Wednesday, June 18, 2008

A Model Opinion Affirmed

On October 4, 2006, I did a blog on a district court opinion in Utah holding, on summary judgment, that computer modeling of Toyota cars did not meet the requisite standard of originality. I was surprised by the opinion. The Tenth Circuit unanimously affirmed yesterday in a well-written and persuasive opinion by Judge Gorsuch, Meshwerks, Inc. v. Toyota Motor Sales, U.S.A., Inc., (Docket No. 06-4222). The opinion contains appendices with pictures that are helpful. I still have doubts about the result, but now believe it a closer question than I previously thought.

The relevant facts are these: for the introduction of its 2004 Toyota hired the advertising company of Saatchi and Saatchi; both companies agreed the campaign would involve digital models that could be used on Toyota’s website and in other media. Toyota and Saatchi hired Grace & Wild to supply the digital models. G&W in turn subcontracted with plaintiff Meshwerks to do some, but not all of the work, in producing the final digital models: Meshwerks would do the digitization and basic modeling; G&W would then take Meshwerk’s work and add details, as well as color, texture, lighting, and animation. The suit arose over a dispute about the scope of the license granted by Meshwerks.

To perform its part of the ultimate product, Meshwerk employees began with the actual cars. The employees put tape on the cars in a grid pattern. Each point of an intersection of the grid was marked by a mechanical arm, to which a computer was attached, resulting in a file of data points. Using off-the-shelf software, those points could be opened up, and once opened up an image of the car would appear. It is undisputed that it took Meshwerks 80 to 100 hours to produce a model for each care; it is also undisputed that Meshwerks altered over 90% of the data points collected during the original digitization process, in order to get a model that looked as much as possible like the actual car. Some elements of the car had to be created from scratch.

If one were to apply a labor theory of originality, there is no doubt Meshwerks would meet that standard. But the labor theory was rightly rejected by the Supreme Court in its 1991 Feist opinion. Two factors were of critical importance to the Tenth Circuit in affirming the district court’s grant of summary judgment to defendants (1) G&W, not Meshwerks made all the decisions regarding “lighting, shading, the background in front of which the vehicle would be posed, the angle at which to pose it, or the like; (2) Meshwerks’ purpose was to “depict nothing more than unadorned Toyota vehicles, the car as car.”

On this last point, the pictures in the appendices help. I agree that if the final result is not copyrightable, it doesn’t matter how much labor or skill went into production of the models. The use of off-the-shelf software to produce the data points did not result in a copyrightable work; instead, the originality, if any, must be found in what Meshwerks did after the original digitization and before G&W took over and added more traditional creative elements. The court of appeals seemed to think that Meshwerks’ contributions were taken the unprotectible data points and then tweaking them so that they looked as close at possible to the real car, an effort in found to be unoriginal under Feist, regardless of how many hours it took.

I am more sympathetic to this view than I was before, but I am still bothered by the premise that efforts to realistically depict an actual object are unoriginal. Superrealist oil paintings are an obvious example. I have a few posters of them, and if you don’t look closely you think it was a photograph. Photographs are of course protectible, usually, due to the lighting, shading, angle and other choices. But in Meshwerks those choices were made by G&W, and this is what seemed to persuade the court of appeals Meshwerk’s contributions were unoriginal.

If I was still writing casebooks for law school classes, Meshwerks would be a must for inclusion in the next edition.

10 comments:

Steve said...

Hi Bill, fun post. I haven't read the opinion yet, but I'm fascinated by this sort of case so I'll surely have a look at it. I have a question that can be posed even without reading the opinion- if Meshworks could copyright the computer model of the car, what would stop Meshworks from suing Toyota for producing a substantially similar derivative work? That can't be right.

The car comes first, and Meshworks admittedly modeled the car- so any copyrightable elements in the model are necessarily derivative of the rights in the car itself!

I'd have to disagree with your opinion that this is even a close question- the necessary outcome dictates the court's approach. Any originality in the 2D model must arise from the design of the car itself, not from the translation. And if Meshworks does its job properly, they don't introduce any additional creativity into the translation they produce.

The only way that Meshworks could reasonably claim copyright is if they introduced some artistic originality or creativity into the translation- but that would presumably ruin the work they were hired to do!

Dan Lewis said...

I saw you speak on a panel re: Bridgeman Art Library v. Corel Corp., and since then the labor theory of originality has fascinated me. Thanks for sharing this one.

My first reaction to the rejection of the labor theory was a visceral one -- it seems that the producer of the copy was "robbed" of his efforts. What I'd like to see is if these types of producers start to draft copyright-like terms into their work contracts. It seems like the next natural step.

Christian T. said...

steve ----

I believe that the copyright protection afforded to Meshworks for its 3d models would be extremely thin. In this way, Meshworks is protected from any appropriation of its actual models, but has no right of action against a party who makes its own 3d models.

The key issue is whether you believe there is a merger of idea and expression. I disagree with the premise that there are only a limited ways of expressing the models and therefore think that Meshworks is entitled to protection for only their independently creative contributions.

