That's incorrect (not the quote, but the application of it) for a number of reasons, but I'm not going into them specifically. Suffice it to say that you can practically point to any object within five feet of you right this instant and it's guaranteed that they are all derivative products, in some form or another, created without consent of the original copyright owner. The line between creation and derivation is non-existent in practical terms, and software is no different.

Basically what that's saying is that you can't take an iPod, etch a different logo on it and claim that it's your own product. However, you can take the concept of an iPod, modify that, and create your own copyrightable MP3 player. Much in the same way, you cannot change a few lines of ROM and claim that it's yours; however, you can take the concepts from ROM's code and create your own MUD (whether this involves writing from scratch using ROM as the inspiration, or physically editing the original code to match your ideas is irrelevant and many people get hung up on the process rather than the end product).



While in a court setting, this may be true, but it's completely illogical. For one, there is no such thing as a truly "clean room". If I were to take this approach to creating a new MUD, I could do a pretty good job of identically replicating many chunks of ROM code just by memory. Would this not be copyright infringement?

The fact is, every product is a derivative of a previous product. That's just how life works (see: evolution, natural selection, reproduction, et al). The conflict comes not from whether your code is derived from someone else's, but whether that derivation is intended to create your own product, or simply to modify the original. As I said previously, it all boils down to intent.

I think you should, because it sounds to me as though you've misunderstood the meaning of "derivative work". A derivative work is a work based upon one or more preexisting works - for example, one which is primarily a new work but incorporates some previously published material.

Ideas, procedures, methods, systems, processes, concepts, principles, discoveries and devices are not protected by copyright law, and can therefore be copied without creating a derivative work.



If you start out with a copy of ROM, then your new version will be a derivative work - to quote 17 USC 103 (b), "The copyright in a compilation or derivative work extends only to the material contributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material. The copyright in such work is independent of, and does not affect or enlarge the scope, duration, ownership, or subsistence of, any copyright protection in the preexisting material."

The only question is whether or not it can be proven that your mud is a derivative work, and as I've said already, that's going to depend on each individual case. There are various tests used to determine possible infringement, and factors such as a documented clean room implementation can also greatly help, but there's no clear cut-off point.



Yes, of course it would - see ABKCO Music v. Harrisongs Music, whereby George Harrison was found to have infringed the copyright of "He’s So Fine,’", a song that he had heard years before, when he wrote "My Sweet Lord."

Furthermore, what you describe is clearly not a "clean room" procedure. The whole reason why it's called "clean room" is because the people implementing the new software haven't been exposed to the original - they are working in a "clean" environment, uncontaminated by any knowledge of the proprietary techniques used by the competitor.



Evolution, natural selection and reproduction are not protected by copyright, and therefore cannot be classified as derivative works under copyright law.

In many ways, it's not even possible for any of us in the MUD community to really do a clean-room re-implementation of one of the existing drivers. For example, I can write a new mud that works the way *I* want it to work without infringing on the Diku driver… but I can't honestly create a driver that is Diku-compatible because I've spent quite a bit of time working with the Diku source code.

In essence, if I tried to make something that works like Diku, I would (perhaps without realizing it) be using many of the Diku algorithms as I wrote my new code.

To really make, for example, a GPL'd DikuMUD, you would have to get someone who had never seen the codebase (or data formats), and describe how things should work while they implemented it. That description would have to be at the cause/effect level, not the algorithm level. So you could say "the who command needs to display all the players who are logged in and would be visible to the player running the command", but you can't say "the who command needs to traverse the descriptor list and display any players who are visible to the one running the command". The latter uses knowledge of the algorithm.

I know what derivative work is, and you're wrong about concepts (etc.) not being protected by copyright law. After all, what is code? A concept. What is literature? Another concept. What is a logo? Motto? Repeat ad infinitum. Copyright law is very hypocritical in this sense.



You're absolutely correct - if there is any original code left (which there wouldn't be in the situation we're discussing).



It can never be "proven", unless for some reason you've video taped yourself every second spent plugging away at the code. The best that can be done is logical speculation. I'd venture to say the tests you're referring to most likely amount to nothing more than code comparison, which is an innately horrible indicator of infringement (see my example about ROM in a clean room).



What you're referring to is perfect ignorance. While it's a nice theory, that's all that it is. It's never used in practical situations, as it would be absurdly inefficient and most likely never cultivate the desired results (not to mention the fact that you have no possible way of proving that the person is, in fact, ignorant of what he/she claims to be).



I spent a few minutes trying to figure out whether you were just being funny or completely missed the point. Still haven't figured it out…

Actually, clean rooms were used to some degree of success for cloning hardware. It's fairly easy to take an engineer who's not worked on a chip, give him the chip and I/O, but no specs at all, and have him compile a list of responses based on inputs. You can then figure out what logic must be implemented by the chip, and implement the new chip.

Clean room doesn't mean you have to have never seen the product: it means you have to have never seen the implementation. You could play a MUD and then go write one that behaved similarly from scratch, duplicating the gameplay exactly, and you'd still be ok.

http://www.copyright.gov/circs/circ1.htm...

What Is Not Protected by Copyright?

Several categories of material are generally not eligible for federal copyright protection. These include among others:

…

Ideas, procedures, methods, systems, processes, concepts, principles, discoveries, or devices, as distinguished from a description, explanation, or illustration



If I translated Lord of the Rings into Russian, none of the original words would be left - but it would still be a derivative work. Why? Because copyright also protects nonliteral elements. This is all explained in the article I posted.



