By the end of today, the Bush administration will have published a flood of new regulations, pushing them into the 60-day pipeline that gets them enacted just before the Obama administration takes office. The blowback on these “midnight regulations” centers on the minimal opportunity for the public to review and provide input into the process. The speed with which these rules have been shoved through mocks public participation: the Interior Department reportedly had 15 experts, in 32 hours, filter over 200,000 comments on proposed relaxation of the Endangered Species Act. Unsurprisingly, the final rules are little-changed from the original proposal.

Can new technologies, and new techniques for applying technology, address this problem?

Balancing asymmetric attention

When the Obama campaign promised to tap into “the vast and distributed expertise of the American citizenry to help government make more informed decisions,” it restated one of the major goals of rulemaking. Administrative agencies take over when Congress can’t provide needed expertise or flexibility. In theory, notice-and-comment is the way this process stays democratic, by letting citizens get directly involved.

(The campaign has already demonstrated its trust in involved citizenry by letting volunteers have access to its vast voter database -- a different kind of transparency and engagement, but a similarly huge leap of faith.)

The problem is that rules are often dry, obscure, and boring. As a result, citizen participation has been losing an asymmetric war of attention. Those with the sharpest interests at stake pay the most attention, invest the most time, and pay for the best research (a/k/a “lobbying”). Your average citizen, by contrast, cares about these issues only a tiny bit, and it’s the odd crusader who has the time and wherewithal to provide contrasting views. For a chilling example of what can go wrong, consider the 2004 SEC hearing in which five major banks supported loosening debt restrictions; one lone citizen dissented remotely, from Indiana, and was roundly ignored. Our crashed economy represents the cataclysmic consequences of that single decision.

The Internet has proven effective at putting small bits of attention to valuable use, which might help restore some balance to this situation. An army of citizen-activists could pore over proposed rules, flagging issues and distributing the work of collating evidence, Wikipedia-like, to support or oppose proposals. Left to their own, citizens tend to gravitate to the same high-profile issues. (Notice how the media converge on the Endangered Species Act when discussing the midnight regulations). But with a little nudge (like Google Image Labeler’s game, for example), the distributed army might wander to more diverse battlefields.

When transparency isn’t enough

However, expertise about a topic like bank capital reserves isn’t as common among the public as the ability to label a picture or decipher a bad page scan. So mass, Internet-enable participation isn’t a panacea, and public interest policy centers that aggregate capital to hire specialists as a proxy for aggregating interest remain vital. Even so, lowering the bar so that the handful of experts, rather than the masses, can join in would help. (So, too, would top-down efforts by policy groups to coordinate armies of volunteers to gather grassroots data and stories). More importantly, citizens can underline the values that agencies should weigh most heavily when choosing among options (save more jobs, or save more trees?)

Of course, the deeper problem is the one we see today, 60 days from the next administration: the mere opportunity to comment does not necessarily create the possibility of being heard.

So making the rulemaking process more transparent isn’t a full solution. Although it takes work, you can already see these midnight regulations yourself. Just knowing about proposed rules isn’t enough. We must have meaningful participation.

Comments and the problem of Astroturf

We put a lot of faith in comments in both our rulemaking process and in our online discussions. Yet anyone who’s seen a badly-managed blog knows that open comments can generate considerable noise that obscures useful information (sometimes on purpose).

Astroturfing turns this problem into a weapon, by generating artificial support for a viewpoint through fake (often hired) grassroots. It happens online, and it happens in real life, too.

Spamming a hearing

There are ways to curtail this kind of activity – comment rating and user reputation jumps to mind – but each technique can probably be circumvented if the hacker is motivated enough. (I can imagine a ’bot that registers and puppeteers supporters in a way that accurately mimics real people. The same can be done with human “bots,” much as many SEOs now user real people to spam blogs.) With cutting-edge developers on the government side (whether they are employed by the government or not), the process might stay one step ahead. But maybe not – which brings us back to the problem of asymmetric participation.

Comments vs. Participation

We’ll need more than just open comments if we want true participation in the rulemaking process. Beth Noveck, a leading thinker on “wiki government” (and Obama adviser) points out that careful design of the process – both technology and law – can enhance citizen participation. In particular, she suggests that we consider shifting our attention to group deliberation rather than individual (and individualistic) commentary.

There are thousands of experiments out there to use technology to enhance deliberation smartly. The efforts that interest me most pay attention to the “social physics” that pertain among people, building around people rather than information. I’m particularly fond of one obscure project, RedBlue, which brought together people from opposite political viewpoints to hold constructive conversations. The technology itself was rudimentary, but it embodied decades of experience facilitating face-to-face discussions among real people (specifically, held by the Public Conversations Project). Another project, here at Berkman, paired up students to discuss class topics, with the explicit goal of overcoming the usual problem of the first commenter dominating conversation or other students shirking contribution (As befits law schools’ reputation for grilling students, the tool is called the “Rotisserie.”) And Ebay’s automated dispute-resolution system deploys considerable insight into how problems arise and are resolved (it turns out that almost all disputes fall into a very finite set of types, for which there are good, automated responses).

The point is that good participation requires first a vision of what that participation looks like: who should be heard, what ideas must come out, what kind of back-and-forth produces the best results. (The Obama campaign had similarly paid close attention to making its voter database available to the grassroots: leaders got access when they proved themselves trustworthy, got training for their appropriate roles -- phonebanking vs. canvassing vs. volunteer recruitment -- and had plenty of online and human support). Only then does it make sense to build a technology to scaffold those social interactions. I’m pretty sure that the typical blog comment system is not one of those optimum solutions.

p.s. Codelaw counts

And a final word on transparency and rulemaking: Among those rules that the Obama administration should strive to keep transparent are the those that are implemented by software. “Code is law,” and as I’ve pointed out previously, law is, increasingly, code. When promulgated rules are executed through machine code, it’s absolutely vital to the public interest that we have equivalent notice-and-comment privileges to the code. Otherwise, the paranoia that surrounds black-box voting machines will become commonplace for everything from food stamps to no-fly lists.