Difference between revisions of "User:Econterms/Wikiscience proposal draft"

From Wikimedia District of Columbia
Jump to navigation Jump to search
(→‎Features to try in next prototype: info about methods suggested by Janis)
Line 21: Line 21:
 
* Use Use relations between works, perhaps from the list defined by [http://purl.org/spar/cito CITO].
 
* Use Use relations between works, perhaps from the list defined by [http://purl.org/spar/cito CITO].
 
* [http://en.wikipedia/org/wiki/PubChem PubChem]
 
* [http://en.wikipedia/org/wiki/PubChem PubChem]
  +
* JG suggests: for each work, identify funders/affiliations, some of whom are interested parties (e.g. tobacco research) ; learn about [http://www.nlm.nih.gov/pubs/factsheets/medline.html Medline] ; most relevant research will be indexed in PubMed ; define terms, like meta-analysis
  +
* idea: note small samples, note meta-analyses, note "randomized" studies
 
; Eventual objective is to create a strongly attended site, hopefully run by Wikimedia Foundation, using sources which are generally indexed in wikidata.
 
; Eventual objective is to create a strongly attended site, hopefully run by Wikimedia Foundation, using sources which are generally indexed in wikidata.
   

Revision as of 23:02, 21 November 2014

Project plan for science wiki

  • Set up at referata. Find out from Yaron & company if this can be done, freely, cheaply, using the same extensions are as available at AADB and WikiPapers, and how.
  • Use/copy source information from acawiki.
  • Use/copy source information from the Mietchen/Klein/Senate project on Wikisource [1]
  • Note that policies prevent commentary on Wikisource, and allow even neutral annotations only with some effort: [2], [3]. However it would be quite plausible to copy PubMed papers which make it to wikisource to an annotated version of them at subpage /Annotated, then use that for something.
  • Their project has a budget and funding; possibly worth imitating: [4]
  • Contact James Heilman
  • Ask User:Sadads about setting up on referata and helping copy all necessary code to new medical-wiki site. Consider contacting the author of WikiPapers.
  • Ask Yaron Koren about setting up on referata
  • User:Groupuscule examined sources on the safety of eating genetically-modified organisms and some of these may qualify as medical: [5]

Features to try in next prototype

  • Links, and footnotes to and/or from retractionwatch, VIVO, ResearchGate, Academia.edu, Mendeley, PubMed, Wikipedia, Google Scholar, ORCID? ResearcherID? BiomedExperts? acawiki.org? Wikisource? . . . .
  • Data on clinical trials, or common data sets, used by multiple papers. Public site clinicaltrials.gov, example: http://clinicaltrials.gov/ct2/show/study/NCT01076764
  • Demonstrate "B disputes A" links and reports
  • Demonstrate "B uses the same data as A" links and reports
  • Can make use of links to OMIM? A visitor can immediately search OMIM (http://omim.org/) by gene (say, BRCA1) or by disease (say, influenza). In response it lists publications that relate the disease to the gene.
  • Use Use relations between works, perhaps from the list defined by CITO.
  • PubChem
  • JG suggests: for each work, identify funders/affiliations, some of whom are interested parties (e.g. tobacco research) ; learn about Medline ; most relevant research will be indexed in PubMed ; define terms, like meta-analysis
  • idea: note small samples, note meta-analyses, note "randomized" studies
Eventual objective is to create a strongly attended site, hopefully run by Wikimedia Foundation, using sources which are generally indexed in wikidata.

Value added

  • Our site is not trying to be a social network like many of those sites are ; it's directed toward the search for neutrally-demonstrable truths. It should link to and from those sites.

Contacts and reviewers

Get views from User:Sadads, James Heilman, Matt, Max, Daniel, and Yaron. maybe also User:Debivort. Announce the effort on Wiki Medicine.

  • invite Janis Geary of U Alberta, per commons conference

Background

  • We've seen a good bit of frontier work in the realm of wiki/commentary and PubMed.
  • PubMed links are easy.
  • ResearchGate employs 120 people (??!) and has $35 million from investors.[1]
  • drop nonmedical fields from the proposal -- they expand our focus too far
  • Cite the project of Daniel, Matt, and Max, with a footnote to their Wikimania 2014 presentation

Effort required

References

  1. Richard Van Noorden. Scientists and the Social Network. Nature 512 (Aug 2014). 126-129.