created by B, whether as a completely original work, or as a copy of something done by the old guys.
    A copying B can be illegal, though, if A copies too much from B, or copies from B in a particularly slavish way, or copies more easily protected aspects of B - like actual splines from inside B's font.
    To describe the details of this is beyond my own competence, as I am not a lawyer, and furthermore the law on intellectual property rights in typefaces varies significantly between Europe and the United States.

    If B copies the old guys, what B contributed that was original in his copy is still entitled to protection - but that protection can't be used to protect anyone else from copying from the same old guys.
    That is the principle that lets this all make sense - and it is the principle without which we indeed would be living in a "funny world".
  • John HudsonJohn Hudson Posts: 2,955
    The 2 fonts are clearly different.
    My understanding is that the case is not about whether the two resulting tyepfaces are different; it is about how Spectral was arrived at and whether that process was derivative of Le Monde.
  • Nick ShinnNick Shinn Posts: 2,131
    The working files to which the defending lawyer refers should settle the matter.

    But props to JFP for going after Google. That company’s type business model may have put a cornucopia of free fonts at the disposal of the masses, but in so doing it has limited the financial opportunity of independent font producers, who are also part of the masses.

    There are other reasons to resist this behemoth.
  • PabloImpallariPabloImpallari Posts: 777
    edited February 2023
    but in so doing it has limited the financial opportunity of independent font producers
    Any way you can prove that? I thinks is 100% bullshit. The very same bullshit was said when Corel Draw included fonts on his Cd. AFAIK Google Fonts propelled webfont adoption to the roof, and in doing so also increased financial opportunity for independent font producers to sell webfont licences... and I have heard from many independent producers that they are making a lot of money from webfonts. Is there any realistic way anyone can probe one thing or the opposite? I'm not sure any of our both hypothesis can be confirmed... I guess they will stay in the realm of the hypothesis world forever...
  • Igor FreibergerIgor Freiberger Posts: 250
    edited February 2023
    It's a very interesting lawsuit because the fonts seem to be clearly different. A more detailed view shows they share a number of similarities:



    When you go to the actual contours, the structure of nodes and handles is also similar. Maybe too similar in some areas, like the terminal of lowercase a. But... can the contour structure be actually very different?



    It will be exciting to follow the results of this judicial dispute.
  • Both fonts seem to adhere to a tried and true style that doesn't leave too much wiggling room. They differ in the things that they can - serifs, the forms of certain glyphs that usually give a typeface it's distinctiveness. IMPO, one could just as "certainly" claim they copied Garamond. 
  • John SavardJohn Savard Posts: 1,088
    edited February 2023
    The working files to which the defending lawyer refers should settle the matter.

    But props to JFP for going after Google. That company’s type business model may have put a cornucopia of free fonts at the disposal of the masses, but in so doing it has limited the financial opportunity of independent font producers, who are also part of the masses.

    There are other reasons to resist this behemoth.

    I cannot agree with this sentiment at all. At least, not as I am understanding it; perhaps I am badly and completely misunderstanding what you intended to say.
    Even if Google is one of the corporate bad guys, if indeed Spectral does not infringe on the intellectual property in Le Monde Journal, but somehow those alleging this manage to get the courts to believe it... then, once this inherently dishonest technique is shown and made to work, it is going to be used by the bad guys against the good guys far more often than the other way around.
    The days of independent font producers will be numbered in such a situation.

  • When you go to the actual contours, the structure of nodes and handles is also similar. Maybe too similar in some areas, like the terminal of lowercase a. But... can the contour structure be actually very different?
    I see the resemblance, but I wouldn't say they're too similar. They have the same terminal design, but the overall form is markedly different, like how the curve-to-curve and curve-to-flat transitions in Spectral are much more abrupt.
  • John HudsonJohn Hudson Posts: 2,955
    Again, the case is not about whether the typefaces look similar.
  • PabloImpallariPabloImpallari Posts: 777
    edited February 2023
    Its about Ego  :p
  • John HudsonJohn Hudson Posts: 2,955
     :D 
  • How much of the contour placement can be attributed to automated contour cleanup processes in FontLab et al keeping nodes at extremes?

    I’m more intrigued by the linebreaking and metrics when the two samples are superimposed, assuming no kerning is turned on.


  • Igor FreibergerIgor Freiberger Posts: 250
    edited February 2023
    Again, the case is not about whether the typefaces look similar.
    But, to be derivative, Spectral needs to share similar structures with Le Monde Journal, no? Otherwise, how one could even consider some link between then?

    For example: the same number of nodes of the same kind with very similar relative positions in the terminal of a is far more important than the resulting look. The terminal indicates a possible derivative work, but the lawsuit needs many situations like this to prove something.

