No, its data – data being manipulated of existing software – software which would be there regardless of whether or not people make something.

But either way, there should be no right, virtually or otherwise intellectual “property” is one of the most sick inventions of the 18th century, and one can only hope the internet will actually help destroy it.

I followed this over here from Tom’s elearning newsletter and my attention is rapt. (Makes note to get in touch with Tizzy in-world for future discussion sessions.)

Thanks for the very comprehensive article and the many links!

How does something like this work if you have your own space in Second Life and do screen captures of your own avatar?

Lots of arguments about the issue are comparing snapshot (or processed snapshot) to RL photograph. While that would make lot of things easier, those are not the same. Usual argument is that all the content of a virtual world is created by someone who holds the copyright and authorship on the creation. But then, what about architectural photography? Are those two the same? If you check the arguments of creators on Bettina’s post, they say that, unlike photography, for SL snapshot there remains a doubt who is the creator of the scene. Is it the person who took the snapshot or somebody else?

But that’s the question of authorship not copyright. For copyright matter (if I understood well) there is a copyright infringement because scene is a piece of software. And somebody has copyright over that piece of software. But we forget that, years ago, there were lot of discussions if architectural photography is copyright (or at least authorship) infringement. But it isn’t. For the very simple reason: when you are taking a photo of an building, or a statue, you are changing the medium. Your work is based on somebody other’s work, but you apply your skills, your vision of the artwork, your feelings and ideas and make something new. for the same reason, photograph of a painting is not an original work. Same goes to screenshoting the website design. For the same reason, making an image out of 3D model shouldn’t be bad thing to do.

I have to take my turn on the question Benjamin is trying to leave to professors: whether the internet will change copyright law dramatically? It’s not the Internet, but digital technology. Some 15 years ago or more, nobody cared about copyright infringement done by common people. Music was stored on vinyl records and text was printed on the paper. Yes, I could copy that music on the tape and photocopy the book. But quality wasn’t the same. And if I had the chance, I would pay the price to get the book or vinyl. If I had not the chance, that is, if all the stocks were sold and no re-releases in sight I could make the copy, which is the infringement but there’s no court on the planet that would prosecute me for that.

Those times are over. Copy of a CD is a new CD of the (almost) same quality. But that almost is not important enough to stop millions of people world-wide to do it. Even reducing the quality to mp3 is not enough. And mp3 is, in many cases, even more convenient than the original CD. Our problem is that our technology is developing much faster than our ethics and our legislation. And copyright is not the only area we feel that problem.

Anyway, I understand both sides. I understand 3D creator’s need to protect their authorship and profit. But I also understand needs of those that use that content for their work. In my take on the problem, I tried to imagine what would it take just to credit all the people who contributed to a single snapshot. Then that should be multiplied to get the amount of work needed to get all the copyright permissions to set the things legal. And that was just a single image. In the case of a comic of machinima, it is less work, time and nerves to forget about the second life, open Blender or Maya and do all the work alone. But then, beauty of SL is in sharing and collaboration even if that collaboration is not always direct. Will all this end up in great failure of collaborative environments just because we didn’t learn to collaborate and share?

Ben’s reference to Google’s low-resolution images in its image search might deserve a bit more attention to explore this point. The case he’s referencing, I believe, is the Perfect10 v. Amazon case, which found that these low-res thumbnails of Perfect10 models were sufficiently transformed from the high quality photos. However, the argument was susceptible to Perfect10′s defense that there was a market for these low-res thumbnails, which at most would probably turn them into derivative works and at least would weigh against a fair use defense on the first and fourth prongs (type of use; market effects).

To make this argument, Perfect10 said that the market was in cell-phone users who would buy these photos for their phones. If you’re skeptical about that market, then you’re in agreement with the 9th Circuit. Perfect10 could not show any existing market for this argument and had not sold any such images to cell phone users.

Lem makes a parallel (though more defensible) argument above, by noting that a 2D representation of a 3D environment is sufficiently transformative (or maybe the argument these days is just “low enough in resolution”) to weigh in favor of a fair use finding.

Ok, so, to argue against transformative use, we have to figure out what the harm to the creator is in this case. Of course, if the photograph is almost solely about the creative work, it’s probably infringement no matter what, just like the Wall Street Bull or a picture of Ben’s book pages. But if it’s a smaller use (a sculpture in the background), then what?

Well, presumably he’s not in the business of marketing people wearing his clothes or displaying his statue. He probably takes pictures of his clothing or statue for use *in* marketing, but not as a separate product, and not for sale. So tangential uses of his work in the photograph may be defensible. Generally, I would argue that you have to say that people aren’t buying his clothing or statue, at least partially because they *see* (or buy) this photograph.

But, there’s one big way in which SL differs from RL in terms of public display of statues or other works of art: in RL, you can’t right click on a statue to figure out who made it and get yourself a landmark to his store. In SL, arguably, this is how most content creators actually get business — somebody sees the product in SL, likes it, right clicks on it, finds out who made it, and goes to his/her store. Arguably, if a “market” develops where SL users are regularly exposed to 2D, non-interactable images of SL items, content creators would be robbed of this inflow of customers. Customers who would explore SL less when they can just look at still pictures of it. (It sounds a little silly, but I’ve definitely seen pictures of places in SL on websites and decided that there was no point to visiting it since the pictures did a good job of capturing everything I would want to look at.)

