User talk:Doc James

From Meta, a Wikimedia project coordination wiki
(Redirected from User talk:Jmh649)
Jump to: navigation, search

IRC office hour for Wikimedia Foundation copyright strategy[edit]

Hi there - thank you for your participation in the copyright strategy discussion so far! In addition to contributing on-wiki, you may be interested in an upcoming IRC office hour the Wikimedia Foundation legal team is holding to discuss the copyright strategy. It will be on September 15 at 14:00 UTC. More information is available on Meta-Wiki. Thanks! Joe Sutherland (WMF) 00:48, 8 September 2016 (UTC)

WikiFactMine project[edit]

Greetings from petermr and the WikiFactMine grant/project. We have just had our initial project meeting with Marti Johnson and she suggested that you and us can work closely together. This would be great - this message is just saying hi, and waiting on Marti to join things up.Petermr (talk) 18:25, 10 November 2016 (UTC)

User:Petermr From what I understand you will be mostly developing software for Wikidata correct? Doc James (talk · contribs · email) 18:40, 10 November 2016 (UTC)
Doc James Yes, that's a large part of it. To increase the amount of useful items in Wikidata and also to promote the value Petermr (talk) 21:14, 10 November 2016 (UTC).
So you are building a system which automatically finds sources for claims already within Wikidata? Or are you building a system that finds claims that could potentially go in Wikidata? Doc James (talk · contribs · email) 21:24, 10 November 2016 (UTC)
I think the first would seem to be more valuable, but - if there is a demand by users - we could trawl for potential claims. Petermr (talk) 21:28, 10 November 2016 (UTC)
Okay. For medical claims one needs to make sure to use high quality references. Low quality references can be used to say nearly anything. Doc James (talk · contribs · email) 21:39, 10 November 2016 (UTC)
Agreed. We wouldn't want to completely automate the ingestion of papers supporting claims. It's more likely that we can create high quality lists that can be presented to experienced editors. I'll be talking with Daniel Mietchen (Q20895785) next week at NIH who is working on the Zika corpus project. I think this is a good archetype for collecting facts.
We are looking mainly for high quality secondary sources rather than high quality primary sources as references for En WP. Doc James (talk · contribs · email) 22:15, 10 November 2016 (UTC)
Fully understood. If you have Open Access sources/reviews that you regularly use we'd be happy to explore those. We'll probably concentrate more on bioscience Petermr (talk) 09:45, 11 November 2016 (UTC)

On the top of all talk pages for disease articles we link to a link to all "Free review articles" on a topic. However not all of those review articles are deemed high enough quality to be a Wikipedia reference. Typically we do not use journals with an impact factor of zero and journal articles which are copied and pasted from Wikipedia (which is happening amazingly often now). Doc James (talk · contribs · email) 17:54, 11 November 2016 (UTC)