Hanging out at the dg.o digital government conference this week has got me thinking about the relationship between a legal-information infrastructure-builder like the LII and digital government projects generally. The LII moved into e-government work quite easily. That happened because both are concerned with increasing public access to legal process, which is often the same problem as increasing access to government process. More important, perhaps, was that the move reinforced the LII’s longstanding role as a convenor of conversations between information scientists and legal specialists. Much of my role in the project has been translating between the two. It’s not as simple as one-speaks-math-and-the-other-speaks-administrative-procedure. The two academic cultures are very different in the way they go about things, for one thing. But one of the great benefits of the LII’s location in a great research university is that we can bring world-class experts together around interesting problems, and provide real solutions to the public.One obvious relationship between digital government and legal information is simply that the LII is a reference library for all manner of digital-government activities. You can see this in our link-census, if you look for government sites that link to the LII. Many of them are using us to provide background information for web pages that support citizen interaction with government. And I have long believed that citizen interaction can be a valuable justification for a legal-information service in places that are inclined to resist exposure of the legal system. The clear benefits of increased efficiency and effectiveness in government interaction with citizens could potentially be seen as outweighing the risks of transparency.
Digital-government research — especially the part of it that involves the computational processing of regulations or other legal text — offers valuable techniques for us. The same advanced language technologies that can be used to automatically categorize and structure responses in notice-and-comment rulemaking can help us organize and navigate judicial opinions. The algorithm that looks promising for extracting metadata from SEC filings can do the same thing with Circuit Court opinions, and so on. So you can get a lot of ideas for techniques and services at a conference like this. And we have our own research and techniques to offer too. Most important, we have a very deep well of experience with non-lawyer professionals who are trying to find out about the law.
Which brings me to the first of two slightly deeper observations. Non-lawyers trying to find and understand the law face two barriers that look a lot like the barriers for ordinary citizens who try to participate in e-government. Neither has anything to do with barriers to Internet access per se, or with the usual run of “digital divide” issues. The first is that they have to somehow figure out how to map a set of terms or nouns found in a description of their problem or interest against a set of more formal concepts that map things in the way that legal or governmental agencies understand the world to be. At the LII we’ve frequently joked that we’re trying to build systems that will find the article labelled “nuisance” when the user submits “barking dog” to the search engine. This is not exactly a new idea; Dan Dabney of West Group was making the point about differences in search style several years ago, though unfortunately none of his remarks were ever published. Similar problems apply when ordinary citizens try to figure out, say, regulations put forward by a particular agency. They have their own way of talking about things.
And of organizing and doing things, too. The second shared barrier or threshold has to do with the naive user’s lack of understanding of how the government, or the legal system, is organized. The farmer who grows peaches is not necessarily going to know which agency regulates their importation or distribution, just as he is unlikely to have much understanding of jurisdiction or civil procedure when starting to pursue a purely legal question. So the question of how we get inexperienced users past barriers rooted in a lack of structural or procedural understanding is one that is shared between these two areas of investigation. And it’s an important one to solve. Public education is part of the answer, as are improved search and navigation systems, as well as tutorial systems that help users frame questions.
By the way, there are 17 separate Federal agencies whose activities touch the problem of food safety. And literally hundreds of regulations that apply to the use of dry-cleaning chemicals — something of vital concern to 85,000 small businesses in the United States, most of which are family owned. That’s a lot of help to try to give people.
I said there were two deeper observations, though perhaps this second one is more intuition than anything else. Ideas about citizen participation in the rulemaking process tend to fall somewhere on a spectrum defined at the outside by two equally impractical views. At one end, we find extreme direct-democracy advocates who believe that any attempt to channel or structure citizen input into governmental process either is now or could become a form of censorship. At the other end, some within agencies believe that increased access to process (via, say, e-mailed submissions) inevitably devolves into spam, anarchy or both. Each can point with some validity toward the excesses of the other. Both are wrong. Completely unstructured, unmoderated input is too confusing and noisy to be of much real use. Anybody who is sufficiently socialized to know that people take turns talking in face-to-face conversations has absorbed that lesson. At the same time, narrowly channeling and control of input from a set of stakeholders limited to those government already knows about is equally problematic; it misses a lot. So a better question is where and how to build reasonable structures for conversation. Likely, it’s somewhere in the middle. And many of the techniques that we create to do that will be equally useful for structuring conversations about what other kinds of law should be.
So there are lots of reasons for the open-access-to-law community and e-government researchers to be talking, and here in Montreal it’s a good discussion. As usual, the LII staff is dreading my return, because I’ll be chock full of ideas that someone will then have to build…..