People don’t really watch advertising anymore. With all the (illegal) downloads, the Tivos and DVRs, there’s no reason for people to watch advertisement. In light of this, I’ve often said that businesses should start to make art-advertisement (maybe I should call it . . . → Read More: The Discovery Channel Artvertisement]]> I’ve just seen this (new?) ad for the discovery channel, and I think it rocks!

People don’t really watch advertising anymore. With all the (illegal) downloads, the Tivos and DVRs, there’s no reason for people to watch advertisement. In light of this, I’ve often said that businesses should start to make art-advertisement (maybe I should call it artvertisement). Adverts I want to watch. Other good examples (if you ask me) are the Cog Ad by Honda, the Sony Bravia “Splashy colours” series of ads and others.

I would add this particular ad to that list as well.

So the wikimedia foundation just announced a deal was accepted by everyone that would . . . → Read More: Wikipedia licensing will be compatible with creative commons]]> [I know I'd promised this post would be about the great idea I have and which I've been dangling around without really being descriptive, but as far as I can see, this has just come out and I need to talk about it.]

So the wikimedia foundation just announced a deal was accepted by everyone that would make the GFDL (the gnu free documentation license) compatible with the creative commons license. Wow, ok, so what does that mean? Two short points.

1- What’s a license again?

The “copyleft movement” doesn’t claim that it won’t use copyright, rather it encourages creators to voluntarily give some or all rights away formally, but within the confines of copyright law. How do you do that? You use a license. As an example, let’s take wikipedia. Formally, everytime you modify a page on wikipedia, you own the copyright to that page as a derivative work of all the stuff before… until the next edit of course. By using the wikipedia framework (the wiki, the site itself, etc…) you agree to post your copyrighted stuff according to a license. In this case, wikipedia uses the GFDL, the GNU Free Documentation License. It says many things, but the summary is that everyone is free to use the stuff freely for non-commercial or commercial uses, at will. What you have to do though, is that whatever is licensed under the GFDL has to be distributed under the GFDL and also carry along with it the whole text of the GFDL (which is pretty long). It also says that if you make a derivative work of the stuff (which is what you do when you edit a page) you have to distribute it under the GFDL, so that if you use something that’s licensed under GFDL as a base, you have to make your end product licensed under GFDL as well. It’s not as bad as it seems as the definition of a derivative work isn’t all encompassing. For example, if you quote something from wikipedia in a book, you don’t have to distribute the whole book under the GFDL. Is the GFDL the only license of that type? Absolutely not. There are dozens and they’re not all compatible, which has caused headaches at time. Another license which is pretty close to the GFDL is the CC-by-SA (creative commons attribution share-alike license). Now you may think, “wow!What’s that?!” Let’s describe it.

2-Wow! What’s CC-by-SA?!?!

CC-by-SA stands for creative commons attribution share-alike license. The creative commons is a not for profit organisation that came out of Stanford and whose purpose is to design license agreements that allow you to easily understand what you’re doing and easily give away some, all or none of your rights away. To achieve that goal, they have crafted a whole set of lienses (and are crafting more) as well as have ported them to different jurisdictions (different countries). So hte CC-by-SA is only one of those licenses. What does CC-by-SA say? Well, basically the same thing as the GFDL. It says you can use the work as you like, but you have to say where it came from and the resulting work has to be licensed with the same terms. Now, it’s nice and simple, but it’s not compatible with the GFDL? So at this point, you may ask “Why?”

3-Why are GFDL and CC-by-SA not compatible?

Simple. In fact, right now, the incompatibility is only one-way. You can take stuff that’s licensed CC-by-SA and integrate it (or just release it) under something that’s GFDL, because the GFDL has the same terms as CC-by-SA. You can’t take something that’s GFDL and release it under CC-by-SA because the GFDL says the derivative work has to be licensed under the same license. CC-by-SA only says the license has to have the same provisions, but it can be a different license. Annoying heh? Two things achieving the same goal can’t talk to each other. So now you may think “Well, that’s nice legal stuff, but why should I care?” Let’s look at that.

4- “Why should I care?”

Well, at least I (Charles) should care because I administer websites and I have to deal with these licenses. If you look at the right bar of this website, further down, you’ll see a little logo with the name of the license. That’s right, this site is licensed under a creative commons license. Neat heh? I don’t actually mind people reusing what I write here. I just ask that they be nice about it and follow my conditions. Now that’s nice, but I don’t often use wikipedia stuff in here. In fact I don’t. One area where this was a problem was wikiscuba. I started that website with a british diver and we had to decide, at the beginning, which license we were going to use. We wanted to have a creative commons license so that we could interface with wikitravel. Wikitravel is a travel wiki and since we have listings of dive sites, we wanted to be able to re-use stuff from wikitravel in ours. As a matter of fact, we did. Way at the beginning, I copied the country pages from wikitravel to give us a start and some structure in the dive site section. The problem is that we also have a techniques/gear/instruction kind of part of the site and it would have been nice to have GFDL licensing so that we could take stuff from wikipedia. We had to make a choice. We picked creative commons. With the hindsight, we could have picked both. It’s possible to dual-license all the stuff. We didn’t. Why? I don’t remember. That’s where the headaches start, you see. Thanks to this new move (I’m not sure whether it’s a done thing, but when it comes, you’ll know for sure, wikiscuba will gain about 1000 pages overnight and I’ll happily add information on there!) we won’t have to choose. Now we can just get back to the business of creating more information, adapting more information and distributing it so people can use it.

5-”That’s nice, Charles, you’ve told us why you should care, but why should I care?”

Ah! Good question. Well, in today’s world, it’s easy to create content and distribute it. Do you have blog? Did you take a picture of your dog and put it on it? Then you’re distributing content. (Just for the record, it could also be a picture of you, your boyfriend, your niece or anything else.) What about the text of your blog? What if you wrote down something, put it on a webpage for everyone to see? You’re distributing content. What if you sing a song you recorded? You’re distributing content. What if you write a piece of software and you put it on your webpage? You’re distributing content. (There are licenses specifically for software, for example linux is licensed under the GPL which is also very popular. There are, for software also, a lot of different licenses and they’re not a priori compatible, either because of the terms they use or of some technicality. Firefox, for example, is free and open source software, but is not distributed under the GPL.)

So as you can see, in today’s world, it’s easy to create and distribute content. (As a corollary, it’s also easy to copy someone else’s copyrighted content and distribute it.) An example which is hopefully more interesting to the scientist-readers of this blog, is the following:

If you write a report, produces plots, integrate it all in a pdf and put it on your webpage, you’ve created and distributed content. Now, what do you want? Do you want to make sure people know they can cut out a figure and re-use it in their Ph.D. thesis while attributing you as the source? This is probably something most scientists want. One option would be to release your work under the Creative Commons attribution share-alike license, the CC-by-SA. [Just for the record, I don't work for creative commons and I don't get anything from them for saying this or anything else.] In today’s massively interconnected world where fast communication methods are available to many, it’s important to know about copyright. Especially when it’s so easy to break the law by copying someone else’s copyright these days.

So that’s the long story about the short announcement. This is a great day. Interoperability between “open source” or “copyleft” projects is important or, at least according to me, desirable. Sometimes you don’t want interoperability. For example, I believe the license which governs the BSD operating system (most people have microsoft windows, mac people have MacOS, the free and open source operating system normally talked about is linux, BSD is another such free and open source operating system) allows people to take the source, modify it, compile it, package it in a machine and distribute it around, making people pay for it, without ever giving the program away. (If I’m wrong, someone correct me, but I believe it’s the BSD license which allows that.) Now I don’t think that’s wrong. It’s a choice. And the BSD people, at the beginning, made that choice. Linux, on the other hand, says that you can re-use linux code (well, GPL code really) as you will but you have to give the source away. Again, it’s a decision. It has good effects: the license basically tends to multiply and your code would never end up to disappear from the usable-for-free pool (which could happen to BSD although it’s quite unlikely). It also has undesirable effects: there are people who might want to use the code base and create something really awesome that you’d actually want to buy, but they won’t because they don’t want to give the source away. The original linux people made a choice.

My final point is that when you create content and distribute it, you have to make a choice as to how you’re going to distribute it. You can give all your rights away and put it in the public domain. You can give no rights away and simply not license your creation at all (or basically only license it for viewing). You can give some rights away in between those extremes. Still you have to know what you can give away and what your choices are. I recommend the creative commons website as a good start with information although there are more ressources out there.

