Hi Election Hacking Attendees! We'd love to keep in touch, follow up on your ideas, and create a long tail towards building new things. Two captures we'd love: 1) Your Group: What did you work on and where did you get? 2) Your Group members names/e-mail/twitter THANKS SO MUCH. THIS HAS BEEN AWESOME ****** FORMATTING EXAMPLE: Your Group ---------------------- What you're working on: Where did you get: Who's involved: ******* Citizen Voices: HATE MAP: People hate politics, build trust and information flow. 1. where are you from? 2. What do you hate about politics? 3. information you should know 4. start a conversation on that topic Your Group ---------------------- What you're working on: We worked on connecting communities/citizen voices to media? What does the electorate want? How do media organizations use and verify this information. How to connect hyper local (Polygons) to topics/metadata that matters to citizens but can be useful to media? What do the electorants want? Inquiry: HOW DO WE GET VIEWPOINTS ON VIEWPOINTS? Where did you get: We created software requirements. Use Liquid democracy type go Location (Polygons) + Topics/Categories leads to FEED/Link Content provided could be a. original data b. survey/opinion c. verification/rating There are 4 main interest groups that use/need this information a. citizens b. media c. govt d. interest groups/ companies How can editors rate/verify and get more in depth with citizen voice -model in depth vs. "citizen eye" -what is the metatag of a source of a source models could be Public Interest Network or Liquid Democracy Liquid Democracy - open discussion and assessment of voting to topics Public In Network - trusted expert network -provide incentive - how this will effect you and how has this (issues) performed in this post -issue of bias -personal vs community good: why do you vote? -Citizen voice is "apathic" or "bias" = why journalists/edtiors don't trust or use -ignoring youth issues - do you want youth to vote? -trust and corruption issues -lack of knowledge of political process -biased Who's involved: Two journalists, one student, one banker and one idea hacker/mapper 1) Your Group: What did you work on and where did you get? 2) Your Group members names/e-mail/twitter Grainne O'Brien grainneob@gmail.com - @grainneob Heather Leson hleson@ushahidi.com - @heatherleson Rob Bole rrbole@bbg.gov @rbole Stephen Abbott stephen.abbott@guardian.co.uk @stephen_abbott Finlay Craig Finc@me.com @finlaycraig BETTER DEMOCRACY - MOBILE DESIGN ---------------------- What you're working on: Mobile-first stratagies for providing election information to promote democracy and citizenship (over "news") - specifically in Kenya and mobile meaning SMS feature phones Where did you get: An outline of strategies for reaching SMS feature phone users and providing them critical information such as: * Where do I vote? * Who can I vote for? * What positions are up for election? * How do I register to vote? Also, a list of challenges of this effort Who's involved: * Jessica Lord (@jllord) * Brian Boyer (@brianboyer) * Matias Attwell (@tutatis) * Marcus Asplund (@marcusasplund) * Linda Kamau (@lkamau) ******* BEAT THE HORSE RACE Improving Your Information Diet / Towards A Slow Politics Movement Horse race: politics as competition, not democracy. Not helpful. Complicated: journalists, voters and politicians Public asks for more info, less fluff. This is your daily diet. Bryan White (@bwwhite), Mind Candy Joe Germuska (@joegermuska), Chicago Tribune Larry Birnbaum, Knight Lab Nortwestern Mike Tigas (@mtigas), Open News fellow Jaap Meijers (@tjaap / @twisst), freelance journalist * Goal is to encourage readers to reflect on what political info they consume, and nudge the toward a balanced news diet * Browser plugin * server-side text analysis to determine what type of content the article contains * Immediate visual feedback of what types of political information you consume * graph history of your diet by day / week / all time * Periodic email of your diet, calls to action to be informed ******* 1) How many orgs made maps and results lists: redundancy Why do we do the same things and not have time to be journalists? Currently, development for elections is too expensive and time-consuming. If we made this easier, we may have time to do analysis and context... An analogy: Political pundits *should* look at Nate Silver and think to themselves, "hey, I don't have to worry about covering the horse race, this frees me up to focus on context and analysis." That's what we're going for here for news devs leading up to elections -- take care of the maps, charts and lists that are part of obligatory coverage, freeing them up to do context and analysis. For example: Do maps lie? Well, if we don't have to spend so much time building the maps in the first place, maybe we can provide the context necessary for readers to understand what they're looking at. The Ideal World: Centralized open data source from the government normalized, updated in real time The Actual World: Tier 1: Platform access to data through API, middleware admin Tier 2: Libraries for patterns & UI, customizable widgets, educational training, platform plugins Action plan: * open callout for libraries and tech contributions (Actual World, Tier 2) * someone agrees to be open source project hub -- Knight Lab? IRE/NICAR? OpenNews? * this person manages communcation and helps define the community * this group does education and adoption, facilitates future contributions Phase 1: pattern libraries, widgets etc This gives us an opportunity to create best-of-breed visualizations, so readers in Walla Walla, Washington, have access to the same quality as readers of the New York Times. We can also bring in people with dataviz expertise, to help mitigate problems like misleading maps. We can handle maps, charts and lists in a way that's platform-agnostic, responsively designed, and capable of being styled to match a site's preferred palette. If a news org wants to differentiate, they're still free to develop their own, but we're providing a baseline here and encouraging devs to work on other tools for understanding elections. Phase 2: API between existing data (what phase 1 presentation tools need) Some of this already exists. The LA Times released python-elections, which is an open-source Python library for interacting with AP election data. We don't think it would be hard to get news developers to contribute their time to build API libraries for Ruby, PHP, etc. etc. This is work that only has to be done once per data source, and provides a translation layer that provides the pattern libraries what they need for display, but can evolve if the data source ever changes. Phase 3: Ideal world utopia, swap out data sets This is a big, BIG idea, and would require a requisite amount of effort. The goal is opening elections data and making it free to everyone, rather than relying on AP as the sole provider. This would take something like Google getting involved. Or alternatively, creating the system for states and municipalities to enter election results in the first place, so we can normalize that process -- a lot of conversations would have to take place to make this happen, of course, but the potential is there to build a better system than the one everyone has to use right now. 2) Miranda Mulligan @mirandamulligan Aron Pilhofer @pilhofer Ryan Pitts @ryanpitts Brian Abelson @brianabelson Dan Hill @danhillreports Noah Veltman @veltman Marianne Bouchart @Maid_Marianne Martin Stabe @martinstabe Ændrew Rininsland @aendrew ******* Depricating the AP/Exit Polls: Problem: Getting AP's elections result data is hard (though getting easier, see above) and expensive (~60,000 for access to everything). These costs prohibit smaller news organizations and citizens from generating their own election coverage. Question: Can we reverse engineer the data-collection process for election results / exit polls? If so, how accurate would it be? Methods: * Generate reporting tool (mobile web app) which communicates with a server, accessible via API * Recruit volunteers / paid data collectors