Crowdsourced accountability project: Progress, but we still need help from developers

by on December 14, 2008 · 34 comments

Mayor's Project Napkin SKetchUPDATE: I’ve created a Google Group for this project. I hope you’ll join it and help us build this tool.

Last Thursday I asked for help creating a site that would facilitate crowdsourcing the task of prioritizing the 11,000+ projects proposed in the U.S. Conference of Mayors’ $73 billion “Main Street Economic Recovery” stimulus plan. The point of doing this is to help President-Elect Obama keep his promise that any stimulus spending will be directed at critical infrastructure, and not pork. Roads and bridges and schoolhouses are infrastructure, but dog parks and tennis centers in wealthy neighborhoods probably don’t count.

Software developer Kevin Dwyer stepped up to the plate, took the mayors’ report, and parsed out the projects into an SQLite database. You can find the database here and Kevin’s take on what he did here. Now that we have the data in an easy-to-remix format, I’d like to ask for your help developing the backend for the site.

Going forward I can offer graphic design and copywriting to the project, as well as cat-hearding, which are my comparative advantages. What I don’t have are the technical chops to code the backend. If you are a developer, or know someone who is, and might be willing to help, please read on.

The functionality I’d like the site to have is a lot like what WashingtonWatch.com offers. Each proposed project would have its own item page. That item page would list the project name, city and state, cost, and estimated number of jobs it would create, all of which are included in the database. Then each item page would have a wiki section where users could write (hopefully) neutral POV descriptions of the projects to put them in context. Under that there would be a comments section where users could trade their opinions on the merits of the project. (Perhaps these could be threaded, maybe using Disqus.) Finally, and importantly, each item page will have an up-or-down voting mechanism that will let users register whether they think the project is critical infrastructure or not critical infrastructure. This voting is what will let us rank projects from critical to porcine.

(Here is a sketch of the interface with notes. It’s a lot like an individual bill page on Washington Watch.)

Now, apart from search functionality, we would have to offer easy browsing for folks to find projects that interest them. I think the home page should offer a link to search by city and state. Clicking on that link should offer a list of states. Clicking on a state should offer links to all cities in that state, as well as a list of all projects in that state in case a user doesn’t want to drill-down any further. Clicking on a city will display a list of all projects in that city.

Each of these lists I’m describing should be sortable by name, locality, cost, and estimated number of jobs created. That way someone can click on a state then sort by cost so they can see the most costly projects first. On the home page there should also be a way to browse all projects in the country ordered by cost. Another possible sorting option might be how a project is ranked by users.

So what do you think? I’d love to hear any thoughts or criticism you might have on this proposed interface. I’d also love to hear any ideas of the best way to technically implement this (especially if you’d like to volunteer to help out). I’ve been told one way to do this is to use MediaWiki and create a page for each project. That sounds good but I’d like to make sure we have the ability to rank and sort like I’ve described. Is that possible? I’ve also seen Pligg, a Digg clone, which might do the trick, but it while it has commenting and voting, it doesn’t have a wiki component. Finally, we could beg Jim Harper to let us use his Washington Watch software, but it’s custom-made and I think it’d probably be easier to use something off the shelf. I’m sure there are other ways to do it and it would be great if we could hash them out here. Thanks for your time!

Previous post:

Next post: