But when discussing “the future of science” then I guess it should be allowed to assume that science could select for something else. In my view what you describe is not really “science” selecting. But rather the old-world institutions around science. The “nature-paper” credit of now, is a relic of when dead-tree publishing was a physical necessity. And the hirering based on nature paper authorship is a relic of the limited information available in a dead-tree limited world.

In my view this is why the brainstorming around openness, credit, timestamping, incremental contribution etc. is important. The goal is not to make open science work inside a world limited by dead-tree distribution of information. The goal is to think up a system that takes advantage of digital information to improve how science is done. Right?

Delusions of grandeur do indeed lead to assholery, but then again, most of the great scientists I’ve met (and I’ve worked directly with half a dozen or so nobel prize winners) have pretty big egos (and had them before they won the prize). Science doesn’t select for friendliness or personality, it selects for achievement, and being a driven jerk is a selective advantage. It would be a lovely world if everyone made cupcakes for one another, but I don’t see it happening any time soon.

What is the incentive not to share credit? Well, funding and jobs are limited commodities. The more credit one has for oneself, the more of those things one is likely to garner. There’s also the matter of being right, being the actual one to make a discovery and getting the credit for that. In the case we’re discussing, Franklin is rightly credited with collecting a key piece of data but not with discovering the double helix because she didn’t discover it.

Never heard this, and it would change my view of the situation considerably; can you point me to a source? (Pre-emptive: Watson’s book doesn’t count!)

Would you rather be a great scientist or a cog in the machine?

False dichotomy. Like 99.9999999% of everybody, including scientists, I’m muddling around in the middle somewhere; choosing to be “great” doesn’t enter into it. Delusions of intellectual grandeur are behind most of the type-A assholery I see in science.

When you talk about getting credit, it sounds like you’re thinking about discovering the double helix, when the vast majority of advances are incremental. We’re not talking about seeing photo 51B and choosing to bogart the Nobel, we’re talking about making some pedestrian addition to the knowledge base and choosing not to be an asshole about it. What I don’t get is, what’s the incentive NOT to share credit?

Good points though on trying to introduce tools where there's no obvious need. There's a nice recent set of articles on this subject here, here, and here.

deepak–
Again, I’m not so sure how happy I would be if I were shuttering my own lab while someone else was using my data and winning funding and prizes. I just may not be as noble as you, but then again, I’ve got a family to feed.

Bill–
Don’t get me wrong, I’m a big fan of collaboration–the majority of the papers I published in my non-illustrious career at the bench were collaborative efforts with other labs. But I do think there needs to be some balance. I think one needs to exploit one’s own work first before turning it over to the rest of the world. I don’t see anything wrong with collecting enough data to write a paper, then opening things up to everyone. Sure, it may take a little longer, but it keeps people employed and rewards them for their hard work. And one should collaborate, but there’s a big difference between sharing data amongst collaborators and throwing the doors open to everyone before you’ve thoroughly analyzed your own results.

Perhaps “smarter” is a loaded word. Let’s instead say “someone who had the insight that you missed/lacked at that particular moment.” Again, Franklin was offered coauthorship as I recall, but chose instead to let credit fall to those who had the insight she lacked at the time. My opinion is that if she continued on, and Watson and Crick hadn’t seen her data, perhaps she would have reached that insight herself, and been able to rightfully claim the credit for both. The world got the double helix a little faster, and she became a footnote, rather than getting the spotlight. Good for the world, good for Watson and Crick, not so good for Franklin.

And I don’t think people are wrong in looking out for themselves, protecting their careers, their labs, their employees and students. Would you rather be a great scientist or a cog in the machine? Most people I know would choose the best chance at greatness.

Also, there are always going to be people who look to take advantage of the system, particularly an open one–that’s human nature. Take a look at the gaming that’s going on with arXiv. Not everyone is going to play fair, and Open Science gives advantage to the unscrupulous, and to big labs that can afford the manpower to quickly replicate others’ data. Timestamp or no, one could always claim never to have seen the grad student in question’s open notebook (probably likely, given the flood of information which would occur if everyone suddenly went open and online), and that it was just coincidental that they were doing the same experiments at the same time. This happens all the time in the journal biz–lab head A gets sent a paper by lab head B to peer review, and realizes they’re about to get scooped and quickly puts out their own paper while delaying the review process.

You guys may just have more faith in human nature than I do. But every scientist I’ve mentioned the idea to scoffs at it. They want to keep their employees and students working, they want to keep the lights on and keep their families fed. This probably slows the course of science, but then again, it makes it a viable career.

timestamping won’t help in this circumstance, unless the point you want to prove is that you did the legwork and someone else smarter than you figured out what it meant first.

Why is it *necessarily* someone smarter than me? In my experience of the reverse situation, when I realize what someone else’s data mean before they do, it’s simply because I happen to know something they don’t — usually an odd, unexpected connection with my own work. It’s got nothing to do with being smarter — just as it would be ridiculous to claim that Watson and Crick demonstrated by their actions that they were *smarter* than Franklin.

Confronted with someone else’s data and the realization that you know something about it that they don’t, you have a choice: you can share, or you can try to grab as much of the credit as possible. How are you harmed by sharing? If the data are the Big Deal, you ride the coat-tails; if your intellectual leap is the crucial factor, you get the spotlight.

To put it another way, since Watson/Crick/Franklin is “the classic example”: should Rosalind Franklin not have been included as an author on that famous paper? And in what way would Watson and Crick have suffered if she had been?

David, you seem to agree that the general idea is sound — that collaboration is more efficient and productive than dog-eat-dog competition. Would you consider, then, opening up some aspect/s of your research, while keeping closed enough of your ideas that your family would not starve? Perhaps if enough people would dip a cautious toe into Open waters, the kind of cultural change Shirley describes could begin to occur.

We differ in philosophy. I actually would love if someone did something cool with data that I collected. All power to them.

Alternatively, we need to figure out how to apply copy left licenses, but locking it up for your eyes only is limiting yourself to the confines of a broken system.