Learning From Beethoven: Speeding Up The Exchange Of Scientific Knowledge
from the it's-good-to-share dept
There is a general belief that science proceeds by smooth cycles of discovery and sharing ? that scientists formulate theories, investigate problems, produce data and then publish results for other scientists to check, reproduce and then build on.
That may be the theory, but in practice the frictionless sharing of scientific results is greatly impeded by two factors: the huge profits that scientific publishers make from acting as a tollgate for knowledge through their journals, and the Bayh-Dole Act that encourages educational establishments to try to make money by not freely sharing the discoveries of their academics, but patenting them instead.
Open access arose in part to combat the first of these, and now there’s a new project that wants to build on its achievements by updating scientific knowledge more rapidly:
We want to change the way research is communicated, both amongst researchers, as well as with health practitioners, patients and the wider public. Inspired by Beethoven, we want to build a research version of his repository and try to tackle the question “What if the public scientific record would be updated directly as research proceeds?”
“Inspired by Beethoven” refers to this quotation from one of his letters, written in 1801:
There should be only one repository of art in the world, where an artist would only need to bring his creations in order to take what he needed.
Here’s how this new project hopes to start creating a repository for science:
There are already over 100,000 scholarly articles available online under a Creative Commons Attribution License and thus free for anyone to read, download, copy, distribute, modify and build upon, provided that proper attribution is given. We will start building Beethoven’s open repository by taking 10,000 of these (especially review articles), convert them into a common format, interlink them like topics are linked on Wikipedia, and update them with fresh information as new research findings become available. This will turn the original 10,000 articles into Evolving Review Articles – in other contexts called Living Reviews – available under that same Creative Commons license. We expect that this will help research to be communicated faster, with the ability to promptly correct errors or misconceptions, and in a way that better incorporates the interests of the public. The Evolving Reviews will have a public version history, so that anyone can see in what state the article was at any given time in the past. Over time, this feature can develop into an important tool for exploring the history of science, or of ideas more generally.
Obviously this idea is close to that of wiki-based projects like Wikipedia, and that’s no bad thing, since the format has proved its power in multiple contexts. Interestingly, the people behind what they seem to be calling “Beethoven’s open repository of research” want to write some new software for the job:
we think that Beethoven’s open repository of research should be federated rather than centralized. This means that if you edit a page in the repository, this act will create a personal copy for you. You can decide whether you want to feed these changes back to others, they can decide whether they accept your changes, and there must be options for authorizing certain versions for certain purposes. Such federated systems for the collaborative structuring of knowledge are only just emerging, and producing a working prototype platform that allows anyone to contribute to Beethoven’s open repository is an important milestone in our project. Once the platform is up and running, the 10,000 seed articles will have to be imported, and a selection of them will be used to demo the Evolving Review concept. You can help shape the project by making suggestions as to what topics we should concentrate on. Finally, we want to facilitate the reuse of the Evolving Reviews in contexts outside research, especially in education and in supporting patients.
They’ve launched an appeal for funds using RocketHub, a platform similar to Kickstarter, and are seeking a fairly modest $12,000. But you have to wonder whether that’s really enough for what sounds an interesting but ambitious project.
Follow me @glynmoody on Twitter or identi.ca, and on Google+
Comments on “Learning From Beethoven: Speeding Up The Exchange Of Scientific Knowledge”
It’s not a bad idea, but I don’t think a lot of people realize how research works here.
It used to be get a grant, do some research, get a result, publish.
Now it’s do some do some research, get a result, get a grant to retroactively pay yourself, then publish.
If they published results before the grant money came in, who would pay them?
How about studies with contradictory results?
I have a concern over the updates. What if two (or more) researches have findings that are contradictory to each others? Will there be someone to review and “judge” which one is more likely to be correct, or will both piece of result be added to repository for others to check?
hypothes.is
The world wide web is the repository. That was Tim Berners-Lee’s vision.
You may be interested in hypothes.is which aims to provide open, web-based peer review for pretty much anything.
Note to self:
Add Beethoven to list of piracy supporters.
Re:
Yeah he was a socilaist too – he was awarded a lifetime stipend by the City.
(Worth noting that the 9th Symphony was financed not by copyright on his previous works but by the London Philharmonic Society – which was the Kickstarter of its day!)
How about studies with contradictory results?
As happens with traditional publishing models – both will be published provided they pass a bar of not being obviously wrong. Later work will then resolve the issue one way or the other – or maybe find that both are right – and it is only the idea that they contradict each other that is false.
hypothes.is
In in the world where TBL worked – high energy physics – scientists had already routed around the publishers in the 70’s – we had a system called the preprint system whereby new papers were distibuted around to all the institutions in draft (photocopied) form. Major institutions like SLAC, CERN etc maintained lists of preprints – which were distributed weekly to everyone. There was also a program called STAIRS (on the nascent internet) that enabled you to find older papers.
What TBL did was to mechanise and streamline this system.
Noooo!
Hopefully SOPA/IProtect will put an end to this sort of sharing nonsense.
/sarc
Re:
Beethoven is killing the music industry!
…
Wait, what?
Re:
Beethoven is killing the music industry!
By writing stuff that’s so good that the modern composers can’t compete!
Patent Case?
I can see them getting sued for violating quite a few vaguely-worded, unused patents with this plan.
One central repository
We need a central scientific learning academy so that institutions and research aren’t focused on their prestige but on the research and advancing mankind’s knowledge. Something like our own version of the Vulcan Science Academy.
Re:
If they published results before the grant money came in, who would pay them?
Actually you need to publish before you can get a grant anyway – otherwise no “track record” so no money.
So the sequence is:
Do work – publish – get grant – “enhance work” publish again.
Re:
I stand corrected. I was assuming you started after you developed a reputation in your chosen field.
Re:
Given what I’ve heard from some people it seems to be:
Do some research
get a grant unrelated to your research
continue doing original research using grant money
suddenly realize your grant is about to expire,
quickly slap together something vaguely related to the grant
get the grant renewed
go back to original research
publish original research
It’s a bit of a caricature obviously…
I think this could solve the problem on unpublishable results. I have a friend who does research. He was telling me how after 6 months working on a very promising project that everyone around him was excited about, he randomly mentioned it to some fellow scientist who immediately pulled out a stack of papers a mile high and said something along the lines of: “Yeah, we tried that about 5 years ago. Took us 2 years to figure out it was a dead end.” Since the result was negative, it was unpublishable. I cringe at the number of scientists that must be right now working on something that has already been demonstrated false by somebody else somewhere. Having research come out more in real time could help alleviate that problem.
Re: Re:
that also brings up the issue of the lack of ability to publish negative results. Whilst i can see potentials for abuse, it would be a useful process, and you could prove you were doing /something/.
They've heard of Git?
Seems like it, together with a web server, would take care of 90% of this.
Put a wrapper around Git, perhaps, to make it friendlier, and standardize how to contact authors (and where what each author considers authoritative resides; perhaps Git has already tackled each of those problems, even), and you’re pretty much there.
There's no doubt but what this project - or something like it -
would be an immense benefit to researchers. For that reason it is likely to be banned by the US Congress and its proposers thretened with a one-way trip to Guant?namo….
Henri