[Originally seen on slashdot and Jamendo Blog]

Part of the reason I’m . . . → Read More: Science and copyright]]> I swear, in my next post, I’ll give all the details I think I can give (that is, all the details which are set in my head) about my “great” idea for the future of scientific communication. In the meantime, I just want to have a little chat about science and copyright.

Part of the reason I’m going through this thought process and devising this master system (I’d use the word framework, but for personal reasons, I hate the word framework) is that I believe there’s something wrong with how scientists communicate. I’m not saying that scientists are doing everything wrong when it comes to communication. In fact, I think many things scientists do are right! Let me outline this in 3 points:

1. The fundamental spirit of how scientists approach communication is perfect. The open-ness of scientific communication should never change. All scientists I know, not only recognize this fact, but probably worship that principle. I don’t think I’m teaching anyone, anything here.
2. The current use of copyright is dangerous. Basically, we’re going about our business, applying number 1 just above to our results, but giving our copyright away to scientific journals, some of which are looking to use it, as the PRISM initiative shows us.
3. The tools and methods we use to disseminate our scientific results and ideas are outdated and inefficient. This is where my framework idea comes in. I’m really not looking at revolutionizing scientific communication (although I’ll claim it at some point so I make noise and people notice what I’m doing), but rather to evolve it to the next step. The underlying principle (number 1 above) will remain. Only the packaging and the speed at which it traves will change.

Let me discuss points 1 and 2 a little here today. More explicitely, let me depart from the usual and try to take an idea from science and apply that to another field.

In number 1 above, I praise the principle of open-ness of scientific communication. What I’m saying is that the founding principle we use to do science is that anyone can take what I’ve done, scrutinize it, critic it, replicate it (in fact, we’re happy when someone redoes something we’ve done and confirms we’ve done it properly), expand on it, etc. All that scientists ask is that when you re-use what someone’s done, you mention who’s done it first, done it wrong, done it partially. Expressed in this way, it sounds a lot like the attribution creative commons license. What’s a creative commons license?

Simple. The Creative Commons is a non-profit organization whose sole purpose is to create lienses which can be used by content creators. These licenses come in different forms, a machine readable form, a lawyer-ese form and a human-speak (for laymen) form. As a content creator, you select which rights you WANT to give to people over your work and you attach the appropriate license to your work. This indicates to potential (re-)users of your work “These are the things you can do with this copyrighted work.”

The attribution license says that you can re-use, use, distribute, mashup (the new buzzword on the internet), remash, paint over, trace over, deform and enjoy the current work. All you have to do is attribute the source properly. This is exactly what we do in science.

One point of clarification though. Legally, copyright extends over EXPRESSIVE WORKS and not over IDEAS AND FACTS. So technically, someone can take your paper, extract the actual data points and re-use that information. Information is not copyrightable. So people don’t even need to attribute the source. In this case, the “science code” goes beyond copyright. When re-using INFORMATION laid out by a previous scientist, we still attribute the source.

On the other hand, there are behaviours scientists exhibit, which don’t respect copyright. For example, we often re-use images from other scientists and only attribute the source. In my master’s thesis, there are images from the phd thesis of a student at Caltech. I still attributed the source, of course. I’m sure I can pull loads of master’s/ph.d. thesis in my field with copies of images from older papers, probably from many recent papers. In the case where one pulls an image DIRECTLY from a copyrighted work and only attributes, one is technically breaking copyright law. Now, I doubt Caltech is going to sue me for using that image (or the copyright holder for that matter, whom I know well enough by now). In the end, using that image and simply attributing the source FEELS right. This is a case where using the creative commons attribution license would work well. This would have enabled me to use the image without worrying about the copyright detail. I still didn’t care, but technically I should have. Were I to be sued, I would argue fair use (as a critic of the work). Still I’m really not likely to be sued by Caltech for this.

