Colin Delany's comments on Matt Bai's recent NYT article reminds me of so many conversations I have had about how Google killed message control. For a long time, I have argued that campaigns cannot control their message anymore. At best they can only hope to manage the chaos.

But as a result of a recent conversation with my friend Lenny Steinhorn, I am convinced that the internet has not killed message control, but rather language control. Campaigns may still be able to shape the message, but citizens are free to internalize it and restate it in their own language. And with the internet, any campaign's message is able to take root in the polity, in many forms and with greater impact.

So where Bai sees campaigns riding unpredictable and uncontrollable waves, Colin is right that there strategies can make a difference. By feeding the online discourse with messages, facts, resources, nudges, and tugs; by pushing these out to a variety of online communities; and by listening to the response and adapting message and strategy accordingly a campaign can add strategic advantage to the organic chaos of the internet.

Let's look at some past examples, starting with the Dean campaign. Sure, Dean's staff stumbled onto 7500 people meeting up around the country, but they quickly recognized the value and potential of this discovery, watched it closely, and THEN implemented a strategy to enhance it. Hundreds of thousands of Dean MeetUp members later, he rolled into Iowa as the favorite (leave aside his loss there, there are still many lessons to be learned about strategies for managing messages in a chaotic environment).

Because of Dean's experience, we now know to look across the internet to find opportunities to engage voters. Just looking for these opportunities is a strategic decision and not something to write off as serendipity (though that happens, too). And once we find these opportunities, we have strategic frameworks for engaging these opportunities. Whether a campaign "floods the zone"; identifies and engages the mavens and connectors in these online communities; or hires an army of online organizers to organize Facebook, YouTube, and MySpace as others would New Hampshire and Iowa, there are real strategies that will make a difference.

And it is not just a matter of the right strategy. Even the best online strategy depends on having the right candidate and the right message. Jesse Ventura did not become Governor because of his internet strategy, his internet strategy helped push him over the edge because of who he was and what he was saying. You cannot assume that two candidates with the same strategy will have the same results.

For example, the Ron Paul explosion isn't just about crazy wired activists driving the campaign, it is a campaign with a popular message seeding the internet and finding purchase among highly engaged voters who hear Paul as a voice of rationality amidst a sea of platitudes and cliches.

I think in the final analysis, we need to broaden our definition of orchestration. A conductor may orchestrate a group of musicians to play a concert, but if the score sucks and the musicians have no voice of their own, the result will sound flat. A coach may come to a team with a strategy that does not fit the team's personnel. In both cases, orchestration requires a deeper understanding of what resources are available and adapting strategy to those parameters.

Yes, the internet has changed the dynamics of campaigns and it can be the vehicle to catapult a viral explosion, but good strategy can create something out of nothing and it can help push something to its limits. The lessons from Dean are far more subtle than Bai suggests, though he does raise some important ones. His assessment is a good start. Colin's addition takes us further still. Likely, there is much more for all of us to learn.