This podcast is brought to you by Rezzed.TV Podcast for the metaverse. Blue Hair Episode Number 19 Copy, Copy, Copy, Copy, Copy

Hello everybody, this is Dedric Mauriac; where in the real life, everybody calls me Lewie. Mainly today’s topic would be pretty much copyright infringement, Copybot, Content, Theft, that kind of area of subject matter. Copying has always existed in Second Life probably since the early days, but not as an easy method in which it is done today.

I would imagine at first people manually had to copy things. They would see something and try and build something similar in nature. I see a red prim. Well, maybe I can make a cube and color it red.

Hey, what do you know? I now have the same object, but now it has my name as the creator. If you got your hands on an object with full permissions, you could actually see all the settings for each individual prim, and then you could actually duplicate that manually one prim at a time.

Although it would take a lot of work. Textures were a different story. However, if you got a texture that you had full permissions on, you would be able to save that texture to your hard drive and then upload it as if it was yours originally, and then apply that texture to your objects. So yeah, copying has been around since the beginning. Now, Linden scripting language actually gives you quite a few methods to increase your output and that they’ll be able to sniff out most of the settings of each prim, and then you could export those settings to another prim and make it look exactly the same in far less time.

In not as complex of a scenario, one example would be the LSL import-export scripts I have made and make freely available. However, it’s not able to read everything about your prims. Mainly, it’s not able to replicate textures. It can replicate the ID of those textures, but it can’t copy the textures themselves. The other problems would be you can’t copy scripts, animations, sounds, just anything that’s in the content of the objects themselves.

You can’t really replicate them, but you can replicate the prims individually. Now, after LSL script, the next easiest way to copy things would be the OpenGL extractor. Now, there’s a few of them out there, and pretty much what this does is Second Life uses OpenGL to render its content. In a lot of games such as World of Warcraft, use this OpenGL. OpenGL is a rendering engine found on many 3D cards to render the contents of a game quicker than a regular CPU could do it on its own.

I think the other ones are DirectX or Direct3D from Microsoft, but the OpenGL is more open to more platforms like Linux and Macintosh. So, what this had done was it would actually take anything in your graphics card’s memory and just save it in a folder. There would be tons and tons of files that it would actually save. It’s not just images alone, it would be 3D objects. Now, you couldn’t import those 3D objects back into Second Life, but you could work with the images.

Let me just remind you again, there were tons and tons and tons of images. One of the benefits to this OpenGL extractor program would be people could make 3D printing of objects within Second Life. Now, I’m under the impression that Fabjectory was actually implemented or it was started based on this idea of being able to capture the objects within Second Life and running them through a 3D printer, which now they actually do the textures as well. So, you actually have a 3D print of your avatar with textures. So, next on the list was the cloning bot by the LibSL folks, which these folks gave a demonstration of a bot that could actually look at the information that your avatar is broadcasting for its appearance, and then it would turn that around and say that it looked the same way.

Although it didn’t have any of the objects in its own inventory, it was able to say, you know, I am this shape and size and I’m wearing this texture for my clothing. So, it wasn’t really copying the information locally, however, it was announcing that it had this information. So, everybody else who saw this bot would actually see that it appeared just like them. This wasn’t such a problem, but very quickly after the announcement of this library, some people got wind of all the programs within the library, and one of them was a little debugging tool called CopyBot. And this led to the explosion of a lot of people getting upset over this bot copying content from their shops. In simple words, pretty much it took everything that it saw, regardless of permissions, and it saved it to its local hard drive. Now, scripts and I believe animation sounds and other content within the inventory was protected in a sense.

However, if somebody starts dancing a dance, well, that animation has to be downloaded into the client in order to render that dance as well. So, I’m not certain that everything was protected, but I’m pretty sure that scripts and no cards were pretty safe. A lot of people got really upset over this, and the Linden’s reaction was pretty much, you know, copying doesn’t always mean theft.

There’s legitimate uses for copying. The second life viewer is just like an internet web browser, and pretty much if you have an issue, follow the MCA claim. Lots of people closed up shop because they were not confident at all with the response from Linden Labs. There were protests, there were people carrying up protest signs, it said, you know, down with copy bot and stuff like that. Which, I don’t really feel it did anything to help the situation other than the Linden’s came out and said, okay, fine, that’s it. Copy bot and anything similar is a violation of the TOS.

Well, you know, a couple years later we see how far that got. There are still very legitimate uses for these bots. Now today we have like, well, Second Inventory came out, which allowed you to download the contents of your own inventory and back it up to disk and then restore it. Or that still posed a problem, which if things were not yours or you didn’t own them or you didn’t create them in the first place, then it became okay, well, you can’t download the content unless you created it or somebody you know that created it has given you full trust to work with this information. So there have been fine tweaks on it to become more in line with permissions and such. But you know, if you download an image on the internet and you start working with Photoshop, it doesn’t say, hey, you can’t edit this image because you know, you didn’t create it. But I don’t know.