Is that a copyright argument? I really don’t know. Misattribution has a role in copyright, but in the US sans-moral-rights world, the argument is tangential and sounds more in consumer fraud. And this isn’t really misattributing (or claiming the statue as the photographer’s creation), so much as preventing customer exposure to the full work and the work’s primary stand-alone method of marketing.

But it is a harm. Arguably, to the “market” for the work. And it could be substantial.

Although I’m generally in favor of sticking to RL precedents when evaluating SL legal issues (rather than trying to argue that SL is completely different, as some of our virtual brethren do), this could be a case where we really need to evaluate how the SL interface affects and alters the rights of virtual content creators.

Forgive the ramble, but I thought this was an interesting way to look at the problem. :-)

First, this post is descriptive (what the law says now, at least as I analyze it) and not normative (how the law ought to be). I’m not saying that finding that screenshots violate content creators’ copyright in their work is a great outcome, just that it’s likely under the current copyright framework. I try to keep my opinions out of this because, frankly, why would you care? There is a raging debate about whether the internet will change copyright law dramatically, and I tend to leave it to the professors, figuring that readers here are more interested in practical advice.

Second, there are definitely good arguments against this finding, as outlined above. The best argument is that some of the public policy considerations that support restrictions on copyright for photographs apply to virtual world screenshots too. I suspect, however, that at least at the trial court level (where this will be decided initially) they’ll find infringement the same way they would certainly find infringement if somebody at Lucas Film re-rendered a few shots from Revenge of the Sith with different lighting and angles using the original data on the computers at the company and then posted them on his website. The arguments against this finding pertaining to virtual world screenshots are reasonable, but they’re much more likely to prevail on appeal (if they prevail) than at the trial court level. And actually, that’s often how law gets changed to account for new technology. Could happen here, but until it does, I believe that most district courts would find that screenshots that don’t qualify for fair use exceptions violate content creators’ copyrights.

Third, somebody asked if screenshots of websites would infringe copyrights on these grounds too — and yes, they would. I hold the copyright in Virtually Blind’s design and content — taking a screenshot of it violates that copyright the same way that copying and pasting the text would. There’s nothing magic about screenshots that makes it okay to copy things that way.

Finally, some people seem to think that because there’s no case law directly on point either possibility is equally likely. That’s just not how it works. When there’s no case law, courts will look for the most similar situation, and here, I strongly suspect that at least at the trial court level, they’re more likely to see Second Life creations as similar to other renderings of 3D objects than to real life photography.

That all said, it is undeniably a close call and there are good arguments both directions, as noted above. While under the law as it stands I think that screenshots violate the copyright of the creators of content pictured in those screenshots, I would absolutely be comfortable taking either side of this argument to a judge or jury.

What would be sad is to kill all in-world photography, machinima, etc, in the name of protecting property and rights when in fact those were never in danger from the activity in question to begin with. If I photograph a sim for a blog article about it, or use a build in a piece of machinima, it seems to me it would more promote the build than somehow benefit myself at the cost of someone else. But then again, for any kind of machinima that’s a story and not an exploration deliberately describing the build, attribution should be required.

But I find it hard to believe that making a machinima wearing a certain prim hair would be considered a derivative work of the prim hair. Maybe it would, but to me that just seems like content creators being anal for the sake of it. I know for my own meager creations, I don’t give a rat’s rump if they attribute me for pictures/machinima that includes it, so long as they aren’t trying to pretend they created it. Know what ah mean, Vern?

On the technical side, I think the description is rather over simplified. I’d like to walk through some of the the steps which happen between content creation, its blending into a scene and then its presentation to users. I’ll then leave it to lawyers to see if it changes the legal parsing of the situation.

The items in a SecondLife (Or pretty much any Virtual World) scene are not, in fact software. They ares software renderings of a bunch of digital artifacts. Each prim is described as a bit of constructive geometry, or with a mesh map. Each surface is assigned a digital texture. Content creators use various tools to create and combine these into artifacts, either textures, prims,or geometrically linked sets of these These are then all loaded into the region simulator which keeps track of where they are, and sends all this information down to the user’s client.

On the user’s client, the current scene graph is finally rendered. The prim shapes are mapped onto the 3d space textures are painted onto the objects, and light is then simulated. Depending on the client’s settings, graphic card, and things like effects tweaking, a scene is now rendered.

A user, taking a snapshot of the scene, is taking a composite of:

a) The original constructive geometry.
b) a scale and possibly color transformation of the textures on the prims
c) The programmatic rendering of, sky, water and reflectoins
d) The programatic rendering of flexies and particles

These are subject to further modification by the client’s interface, such as turning on and off rendering effects, moving the camera around the scene.

For added sport, many of the items in the scene are changed by the object’s owner (stretched, resized, and so forth) System Clothing is painted onto the Avatar’s shape. Prim attachements location is dictated by the avatar’s shape, and by it’s animations. (Either system default, Animation Override, PoseBall, Dance Machine, or played in world gesture or animation, or the combination of all of the above.

Clearly, there is a path between the original content the content creator uploaded, and (within the context of the second life service, offered for sale to some users) It is not, however a simple, or even deterministic path.

I don’t know how this rather more complex description impacts the legal implications. But, the scenes which we see on Second Life are complex combinations of multiple inputs, modified in multiple ways, and then rendered by very sophisticated software. The original inputs are part of this process, but they are hardly delivered to the user in their original form.

This description is somewhat simplified, but I hope it will stimulate further discussion.

- Zha Ewry