This is where my point 2 comes in. We’re using our traditional distribution methods (archival journals and the such), knowing full well that we’re signing our copyright away in most cases, yet without caring about the consequences. Basically, because there have been no real consequences so far. Once in a while, there’s the odd over-zealous journal staff who sees infringiment and sends a letter to the infringer without clearing it with a boss first. Yet, I’m not aware of anyone getting really screwed over this. I’m willing to bet most (all?) scientists are more worried of being discretited because of accusations of plagiarism than because of lawsuits over copyright infringiment.

So in light of this unwritten policy in the scientific domain (we could call it the no-sue poliy), why should we care? While the danger isn’t imminent, groups like PRISM are worrisome. They show that someone, somewhere cares. That person hasn’t used their right yet, but they know it exists. I don’t like the status quo, especially when I’m not in a situation of power. My solution to this is to say that while we devise a system to revamp our distribution methods, let’s take the opportunity to cut-out the middle-man and use licensing schemes which solve this issue. More on this next time.

For now, as promised, let me draw a parallel between what happens in the scientific world and what happens in the music world.

As I outlined above, the way science is done (on a global scale, not on a laboratory scale) is that we constantly re-use, re-hash, re-mash other scientists work. Joe A, in Italy in 1971 publishes a paper which is really interesting. He shows that a species of frogs in Madagascar has the ability to spontaneously change sex. (I’m making this up as I go along now, in case you hadn’t noticed.) Fast-forward to 2002, when Jane B, in South Africa, shows that all frogs have the genetic makeup to do this, it simply happens that only a fraction of those species have the need to use spontaneous sex change. (Again, it’s completely out of thin-air, just an example. I’m pulling this out of the script of a bad science-fiction/cataclysm movie right now.)

In the end, Jane B simply writes in her article that “Joe A has previously shown that…” and maybe pastes in a graph from the original 1971 article, showing the main conclusions. Re-use of previous science to make new science.

What about music? Listen to this clip about the song “Amen Brother”.

Basically, from the sampling of a few seconds of a rather unknown 1969 song, a whole genre of music was born. A spoon-full of copyright infringiment over a bed of creative re-use led to a whole genre of music. In the most basic sense, the birth of hip-hop was basically the same as the evolution of science.

Does the comparison really apply? I would say that it does. Joe A might publish an article with a picture and say, “I have no idea what this means.” Jane B then comes along, realizes what the picture is, scans it from the paper, applies some image transformations to get some features out and say “Ah ah! It is now clear that the large feature in the image from the article of Joe A is …” Remember. FACTS are not copyrightable. EXPRESSIONS OF FACTS are copyrightable. Images are expressions of facts. Scientific images may be argued either way. They’re data, really, we (scientists) wouldn’t differentiate them from raw fact, at least conceptually. But they’re still images. In the modern scientific community, we wouldn’t think twice about this. We might phone up Joe A and say, “Look at this! I’ve figured it out!” Joe A will be happy. And so we’ll publish our new result, with attribution, feeling quite content. Unfortunately, Joe A has signed his copyright away to the international journal of frogs back in 1971. Not too unfortunate, because we don’t really enforce it. In comes PRISM… and I get a little worried. Not that much, but a little. And so I ask, given that we have the technology to solve this problem easily and make our system more efficient, why not?

Now, to give you something to chew on, think about what would happen to science if somehow, we weren’t allowed to re-use other scientists stuff. Think about what would happen if we started suing each others every time re-use occured.

Now, having thought about that for a minute, think about what would have happened to music if every DJ in the 80′s had been sued out of existence?

Consider the approach the mainstream expressive industry is taking to the application of copyright law. The “copyright should last forever and nobody should be allowed to use even a single second of my work withouth paying me” camp. Forward the clip above to the 14:00 mark and listen to what the author considers as the significance of learning the origin of the amen break. When samplers came to light, the industry in place let it slide. They may not have regarded the technology and its uses as viable in any way. Now, copying and sampling possibilities are at everyone’s fingertips. It doesn’t seem they’ll make that “mistake” twice.

Now, what if all musicians agreed to say that “You may reuse portions of my work for the purpose of sampling and creating new expressive works.” How would hip-hop and electronica evolve?

[Update 1: I edited this post to add some links to the PRISM webpage and the Creative Commons website. --CBK]
[Update 2: Fixed link to the youtube video. It seems wordpress' visual editor doesn't quite like the embed tag. --CBK