User stories for open government?

Posted: December 1st, 2008 | Author: | Filed under: Design, Transparency | Tags: | No Comments »

As we build tools that make government more accessible, are we addressing the actual needs and wants of citizens, or the needs and wants we wish they had?

In Sunlight Foundation’s Open House/Open Senate Project discussion, Clay Shirky wrote:

Without a middle step that helps large, disorganized groups take advantage of the newly transparent information, transparency may in fact further increase the net asymmetry betwee ‘interest group with lobbyists’ vs. ‘interest groups without lobbyists’ in getting the Government to craft the needed bargains their way.

The extreme programming technique of defining “user stories” to be
handled by the software could be useful here. One format is

As a (role) I want (something) so that (benefit).

Possible user stories:

  1. As a voter wondering whether to respond to Politician X’s
    fundraising appeal, I want a summary of his track record on Issue Y so
    that I can decide quickly whether to send him $25.

  2. As an “Issue Y” voter, I want to know whether today is the day I
    should spend the four hours a year I have budgeted for calls and email
    to my Senator, and what I should say to him, so that my four hours
    have as much impact as possible on the Federal Government.

  3. As a policy analyst, I want to identify the Senator who has been
    most vocal and consistent with my position on Issue Y, so can I get a
    Senatorial hold placed on a bad bill. (e.g.
    http://www.salon.com/opinion/greenwald/2007/12/18/victory/ )

  4. As an investigative reporter, I want to know what changed in seven
    months, so that I can explain why FISA legislation that failed in
    December 2007 then passed in July 2008.

  5. As a mash-up programmer, I want access to the geographical
    locations of all the sites earmarked for funding in Bill Z, so I can
    place them on a Google Map.

More on user stories:
http://www.agilemodeling.com/artifacts/userStory.htm.


Friend to friend canvassing: more fun, and 50% more effective

Posted: November 8th, 2006 | Author: | Filed under: Phone banks | No Comments »

12/11/2006 update
The Grassroots Champions Coalition has used Advokit to organize precincts throughout California.

Bruce Daniels describes their implementation approach here.

What if you had tools that allowed participants to share their networks with your effort, in addition to their time, thoughtfulness, and energy?

Robo-calling: tempting for campaigns, annoying for voters, ten times the expense

In the election just concluded, I received countless robocalls, mass mailings, and email messages from candidates and elected officials. I voted, of course, but the only thing that moved me to action was a friend’s emailed account of his own door to door efforts in the Virginia Senatorial campaign.

I’m not alone. The news is filled with stories of voters in Arizona, New York and other states who are tired of getting calls.

It’s some comfort to know that robo-calling, at $275 per new vote generated, is 10 to 20 times more expensive than canvassing or phone banks.

Friend to friend

Pat Dunlavey has found a way that’s more humane and more effective.

In 1998, 1999, and 2003, Dunlavey organized ballot measure campaigns in Massachusetts to allow local governments to increase taxes to fund needed local services.

1998: Ten supporters identified per canvassser via cold-calls

In his first effort, call lists were assigned by the campaign based on geography. 85 volunteers each made 30 contacts per person, and identified only one third of the contacts as ballot measure supporters. Volunteers generally completed only 60% of their assigned contacts and found the cold calling an unpleasant experience.

Although the ballot measure passed, Dunlavey was disappointed with the results, and sought ways to avoid cold-calling.

2003: Fifteen supporters identified per canvasser via friend-to-friend

In 2003, his volunteers used a custom web application (reborn in 2004 as Advokit) to scan the town’s voter database and “tag” or claim voters that they felt, based on personal knowledge, would be supporters of the ballot measure. The 55 initial canvassers recruited an additional 40, and then each canvasser contacted more than 26 people on average and identified 15 supporters. The 2003 group was both happier — completing 85% of the calls assigned — and more effective — they converted almost twice as many of of their contacts to supporters (57% in 2003 vs. 31%in 1998)

Netalyst worked with Advokit (and CivicActions) during the fall of 2004 to support the VoteAllYourValues effort , to design an easier to use “lite” interface and to develop strategies that would allow the campaign’s ambitious targets for calls generated to be reached.

Advokit was also used with good results in that year’s New York State senate races.

The Advokit website provides an online demo and free download of this open source software.


Cellphone text messages + passion = results across the digital divide

Posted: November 3rd, 2006 | Author: | Filed under: Demographics, Text messaging (SMS) | No Comments »

Text messages sent by cellphone — also known as Short Message Service or SMS — are becoming a powerful tool for political and civic participation.

Last fall, activists opposing the Supreme Court nomination of John Roberts recruited 25,000 volunteers for their Massive Immediate Response effort. Each volunteer agreed to respond immediately to a text message requesting that they call their Congressional Representatives. The 27% opt-in rate was five times that observed for the most successful commercial entertainment campaigns.

Rick Santorum, the Republican incumbent running for U.S. Senate in Pennsylvania, used SMS to reach out to his supporters.

On Halloween, Pat LaMarche, the Green Party candidate for Governor in Maine, asked her supporters to provide topic suggestions via text message for a speech at an upstate university.

Note that all of these initiatives relied upon participants who were already fired up for or against particular candidates and issues — SMS is less likely to be useful in converting a luke warm supporter into a passionate participant.

Almost one third of all cellphone subscribers send at least one text message per month, with younger Americans leading the way. A recent New Politics Institute study notes that cellphone subscribers comprise more than two-thirds of the population, that more than half of 18 to 34 year olds use text messages at least occasionally, as do one in five 35-54 year olds.

Among those 35 and under, text messaging reverses the digital divide. Five out of ten Hispanics, four out of ten African Americans, but only three of ten whites use their cellphone text messaging capability. So it should not be surprising that Voto Latino announced plans in July to register at least 35,000 Hispanic youths nationwide using cellphone text messaging tools.

These tools are rapidly reaching the “point and click” stage for campaign organizers. For instance, the Mozes service allows campaign manager to set up “text message ballots” easily. A “do it yourself” American Idol is using this service to let listeners to indicate the band they prefer in one on one matchups.

To track developments further, visit mopocket, textually, and the MobileActive Community Blog.

Update: 2006/11/17
The Democracies Online wiki provides a quick overview of SMS pros and cons and describes how SMS was used the Philippines, Lancashire, Bristol, and South Korea.