Or if you (or others) had retained earlier copies, perhaps in a version control system. Or if the filtration test reveals any infringing protectable nonliteral elements.



It was approach used to create the IBM clones, by reverse engineering the BIOS and creating a technical specification, which was then passed on to a second team of programmers who had never seen the original code. It was also used to a limited degree for software in Computer Associates International, Inc. v. Altai, Inc., whereby Altai had the software rewritten by programmers who had never seen the infringing code (a clean room rewrite). The Court held that there was no infringement in the second version, although the focus of the case was more on the "abstraction/filtration/comparison" test.



That depends how serious you were when you claimed that "every product is a derivative of a previous product". We are talking about copyright law here, and under copyright law your statement is simply wrong.

Apparently you don't quite get what copyright is actually. You're describing things that are covered by patent and trademark law, not copyright law.

Yes, I've read it the same as you have. What you completely ignored (and what I attempted to explain) is the discrepancy in the definitions of things like concept, system, process, etc. For all intents and purposes, code is a concept. It's an intangible thing. You can't "create" it, because it's already been created. You're simply arranging pre-existing syntax in the same manner a child arranges puzzle pieces. This is really a non-issue, though, because no matter how illogical the laws may be, it would be impractical to classify things like this as concepts as defined by copyright law.



Yes, but we're talking about your own code here, not ported/re-stylized/paraphrased code (which, I agree, would still be copyright infringement).



Negative - such copies are not uncontested evidence. There's absolutely no way to "prove" that you (or anyone affiliated with you) created them. As I said, the best that can be done is logical speculation.



That may be what they called it, but there's no such thing as a clean room method in practicality. There could have very easily been an information leak and none of us would have ever known. I'm not saying that there was, but just because nothing does go wrong, doesn't mean that it can't. For a better explanation, ask anyone in the medical profession whether or not a quarantined area is ever actually 100% quarantined, or whether a sterile environment is ever actually 100% sterile.



I was quite serious (and right, in the context that I was describing); however, I was merely trying to show the hypocrisy of copyright law as it pertains to derivation. It's rather illogical for a legal system to be able to pick and choose what derivation is legal and what is not, as does copyright law. It all boils down to their definition of derivation, though, which just so happens not to completely coincide with the actual definition. Given the actual definition, everything is derived from something else. Nobody just plucked the idea of an automobile out of thin air, or a computer, or a mattress, or a cinder block, or <insert anything here>. The very basis of invention, technology, adaptation, evolution, etc. is derivation.



Copyright law and patent/trademark law are identical with respect to derivation. I think you nitpicked before getting to the point of my comment ;)

There's no point continuing this, Drrck isn't going to listen to anyone else, he's sure that he's right even if he isn't.

How can a long and protracted copyright argument possibly not be worth continuing, Kayle?? Surely you jest! Or are you ill? :tongue:

There's also no point in cheer leading. The great thing about law is that there are plenty of gray areas that are open to interpretation. Such areas are never as simple as right/wrong, and copyright/patent/trademarks happen to encompass one of the biggest such areas.

Alas, but I am ill, and tired, cranky, pissed off, and Um.. I'm sure there's a couple other things that qualify.

This is why my general response to these arguments is to ask for case law, and if none is presented, to make it very clear that the argument is really just (more or less informed) speculation…

Speculation. That's part of my problem. It's always speculation. None of us have law degrees but we all pretend to be lawyers. And it's tiresome to start reading a thread, thinking it's about the merits of one base over another, and have it really end up being 3 pages of BS and one page of information pertaining to what the topic is titled.

Can we all just put our ePeen aside for a while and act like adults without slinging about speculation on copyright/patent/trademark?

[Edit for humors sake]

I was happy to read your post until I saw a certain mention of male anatomy…

Would you like me to change it to ePeen? XD

No, code is an original work of authorship fixed in a tangible form of expression. It is not protected while it sits in your head as a raw concept, and even after writing it it's not the concept that's protected, but the tangible expression of that concept.

This is all explained in the "Copyright Office Basics" site I quoted previously.



It was an example of how copyright protects non-literal elements.



No, the best that can be done is the "abstraction-filtration-comparison" test, which was adopted by the second Circuit to analyze non-literal infringement claims in computer software. Once again, see Computer Associates International, Inc. v. Altai, Inc.



Yes, there is. It's been tested in court.



Copyright law doesn't apply to everything. Why should copyright terms apply to things which fall outside of copyright law?



No, they are not.

Concepts (by definition) are not tangible. Expression (by definition) is not tangible. Code (by definition) is expression of concepts. The good old transitive property of equality dictates that you're wrong.



Ah… non-literal elements… such as concepts? ;)



…which is logical speculation.



Not going to debate this point any longer, since it appears to be degenerating into "yes there is, no there isn't, yes there is, no there isn't."



Because the basis of copyright law is protecting an author/creator's rights. You don't find it the least bit odd and subjective for a system to be able to pick and choose which author/creators are protected and which aren't?



Again, degenerating into yes/no nonsense.

Guess it is too much to ask that the ePeens be put away.

They are not the same, and so it wasn't a nitpick to point that out. Copyrights, patents, and trademarks all have very specific applications. Since we're discussing derivative works of MUD codebases, the chances of anything but copyright applying are pretty slim, as I doubt anyone spent the money necessary to obtain patents on the algorithms in the code, or to obtain trademark rights to the name of the codebase or logos associated with it.