    My point is that the fonts are more similar than they look at a first sight.
  • Conversely, you can have matching glyphs with non-matching metrics. Here’s a pair I always had my suspicions about, which thanks to recent developments I can now compare via simple Myfonts preview:

    Berthold Bodoni Old Face (G.G. Lange, 1983) vs URW “Bodoni Old Fashion” (“URW Design Staff,” 2001)

    The glyph shapes line up exactly, but the metrics differ in the non-regular weights, such that you have to slide the layers around to see the glyphs match. Lange’s version has the best *†‡§.

    You can now buy both from Monotype on MyFonts! I wonder whether Monotype can do a tax write-off by suing themselves.

  • John HudsonJohn Hudson Posts: 2,955
    But, to be derivative, Spectral needs to share similar structures with Le Monde Journal, no? Otherwise, how one could even consider some link between then?
    Yes. My point is that the case is not going to be decided on a subjective opinion that the two typefaces ‘look similar’, whether to an expert or to an uninformed observer. If one looks at the two side-by-side, it is easy to see all the ways in which they are different in terms of proportion, spacing, treatment of many details, etc. If one starts looking closely at the shapes, though, perhaps certain patterns of relationship seem to be present? One can begin to see how it is possible to arrive at the one design starting from the other. Whether that is what happened is what the case is about.
  • Nick ShinnNick Shinn Posts: 2,131
    edited February 2023
    Pablo:
    Any way you can prove that? I thinks is 100% bullshit. 
    The idea is quite basic: if you give people something for free, they will be less inclined to pay somebody else for a comparable product. Hardly bullshit.

    Your “developing the market” theory is hogwash (a.k.a. bullshit), similar to the “we won’t pay you for your graphic design/illustration/photograph, but you will get plenty of exposure”.
    Sure, the use of web fonts has been promoted by Google, and there is a part of it for which independent foundries are paid, but the vast majority of web sites use free fonts.

    Let me give you one example of how corporate concentration discriminates against the independent operator: Wordpress.

    Wordpress is the Google of web site design apps, used by 43% of all web sites.
    Not surprisingly, if you use it to design your web site, it will direct you to using Google fonts, or Adobe Fonts. (At least Adobe pays royalties, if you are fortunate enough to be distributed by it.)
  • PabloImpallariPabloImpallari Posts: 777
    edited February 2023
    Yes, nick, but on the other hand... I know of about about lots of independent font producer that started new foundries thanks to Google fonts.. and so far I have never heard on any independent font producer that went broke because of it. Even our worst critics back in 2010 are now a days releasing fonts at GF.

    For example, in my country, Argentina... the only foundry was Ale Paul... now a days there are more than 50 people making a live out of both Libre and Commercial fonts. GF helped very much all of those people to begin and now they have found their own path into selling commercial licences. For all those people, it was the opposite of "limited the financial opportunity".. it was his only "finantial oportunity"... otherwise they all will be doing graphic design instead of type design.
  • Concerning anchor points, as a general rule I think most type designers tend to keep them to a sanitary minimum in fonts with simpler contours. In the two examples of the letter /a, the anchor points seem to be where they should in order to describe the form properly. It would be the points placed at odd places that give away derivative work. 
  • Of course, the two fonts are similar; and of course, they have significant differences. (I find Spectral to be the superior font, easier to read, but that’s not the point here.) If I were arguing for the defense, I would cite the Infinite Monkey Theorem, explaining that the world of type design, particularly in the realm of Latin fonts, has become so bloated and overpopulated, with so few models for types intended for reading, that it is inevitable that more than just a few of the monkeys creating them will produce fonts that look similar. And, if all of the monkeys are French, they would produce types with trapezoidal serifs.

    It is inevitable that, when a master craftsman takes on an apprentice or employee, he is aware that the apprentice will become his future competitor, parlaying that which is learned from the master into his or her own product. And so on . . . 

  • Scott-Martin Kosofsky said:

    ... If I were arguing for the defense, I would cite the Infinite Monkey Theorem, explaining that the world of type design, particularly in the realm of Latin fonts, has become so bloated and overpopulated, with so few models for types intended for reading, that it is inevitable that more than just a few of the monkeys creating them will produce fonts that look similar.

    But why so many monkeys ? Their work(?) is not creation, it's re-inventing the wheel.
    Which is difficult because the wheel has a definite curve tension of 55% (except for circus bicycles of course).

    So, please, let's admire Igor Petrovic's original work in the post "My first variable font is out"
    That's Typedrawing! 

    By the way, I'm French speaking and don't use trapezoidal serifs :)

Sign In or Register to comment.