Sunday, November 20, 2011

Introducing Scientifiqa, the science-based Q&A site

I just launched Scientifiqa, yet another general Q&A site (and stackexchange clone). What makes this one different? Every answer must be a citation (and summary) of a peer-reviewed academic article or survey. The hope is that this requirement will ensure high quality discussions and give people an easy and quick way to come up to speed on current scientific understanding in a particular area.

Some questions already posted:


Or ask your own question.

Please share any suggestions or feedback!

Monday, October 3, 2011

A better arXiv

Imagine if arXiv was a fully functional online community.

Here’s one vision:


Member profiles 
The community would be open to anyone (possibly contingent upon approval by existing members.) Profiles would hook into Linkedin, Mendeley, Researchgate, etc to pull in your education, institutions and publications.

Reputation
Every member of the community would have a (private) reputation score based on their “impact” in their field. All actions on the site would be weighted by a member’s reputation, it’s calculation based on:
  • published papers (traditional impact scores)
  • education, employment and institutions (verified by community)
  • actions within the arXiv community (peer reviews, upvoted comments, well-rated submissions)
This would have to be handled openly and with care, but would help keep quality high and entice well-respected academics to engage in the forum.
Peer review
Members could write and request (public or private) reviews of submissions. This would include “rating” the submission across various dimensions. Possible dimensions:
  • methodology
  • novelty
  • expected impact
Comments
There would also be an outlet for less formal discussion, with upvoting and threading.
Article discovery
There would be support for sorting and filtering submissions by ratings and reviews. There could be a published “journal” of best articles every month, possibly selected by weighted vote, if people felt the need for a formal publication.
In addition there would be a personalized feed of submissions, based on topic filters and a recommender system.
Open access
Most importantly, ANYONE could read the output of this community. Even the taxpayer/college student who funded it all...
In addition to these community features, there could be added capabilities for submitting papers. This could be simple things like support for attaching code, data or arbitrary media; or a more significant overhaul of a “paper”, turning it into what is essentially a web page, with inline links for references, expandable sections, embedded media, etc. Unfortunately, there is no standardized typesetting solution for the web yet, so some work might be required to make this happen.

If this got built and people adopted it, I think it could deliver a swift, fatal blow to the academic publishing industry -- something it desperately needs.

We built science.io with this end vision in mind. If you are involved with arXiv or are interested in making this or something like it happen, I’d love to chat. Get in touch kvh@science.io