I create products for the web.
I am deeply interested in problems of collective behavior and future models of open scientific collaboration.
My work since 2011 has focused on designing and building open source platforms which enable people to visualize and interactively explore high dimensional spaces.
I am @colinmegill on Twitter, and open DMs are the best way to get in touch with me, though I am not able to respond to all inquiries (short messages of a few sentences that clearly identify your organizational affiliation — and a clear ask — are helpful in that regard).
I love backpacking, surfing and parenting 3 boys. I travel frequently and love seeing new places.
Polis is a real-time system for gathering, analyzing and understanding what large groups of people think in their own words, enabled by advanced statistics and machine learning. Polis has been used all over the world by governments, academics, independent media and citizens, and is completely open source.
The development of novel methods for gathering voice, and the development of the platform were a response to communication challenges within large scale decentralized movements evidenced in Occupy Wall Street and the Arab Spring.
Our startup (also called Polis after the technology) open sourced the technology at the 2016 g0v summit in Taiwan, and the organization transitioned to a 501c3 between 2018 and 2020. I transitioned from CEO of the startup to the board of the nonprofit, called The Computational Democracy Project, which stewards the codebase, the main deployment hosted at pol.is, and the mission and vision.
The AGPLv3 licensed codebase can be accessed on GitHub.
There are now multiple deployments of the technology, including at national scale by Taiwan's government and at city scale by the municipality of Amsterdam in the Netherlands.
Case studies and outcomes have been featured in The Atlantic, NYT, Wired, The Guardian, MIT Tech Review and The Economist among others, and have been the subject of a short BBC documentary as well as featured in multiple books from major publishers.
From nextstrain.org: "Nextstrain is an open-source project to harness the scientific and public health potential of pathogen genome data. We provide a continually-updated view of publicly available data alongside powerful analytic and visualization tools for use by the community. Our goal is to aid epidemiological understanding and improve outbreak response."
My personal involvement with Nextstrain began with a weekend evening hack night project in 2014 to pull together real time statistics on the Ebola epidemic. This was very similar, in retrospect, to many of the hack projects that resulted in real time dashboards for COVID.
The idea of visualizing evolution in real time was completely new to me, and I was fortunate and humbled to be able to learn a lot of biology while working on platformitizing, generalizing and expanding the nextflu prototype.
Nextstrain went on to win the Open Science Prize, and contribute to the surveillance of viruses like COVID. I was interviewed by Shirley Wu about the tool and how it's used to monitor virus evolution. Here, as in that video, I will reinforce that I brought the interactive web bits, but did end up learning enough general background to perhaps be an interesting translation layer for non biologists.
Visualize the virus radiating out from China, mutating as it spreads.— balajis.com (@balajis) February 28, 2020
Right: map with different strains color-coded
Left: phylogenetic tree, showing how individual strains mutate and diverge
Very impressive work by @trvrb and the group at Nextstrain. https://t.co/p3wkhvtDPL pic.twitter.com/BuoxBaW81A
The Nextstrain paper was published in Bioinformatics, Volume 34, Issue 23, 01 December 2018, Pages 4121–4123.
The AGPLv3 codebase is available on GitHub.
cellxgene is a highly general, MIT licensed tool for visualizing, exploring, and interactivley running computation on high dimensional data (that is, large, sparce matrices, their features, and their metadata).
cellxgene was built for the purpose of enabling pre-publication (benchtop) data annotation and post publication sharing of the many scRNAseq datasets emerging from The Human Cell Atlas, a project described at length in the 2018 MIT Tech Review article The Cartographer of Cells.
Additionally, cellxgene was used to facilate rapid dissemination of many single cell datasets related to COVID-19 illnesses.
The paper, which describes the purpose, features, system architecture in depth, is available on bioRxiv.
The MIT licensed codebase is available on GitHub.
Above screenshot: four million cells interactively visualized from the Survey of Human Fetal Development.
See also: Mapping Cells To Build a Complete Atlas of the Human Body from The Economist.
Boston, USA (remotely), 2021 May, Internal seminar, MIT Media Lab Center for Constructive Communication Polis: Scaling deliberation by mapping high dimensional opinion spaces (pre-publication paper talk) Video
Oxford, UK (remotely), 2021 April, Internal seminar, Future of Humanity Institute Centre for the Governance of AI, Oxford University Polis: Scaling deliberation by mapping high dimensional opinion spaces (pre-publication paper talk)
Venice, Italy (remotely), 2020 December, International Forum on Digital Democracy A CASE STUDY ON COMPUTER-AIDED POLICYMAKING
Stanford University, California, USA, 2019 March, Center for Advanced Study in the Behavioral Sciences Machine Learning and Democracy
Amsterdam, Netherlands, 2018 December, Netherlands Institue for Multiparty Democracy Annual Conference AI and the political party
Aptos, California, USA, 2018 April, CZI Science Computational Biology RFA Gathering A Modern Toolkit for Scientific Data Visualization
San Francisco, USA, 2017 October, Foo Camp 2017 AI and the political party
Mountain View, USA, 2017 March, Ed Foo
Washington D.C., USA, 2017 January, Internews An introduction to pol.is
Naples, Italy, 2016 November, School of Civic Tech AI & the future of participatory democracy
Turin, Italy, 2016 November, School of Civic Tech AI & the future of participatory democracy
Sebastopol, USA, 2016 June, Foo Camp 2016
Jeju Island, Korea, 2015 December, By the Crowd New methods in participatory democracy
Boston, USA, 2015 November, Future of Web Apps Victory.js, A powerful data visualization library for ReactJS Deck
NYC, USA, 2015 June, QCon Data visualization in ReactJS
Las Vegas, USA, 2015 June, Future of Web Apps RadiumJS & VictoryJS: Data Visualization in ReactJS
Seattle, USA, 2015 April, SeattleJS @ Facebook Learn React and Flux: Part III Video
Seattle, USA, 2015 March, SeattleJS @ Facebook Learn React and Flux: Part II Video
Seattle, USA, 2015 February, SeattleJS @ Facebook Learn React and Flux: Part I Video
The Death of the Gods: The New Global Power Grab by Carl Miller. 23 August 2018 ISBN13—978-1785151330
WTF?: What's the Future and Why It's Up to Us by Tim O'Reilly. 10 October 2017 ISBN13—978-0062565716
To save democracy, we must disrupt it
UnHerd 03 September 2018
The simple but ingenious system Taiwan uses to crowdsource its laws
MIT Technology Review 21 August 2018
Taiwan’s revolutionary hackers are forking the government
Wired 4 May 2018
Women's March anniversary sets the stage for new political campaign with a tech twist
Mashable 19 January 2018
The Internet Doesn’t Have to Be Bad for Democracy
MIT Technology Review 2 June 2017
VTAIWAN: PUBLIC PARTICIPATION METHODS ON THE CYBERPUNK FRONTIER OF DEMOCRACY
Civicist 11 August 2016
Changing the relationship between citizens and government
Daum (Korean) 2 March 2016
Blooming digital democracy in Taiwan’s Sunflower movement
Mobilization Lab 25 January 2016
New Community Tools: Polis
Coral Project 9 December 2015