User:OpenScientist/Open grant writing/Encyclopaedia of original research

From Wikiversity
Jump to: navigation, search
There should be only one repository of art in the world, to which the artist would donate his works in order to take what he would need

Beethoven.

We are trying to build such a repository for scientific research, and to design it such that non-scientists can benefit from it as well.

An animated description of what open research is about, featuring Beethoven's repository idea.

This page is part of the Getting your CC project funded course at Peer 2 Peer University. It is managed by the team of @Kubke and @EvoMRI. We want to draft, in the open, a strong grant proposal that is CC-licensed and intended to be reused. The topic: An encyclopaedia of (or GitHub for) original research. Of course, any interested party is invited to join in.

A wiki version of Rembrandt's The Anatomy Lesson of Dr. Nicolaes Tulp, by Alasdair Forrest. What if the scientific record would be updated directly as research proceeds?


Overview[edit]

A basic outline of the course and of our project within it
Idea 1: An encyclopaedia of original research - a collaboratively editable database of interlinked and semantically enhanced openly licensed scholarly articles, be they from OA journals, open notebooks, blogs or the legacy literature that has entered the public domain (this follows up on an earlier proposal along these lines).
Idea 2: Delivery of scientific and health-related information to remote areas. Based on Idea 1.
Idea 3: Exploring the potential of databases as museums of the future. Could be based on Idea 1 or not.
Idea 4: Testing the efficiency of open vs. traditional science. Broadened in scope with respect to Idea 1.
A more detailed characterization of Idea 1 in a standardized format suitable for funding requests.
Briefly introduces the plan of drafting the TeX documents via a dedicated GitHub repository, invites comments on how to find suitable funders
See also the dedicated section below
On ideas 4 and 3 and illustrations, with a suggestion for a crowd-sourced environmental sensing project on the basis of OLPC
Provides an Open Research perspective on the SciFund Challenge and lays out some of the rationale behind our proposal, which participated in the challenge (detailed comments: Creative Commons New Zealand/ Aotearoa, Rory Macneil, Mary Canady, Jarrett Byrnes, Mike Linksvayer)

Current status[edit]

The project has received some crowdfunding in 2011, but this was by far not enough to get things going, so that money remains parked at the University of Auckland until an opportunity comes up to put it to use in the context of something that would fit under the umbrella of an Encyclopaedia of original research or a GitHub for research.

Drafting the proposal on GitHub[edit]

We have decided to use GitHub for drafting our actual proposal (for some earlier brainstorming, see here). The git repository is at https://github.com/Daniel-Mietchen/Open-Research-Proposals (see also Ohloh listing), and the latest PDF version can be downloaded by clicking on the "view raw" link here.

How to get involved[edit]

Getting involved with the proposal drafting[edit]

  • For each of the individual project ideas (outlined in post 2; details added in post 3), you could provide:
    • Feedback on the subject
    • Feedback on related projects that have already been drafted or even funded or finished, or are in the planning stages
    • Feedback on potential partners (individuals, institutions, organizations)
    • Feedback on budget
    • Feedback on potential funders (see also the dedicated section in the draft)
  • For the whole series, you could
    • Spread the word (some inspiration); see also Feedback
    • Provide feedback on online platforms suitable for collaborative writing of this kind
    • Help in drafting or illustrating (the proposal itself, or the blog posts about it)
    • Donate ideas, money, or infrastructure
    • Propose other improvements
    • Chime in on specific requests for help, as listed in the next section

Coding issues[edit]

The issues listed here are essential for the project, once funded, but nice-to-haves during the drafting phase. The more of them we have solved, the stronger our proposal will be, as we can present them as preliminary data or proof of principle. Neither of us are programmers, so if you think either of these issues are easy to solve, we would be glad to hear how.

  • How to parse Google search results to get the PMCID of the listed articles? (some more background)
I don't think going through Google is required, since you could use the PubMedCentral OAI-PMH endpoint that was cited in the WiredScience article to basically fetch all the records (using the ListRecords OAI verb, limiting to records in the pmc-open set), and then filter based on the <license> element right? For example this will get you the first response, and you would need to follow the resumption tokens to get the entire set. I could help pull this content if you need any assistance. Edsu 10:26, 1 August 2011 (UTC)
Good to know that there is a more direct way than going through Google - I wasn't aware of that. I suspect, though, that the <license> tag is not used consistently across (and may be even within) publishers, so if you could check that somehow, this would be an important initial step already. If it is - or at least for certain publishers - then that paves the way for a tool based on that information. In the end, we need a list of PMCIDs (as at http://www.science3point0.com/coaspedia/index.php/Technical:XML_for_test_purposes) on the basis of which to do the xml2wiki conversion. Any move forward in this direction would be of help. Thanks! --Daniel Mietchen (talk) 22:30, 1 August 2011 (UTC)
It looks like <license> isn't used in all the PMC metadata records but it is the NLM DTD article XML, so it may cut across publishers. I put together a quick python script for walking through all the records, and printing out the identifier and license URL (if found). Edsu 01:48, 2 August 2011 (UTC)
Probably best to start with the JATS Preview Stylesheets. Klortho 12:36, 20 May 2011 (UTC)
Yes, that looks good. Thanks! --Daniel Mietchen (talk) 01:03, 21 May 2011 (UTC)
We are currently exploring possibilities to import into GitHub rather than MediaWiki. --Daniel Mietchen (talk) 07:31, 20 July 2011 (UTC)

Illustrations[edit]

Moved to dedicated subpage at User:OpenScientist/Open grant writing - Encyclopaedia of original research/Illustrations.

Getting involved with the Encyclopaedia of original research[edit]

We currently do not yet have the Encyclopaedia of original research running as such, but prototypes exist:

Ideally, we would like the encyclopaedia to exist in the most malleable format currently available for larger cohorts of scholarly articles (XML), and to combine this with an XML viewer that allows for public version history and easy forkability.

What next?[edit]

Independent of whether this current proposal ends up being funded or not[edit]

The blog posts produced in the framework of this project, as well as the text of the grant proposal itself, are put under a CC0 license, so as to maximize the potential for reuse.

We foresee several reuse scenarios:

  1. Research: our projects 1 and 4 are likely to advance science, irrespective of whether we get funded to implement them or someone else
  2. Education: newcomers to grant writing might find it helpful to have the whole process of writing a grant documented in public; funders might use it as an example for pointing out aspects of a strong or weak proposal
  3. Outreach: all four projects, but especially projects 2 and 3, will help to intensify the exchange between researchers and the public
  4. Science policy: data about open science is scarce, so our project shall serve as one of the pieces of evidence submitted to the Royal Society who are taking a look at "Science as a Public Enterprise" (comments invited "as soon as possible, and before the 5 August 2011")

If the proposal gets funded[edit]

We will then work on implementing it, e.g. on building an environment that hosts open scholarly content, makes it editable, proveds a public version history and is easily forkable.

If the proposal does not get funded[edit]

We will probably work on the implementation nonetheless, but at considerably lower speed. Plus, we would probably try other avenues for getting the project funded, and others are invited to try and get it funded themselves, with or without us.

Feedback[edit]

So far, we received little feedback, but what we got was encouraging. Some examples are listed below:

See also[edit]

Paper on similar approach