You're completely missing the idea of copyright as applied to a paper.
Please do enlighten us!
The work that is copyrighted is the paper itself, not the research described within. You can build your research on the results of others absolutely normally - and that's what Newton meant. Read, do research based on it, write a new paper.
Sure, writing new papers is one possibliity. But it's not the only way to collaborate.
Fair use makes it even possible to cite the parts you want to discuss, if necessary. I can't really think of anything more you could ask for, anything "more free".
Fair use is a defense, not a right. It's not perfect, and it can be messy to defend in court (or so the lawyers tell me). Why wouldn't some scientists want to make their research and their ideas *even more* accessible to those who wish to remix them and/or use short/long excerpts?
On the other hand, building your paper on someone else's paper by just modyfing the relevant parts is not in any way helpful for science - and that's the definition of derivative work here.
What are the "relevant parts"? And are you certain that no useful science can come from borrowing data/charts/text from an existing paper beyond what Fair use would allow?
Example: I write a 100 page paper about monkey butts. Let's say it's about hair color and distribution. I include a lot of really useful pictures of monkey butts and graphs of the statistics thereof.
Someone else wants to write a new paper about monkey butts and realizes that they could use all of my existing charts and pictures as their data and come to some new and very interesting conclusions. Their paper is about cancerous growths, and it's not a critique of my paper, so they can't just suck up my data, pictures, and prepared charts and claim Fair use.
If my paper were under CC-BY-ND, they would probably be SOL. Let's say that I had since died (bad monkey stew), so they couldn't ask me for permission.
If, however, my paper were under CC-BY-SA, then they would be able to re-use all of my lovely scientific shit.
In fact, if you do something like this, you're a lousy, lazy scientist - if you can fit your results into an existing paper like that, you probably haven't done anything new and worth reading.
Or maybe the original scientist missed something? Or maybe there's a new, novel way to use his reasearch and copyrighted materials? (see above)
ND-free licenses are extremely useful for code, potentially useful in art, but worthless in science. There's no value added here.
When someone talks about "value-added" in terms of science, it raises a red flag for me. Why does science have to be about "adding value"? Sure, there are a lot of people and a lot of companies that do a lot of hard science, and bully for them. But I don't have to argue that all of them should eschew the ND clause, I just have to show that some scientists have valid reasons for avoiding it.
Seriously? You consider only completely open or completely closed position as non-hypocritical? What you're saying is just pseudophilosophical mumbo-jumbo, based on a fundamentalist understanding of "information wants to be free".
Any half-open or half-closed position seems like a very hard position to argue. Bright lines, as any lawyer will tell you, are easy: Everything falls neatly to one side or the other. As I stated above in my answer, perhaps the "No Derivatives" clause actually legally accomplishes something quite different from its generic meaning in everyday English, and so some classes of "derivative works" of a paper marked ND are actually allowed. My point is that it is sad that the term is named "No Derivatives" when a scientist seems interested (if not intent) upon sharing their work with the world at large via an "Open Access" site.
What is sounds like everyone wants is a "No Defamation" clause, but that's just too darn hard to write in legal terms, so they just ratcheted down the rights and ended up with "No Derivs".
Is[sic] you believe that ["You consider only completely open or completely closed position as non-hypocritical"], peer reviewed papers should not exist - you should publish everything, whether or not competent peers think it's utter BS.
Whaaa? I'm not following your logic at all. Who is "publish[ing] everything" ? And (perhaps more importantly) who is listening to the quacks/frauds who republish "utter BS" ?
Are you thinking of a different type of peer-review process? Peer review, as I understand it, has absolutely nothing to do with licensing, copyright, patents, or any legal standing. Under a peer-review process, works such as papers, research, (possibly art/music, too?) are evaluated by one's peers and judged for their correctness or competency. Things such as language and originallity may also be judged.
Wikipedia describes peer review: "Peer review methods are employed to maintain standards, improve performance and provide credibility. In academia peer review is often used to determine an academic paper's suitability for publication."
Sorry, but this does not work for stuff as specialized as scientific papers.
I don't get it: What magical sauce is embedded in scientific papers? Why should authors of these works use the ND clause, while authors of software, art, novels, etc... allow derivatives? Is there something special about having a doctorate? Something about the way in which the scientific community can't think up any mechanism to vette or review each other's papers and provide structure and confirmation when they run across a paper written by someone they haven't met?
But we have that system already: Peer review (and, perhaps, a sprinkling of some trust algorithms)
Anyhow, I return to my central point: I am ready to be convinced that scienticsts need the ND clause to protect their work -- I just haven't seen or heard of any cases in which this kind of clause is necessary. Please show me an example of the harm that not having an ND clause has wrought, and I will certainly rethink my position! :-)