Anonymous said...

I find it somewhat entertaining, that the opinion, that you disagree with, contains several citations to your own treatise.

William Patry said...

Dear Anon. I am very grateful for the citations amd respectful toward the opinion. I have no disagreement with the court's discussion of the law. Its discussion of the facts is full and fair. I also indicated that my own view of the facts has moved somewhat too. My reservations are about the facts only.

S. Bielstein said...

As someone who has studied Bridgeman and written about the image trade and the public domain, I have long hoped to see the case applied--and expanded upon--at the Circuit Court level, so this is great news. It may even prove to be something of a landmark since it does grapple to a degree with matters of intent. As to your thoughts about superrealism in painting, it's an interesting subject. I think one might aver that, in a hyper-real painting the execution is so self-reflexive--drawing attention to itself--that that becomes the overriding content of the work, more so than the object painted. At least that's how an art critic would see it. But it is still an interesting legal question, especially since, as we know, a change of medium does not itself confer originality when it comes to making a copy.

Susan Bielstein
Executive Editor
The University of Chicago Press

Steve said...

Christian: you said "The key issue is whether you believe there is a merger of idea and expression."

While that is a valid question in many copyright cases, I don't believe it applies to the facts here. Meshworks was hired to copy an existing form- so anything they produce is work-for-hire regardless of whether there is merger.

My beef is that the copy of the original Toyota form is by definition a copy, so it can't have the requisite originality to merit protection. Either it's a faithful copy, in which case the work does not have originality, or it's not a faithful copy, in which case Meshworks don't get paid because they didn't do their job...

Mesworks producing a satisfactory model is inconsistent with their claim of copyright in the work product. The merger doctrine should never even enter the (heh) picture.

Sunny Kalara said...

I strongly disagreed with this opinion and I think it is wrongly decided.

This case has a lot in common with the copyrightability of a typeface. The typefaces are not coyrightable. 37 CFR 202.1(e). The bitmaps of the typefaces are not copyrightable, as they are nothing but electronic representation of the non-copyrightable typefaces.

However, the digitized typefaces, the scalable true-type fonts ARE copyrightable. And how are these true-type fonts generated? By putting data/grid points along the curves of the typefaces and then tweaking them for hours on end, so the final result is exactly same as the original typeface.

The copyright office opined that:
``... the Copyright Office is persuaded that creating scalable typefonts using already-digitized typeface represents a significant change in the industry since our previous [September 29, 1988] Policy Decision. We are also persuaded that computer programs designed for generating typeface in conjunction with low resolution and other printing devices may involve original computer instructions entitled protection under the Copyright Act. For example, the creation of scalable font output programs to produce harmonious fonts consisting of hundreds of characters typically involves many decisions in drafting the instructions that drive the printer. The expression of these decisions is neither limited by the unprotectable shape of the letters nor functionally mandated. This expression, assuming it meets the usual standard of authorship, is thus registerable as a computer program.'' 57 FR 6202.''

In this case, Meshwerk didn't create a bitmap of the Toyota vehicle; they created a digitized vector model of the Toyota vehicle and inherently introduced original expression in the process.

That's the whole point of creating a digitized model, so you can scale it, rotate it, move it, put a different background on it.

If a true-type font file is copyrightable, there is absolutely no reason, in fact, I would argue that it is inconsistent to argue. that a (vector) digitized model of a real object is not copyrightable.

Carolus said...

This is an interesting point for that relates to a project I've been involved with also. (Just discovered your blog, Mr. Patry.)

In October of last year, the site imslp.org, a Canadian site featuring PDF scans of public domain music scores, was taken down under C & D threat from an Austrian music publisher over (among other things) a 1967 critical edition of Mahler's Symphony No. 1. Mahler of course died in 1911 and is not under copyright even in the EU. Canada's law is not terribly clear on the status of such editions and whether they meet the threshold or originality to qualify for copyright status. US case law (as in Feist and Bridgeman) would appear to hold that such editions fail to meet the standard. Nevertheless, most US classical music publishers (there are about 10 of them) seem to view such editions as a full-fledged copyright with all the trimmings.

The Mahler score at hand is especially interesting because the publisher, Universal Edition, did not even re-engrave the score, but simply corrected the 1906 engraving that had already been in print for six decades. This issue (critical editions) would appear to directly apply to previous poster Mr. Bielstein, as the Univ. of Chicago Press is the joint publisher with Ricordi/BMG of the Verdi critical edition scores. Interestingly, such editions are only protected for 20 years after publication in Italy, and no more than 30 in the EU as a whole.

Lewis Francis said...

FYI, Meshwerks today filed a petition for re-hearing of the Tenth Circuit's decision, on the basis that Meshwerks registered for copyright the computer programs underlying its digital models, not the generated images of Toyota vehicles. As a result, Meshwerks is asking the Court to address the issue of whether the computer programs, which are original and unique to Meshwerks, are protected by copyright, even if the generated screen images of Toyota's vehicles may not be.

Lewis Francis, attorney for Meshwerks.