How should humanity collaborate on answering open questions?

From QuestionWiki
Jump to navigation Jump to search

[note: this question is far too broad. It's tough to precisely define where I wanted to go with this question, so the wording may change in the future]


Here are some of the steps involved in trying to answer a question that could benefit from open collaboration:

  • coming up with the right question
  • listing related questions
  • listing relevant sources
  • making sense of those relevant sources
  • speculating on possible answers to the question
  • designing experiments, studies, or proofs that could potentially answer the question
  • mapping out the logical space of already proposed answers
  • making lists of people or organizations that may have relevant knowledge
  • coordinating replication attempts
  • finding adversarial collaborators

All these things happen mostly in silos with small teams of researchers. Which is a shame. We should be collaborating on all of it.

Why are we not collaborating more on answering open questions?[edit | edit source]

Maybe the biggest reason why we aren’t collaborating on all of these things is a bad incentive structure.

  1. Academics have nothing to gain from contributing to something like an open research projects.
  2. Academics are too busy spending all their time trying to publish in academic journals to contribute to open research projects.
  3. Not only do they not have positive incentives to contribute to open research, they actually have a disincentive. Academics often hoard their ideas, data, code, and even interesting questions because academia is essentially competitive. If someone else publishes your idea before you, you don't get the prestige of being the first person to present the idea. So academics have a disincentive to share their most interesting work before it's published.

How should we handle collaborative citation management?[edit | edit source]

  • One possible way is to just past links to articles in a wiki (like this one), or a google doc, like here.There are a few problems with these approaches. But they generally amount to lack of structure. There are many ways you may want to collect and sort citations. Text documents do well to put sources into fixed categories. But sometimes sources might fit into more than one category. You can copy/past them into both categories, but then if you make notes on one, you'll have to remember to make notes on the other. A tagging system seems to make more sense, where you have one entry per source, but can tag each source with multiple categories.

It would be nice if we could have the citation management well integrated with the wiki itself, but so far that doesn't seem possible. Maybe their's a good plugin for that? Ideally citations would at least be in some sortable table where you could sort by date/author/subjects etc.

Possibly relevant sources[edit | edit source]

Open Science Monitoring – Impact Case Study – Polymath

Factored Cognition

https://michaelnielsen.org/

https://michaelnielsen.org/blog/shirkys-law-and-why-most-social-software-fails/

https://www.lesswrong.com/posts/whXFkYEDiDk8tSBqH/open-problems-create-paradigms

Open Problems in Archipelago

Why is so much discussion happening in private Google Docs?

https://www.mdpi.com/2304-6775/4/2/17/html

There appears to be a generalized trend towards more open access to science, together with citizen science (where citizens act as volunteers in gathering data etc.) and collaboration initiatives--for me this shows that your openquestion wiki might be very much to the point and on time.

This one is interesting! it goes a bit deeper and higher in terms of collaboration in science (e.g. European Union, sharing of research data etc.), but it mentions some common ideas to this project (on infrastructure, on the need for incentives etc.) and I think there's a lot of ideas there on knowledge sharing!

https://indico.egi.eu/event/2452/contributions/4778/

it's from here

https://www.egi.eu/open-science-commons/

https://www.hiig.de/en/the-great-potential-of-citizen-science/  this one is about citizen science--it's not the same as what we're doing, but it does touch on the polymath program etc

https://www.zooniverse.org/  I wonder if something like that (which is awesome!) could at some point work together with an openquestions wiki (when further research is needed for a specific question, this could provide access to many many volunteers)

https://riojournal.com/articles.php?journal_name=rio&id=7573

This one discusses wiki and how it can work as a virtual research environment--I skimmed through it, so I’m not sure how relevant it is, but it looks pretty relevant. It talks about a more scientific research, instead of just literary reviews etc, but why not? The purpose of openquestion.wiki is to get research done, so if many scientists get involved who can actually do research on e.g. chemical reactions, why not have something more elaborate?

https://emckiernan.wordpress.com/2015/11/30/why-open-research/?utm_content=buffer57f23&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer on a project about how researchers would benefit from sharing their work--against individualism and towards a shared communal science!

https://www.biorxiv.org/content/10.1101/031971v1.full

I think this one might be the most relative of all! On how wikidata can be used for collaborative research etc.

https://wiki.dbpedia.org/ Perhaps the community here could be of help? In the “Develop” section.

https://osf.io/

This is another open access research platform--looks more ‘scientific’ and less layman, but could be still interesting to check out.

Related questions[edit | edit source]

Where should we collaborate on open questions?

How do we solve the incentive problem?

Is collaboration always useful when answering open questions, or is it sometimes better for people to work in silos?