A Grater for Democracy

Over at Greater Democracy, they’ve been having a heated discussion this week around who did what and why during the 2004 Democratic campaign, Dean and post-Dean. It starts with a post by Jock Gill regarding the traditional Master/Slave structure to campaigns and the peer-to-peer model that the Dean campaign embraced pretty well (Jock introduced me to the Dean campaign). There are comments by, in order of appearance:

Valdis Krebs “Democrats don’t get “social networks” as well as the Republicans.”
Jon Lebkowsky  ”The Kerry campaign focused on raising money, not building community.”
Aldon Hynes “After the election much of the work promoting P2P activism slowed to a crawl.”
Joe Trippi “It was a fight every day to keep the master/slave beast at bay. [vs. P2P]”
Rayne “We are now in the middle of ‘making it up as we go along’ yet again.”
Zack Exley “We have an excess of ‘ideas’ and a deficiency of developers/engineers. It’s a shame.”
Robert David Steele “Gore needs to form a coalition shadow government and use it to help elect people in 2006”

You can taste the frustration, catalyzing the kinds of who-shot-John claims and counterclaims that feel so wasteful but which probably help energize the group. I’m as amazed as anyone that there is no comprehensive set of campaign tools out there, and that’s why we’re working so hard on ORGware. I’m also amazed at how long it has taken us to get as far as we are. If I’d realized in December 2004 that, in the spring of 2006 we’d still be where we were in the fall of 2003, maybe I would have got moving earlier. As I described last time, our development was as accidental as purposeful. 

We really need better ways of presenting the vital discussions that Valdis Krebs describes in It’s the Conversations, Stupid! (pdf). As David Weinberger said recently, The nits are determinative. Consider how the structure of the Greater Democracy site dissuade the reader from digging into this important conversation, and how ORGware addresses the determinative nits:

• The main post and comments all run together, making the text hard to even approach, much less follow.
• Long posts should be automatically truncated, but expand in place when the reader wants the whole thing.
• Two LONG documents are embedded in two of the comments. Because the comments are not automatically truncated, these lists of political requirements dissuade the reader from continuing. ORGware encourages the writer to upload the original text files, which the reader can open in a separate window to maintain context.
• All blogs and comments are also displayed as threaded discussions, which can be collapsed and expanded as needed.
• Comments must be on the same page as the main blog, condensed to a single line that the reader can expand at will.
• Comments need individual hyperlinks. It’s also helpful if a comment is posted to the reader’s personal blog as a primary post.

Nevertheless, you really should read the whole thing. The rest of this post is a summary that may be a helpful introduction.

From Jock’s original post:

We must understand how the dominant organizing principles of our national communications infrastructure shapes and determines our politics. If we want a truly democratic politics, based on the notions of equality with justice and fairness for all, based upon truly symmetrical relationships, we will have to have a communications paradigm that supports that goal.

Here’s Rayne‘s description:

These needs aren’t restricted to candidates, either; there are groups like Congressional District organizations, caucuses, more, that all have similar needs. I’m involved on the tech team for one caucus, has tech folks on board, but the tech folks end up in a turf-war over the best technology, confusing the rest of the non-techs on the team. The techs are also damned busy with day jobs, can’t afford more time to code. (Same group has spun its wheels for nearly a year…) Part of the problem is leadership and accountability, but part of it is that the question of best practices (for each platform, if multiples are there, including a tool for weeding out the platforms based on capital available and technology on hand) even enters into the equation. What is a best practice, in layman’s terms so that we can cut to the chase and spend the time on coding and content?

We at Open Resource Group think we have the answer, but we’re forced to build a comprehensive set of tools for each campaign-in-a-box, for reasons I described after the Berkman presentation 6 weeks ago: The public will not use any tools that a campaign site does not provide.

Valdis Krebs has thought about this deeply in his PDF:

Using the small-world model, researchers investigated the effect of a single person’s decision to
vote.  A person’s influence spread throughout their local group.  People were 15 percent more likely
to vote if one of their political discussants made clear their intentions to vote.  Within the research
population a citizen would positively affect the turnout decision of up to four other people.  The
researchers called this a “turnout cascade”.  In addition, the increased turnout was found to favor the
candidate of the initiator.  Human clusters tend to contain similar preferences for candidates and
issues, thus an increase in participation was equivalent to an increase in between two and three votes
for the candidate.  Denser clusters tended to show higher rates of voter participation.