So there’s kind of things involved there that are a little extreme, but then there’s always ways to get around it or somebody else is selling you something similar that gets around it. And now we have something called the Mercat Viewer. Which I believe lets you just select any object and copy it to your inventory or copy it to your hard drive and then restore it. So there’s a lot of copying going on in all these tools. I think a lot of them are in response to the fact that, you know, linens don’t allow you as a content creator to actually save your content to begin with. It’s very hard as a content creator to predict potential problems with disaster. You know, I had problems where I worked because, you know, I have all this content that I created and we always put our files into source control. However, when I start working with Second Life, then my hands were tied behind my back because I had nothing to put into source control other than the images. So it’s very hard to work within the boundaries that they set on you. So a lot of people are blaming the economy, the pros of the economy on Copybot.

And I don’t really think that that is necessarily a wise decision. Mainly when Copybot came out, I think in 2006, that was like three years ago, but when it came out, people started closing shop in protest. Well, you know, now your economy is going to go in a downward spiral because you’re not even selling anything. The only people who are selling things are people giving away your content or selling your content.

You pretty much knocked out their competition. The other issue would be a lot of people started throwing up these boxes that would be anti-Copybot theft devices that would shout out a command. And exclamation marks quit. And the reason for that was like the first Copybot versions would actually stop once that command was executed. But then that was easily circumvented right away and that they no longer listened to that command.

And I think to begin with, that command was only listened to if you were the owner of the Copybot. So even in the beginning, it was pointless. What that did in turn was it just drove away customers who were just getting pretty annoyed with that quit statement.

I remember I constantly kept blocking anybody who owned content that was actually shouting quit. So, you know, that’s another one for the designers to take a hit on because of their… It might be harsh, but I would have to say they were ignorant in how Copybot worked. So, there have been changes since then to help people identify when their content has been stolen. For one, you can now inspect any object or avatar to get a list of what objects are being used and who created those objects. So, if you see someone wearing a shirt that you made, you are actually able to find out who created that shirt. Now, that’s important because now you know who to file a DMCA claim against for copying your shirt. So, there are many avenues here that I could go down in analogies with why copying is legal, why it’s not really a problem. I think Corey actually suggested that they would just make the whole linen lab client able to easily view the source of objects in the way that you can view webpages today.

Any browser you get pretty much lets you view the source of a web page and most of them will actually let you save the entire web page, images, style sheets, scripts, everything as an archive to your local computer in some way. If Second Life was the same way, his prediction was that it would be less of an issue to claim copyright infringement and such. Well, you know, that starts to become a really big issue especially if you have a pietary software or code that you don’t want people to see. With webpages, you have what the client sees but you also have server side logic that the client does not see. And unless there’s a way to separate logic between the server and the client, that’s still going to be an issue. The other problem is a lot of the content is really just static client side content.

It would really do more harm to do that than actually being a benefit. The main thing that I see here is they need to be able to control what kind of viewers can actually have access to their grid. Mainly when CopyBot came on scene, I don’t really think there was any way to detect that a client was using CopyBot. Now they do have this string that just about any browser or client that connects to Second Life announces, hey, I am this piece of software and I was made by such and such a person and I am this version.

Well, yeah, you could use that to detect bad software and restrict people. However, that information can easily be changed. Anybody can download the source code, change the value, compile it and then connect to the grid. I think what they need to do is they need to actually provide an API key of some kind to the creators of these viewers. So that when they compile the viewer, they compile the key into it in some manner that also announces what kind of viewer it is and what version. And in this way, the linux can actually have a more accurate list of which viewers are connecting to their servers if the viewers are in beta. And if there is a problem announced with one of the viewers, they can immediately revoke the license so that nobody can connect to it with that specific viewer until the problem is addressed. Mainly like what if they have an issue with the Mercat viewer with the ability to copy content? Well, then they can revoke the access key until they get the go ahead to saying that the information, the feature has been removed. And then they issue another API key for the different version. As far as open source goes, the API key should not be in the source code. It should only be used when the person who is compiling it is actually compiling it. That way, anybody else who downloads the source code would have to use their own API license key for testing and development and such. That might not be a viable solution in some cases.

People can always work around problems. What if someone is able to sniff out the actual key in that viewer? How do you encrypt that key so that nobody can get access to it?

And then what happens if somebody can just sniff the network packets just to figure out what that key is? So, yeah, there is a bit of an issue there, isn’t there? Oh well. Well, anyway, those are my two cents worth in the history and evolution of copying content in Second Life. Maybe someday the Linux will give us the ability to actually copy our objects and import them back in. But until then, we’ll see what happens. Alright, later.