Jon Lebkowsky:

The approach we were advocating could be a factor in future elections, and an online, community-based/grassroots movement, could provide an effective alternative to the current two-party structure that wouldn’t necessarily replace either party, but would provide for greater participation at more levels. We shouldn’t be too idealistic about this, however. Concentrations of money will always be a significant factor, and millions of people with no money will still have less power than a few wealthy corporations and millionaires/billionaires.

I don’t agree, for reasons that will take two years to make clear. Aldon Hynes remarks also on the lack of work being done to bring the needed tools on line:

There is a lot of focus, some of it very important, on who is really adding to campaigns. Some of it can be about ego and not especially productive, but some of it is about measuring activity and helping people become more productive.

My big concern is that after the 2004 election much of the work promoting peer to peer networked social activism seems to have slowed to a crawl.

Hopefully, with the 2006 cycle we can see a resurgence of interest in peer to peer enabled politics.

Joe Trippi thanks Jock Gill for his contributions to the Dean campaign and reminds us:

Two points I would make. 

1. We were making it up as we went along and (initially at least) building it with no funds and little political experience.
2. There is an implied belief among many that there was tremendous agreement inside the Dean Campaign to take the Peer-to-Peer path over the Master-Slave model — this simply was not true. 

It was a fight every day keep the master/slave beast at bay. In hindsight the miracle was that we held it off as long as we did given how many inside and outside the campaign relished master/slave over peer-to-peer.

Zack Exley reminds us that there are far more ideas than working code embodying those ideas:

At Kerry, we would have jumped at any help offered in community building — or any area. But what was most often offered from outside the campaign were ideas – not implementation. Sure, ideas can be a big help. But we needed developers to implement, designers to design, and testers to test. And of course we know that not all ideas actually take off in practice (remember Deanlink?) — so you can’t blame a campaign for not sinking precious time and resources into ideas in the run up to a critical election.

And I’ll say once again, we didn’t just raise money. (Though we did raise a huge portion of the entire campaign budget online.) Most of our team’s effort for the last several months was spent on driving field organizing. We built tools to enable that. It WAS a great tragedy that it was focused almost 100% in swing states, which is why most people who comment on this stuff didn’t see it.

We could have done it all way cooler if we had more talented developers who could build stuff as fast as all of us *idea people* could think stuff up. But we didn’t. So we had stark choices to make: give Deanlink another shot, or raise more than $100m and mobilize 250K volunteers in the swing states.

If we only had one good developer for every Internet strategist/guru/author — then we’d be living in a new and wonderful world of online organizing indeed.

Robert David Steele writes:

I think we also need to go into stealth mode. Gore needs to form a coalition shadow government and use it to help elect people in 2006, and then use the 2006-2008 period to do peer to peer not just in the USA, but overseas as well. If Michael Cudahy can deliver a 20% bounce from moderate Republicans, I believe I can deliver a 20% bounce from other non-Democrats, and a further 20% bounce from foreign relatives of voting immigrants.

And Jon Lebkowsky wraps it up:

What are your requirements? I pointed this discussion out to members of the demtech email list and invited them to drop by, but members of that coalition developed platforms like CivicSpace (for community development) and Advokit (for GOTV coordination), and other Open Source tools have appeared (e.g. CiviCRM). You can usually get volunteers to set these up.

I hate to disagree with my editor and co-author twice in one post, but I must. Unfortunately, campaigns do not have the insights nor the patience to engage in the long and arcane conversations that might theoretically result in the volunteers setting up “CivicSpace (for community development) and Advokit (for GOTV coordination), and other Open Source tools”. They want their site working by Tuesday, not an involved conversation that might start next Tuesday, if the volunteers show up. 

And I cannot over-emphasize that the campaign site’s users will not put up with a User Interface any less obvious than an airport kiosk.

9:03:12 PM