We need Digital Law for a digital world

It’s hard to believe that Canadian law is not available to the public online. Theoretically it is, but not in practice. Yet our lives are governed by millions of legal rules encoded in hundreds of thousands of places. It's impossible for anyone to find and read them all, even when they are published (and they often aren't). The only possible hope for a society in which we comprehensively abide by laws is to have them encoded digitally. No country has succeeded at doing this, or even tried, but Canada could move to a new system of “Digital Law” publishing in a matter of a few years, at relatively low cost.

It will still be a tall order. We have a disparate collection of laws published by the government, case law distributed by the court system, and various rules issued by boards, tribunals, and other arms of our sprawling bureaucratic state. Regulations thoroughly pervade our lives - to the point where in 2019 “unregulated” is a four-letter word for many. But how we can reformat our rules to bring law into the 21st century?  

Here are the features of what a Digital Law publishing system would look like.

First, publish everything

The first step to Digital Law is to publish everything that's intended to be a part of the law. That's not the case right now and Canadians are being left in the dark about what the law that governs their life is. Many administrative boards don't publish cases at all. Even the Ontario Superior Court doesn't post all of its decisions. This needs to change. Law must be made available to the public online. The next step is to public in a useful format, but if the laws aren't published comprehensively then everything else is moot. Every legal rule must be available to the public.

Clean up messy and unstructured data

Courts, agencies, boards, ministries, and governments publish laws in thousands of unstructured formats, without any standardization (if at all).

It’s problematic because laws are data, and right now that data isn't structured in a way that computers can understand ("Garbage in, garbage out"). The reason our legal system suffers from a “garbage in” problem is that law-making organizations don't work with each other. They also don’t know why they should structure their data, or how to go about it. Tech is a second-class system in law. The legal system continues to publish its output as it always has, even while tech companies invest billions into Big Data to tap into the new oil of the digital era.

We need to format our laws consistently and properly. In the same way that people must fill in the forms correctly to get services from the government, the government needs to fill in the "forms" for us about what the laws are.

Access in many languages

An important benefit of standardized formats is that we can then translate laws, for non-native speakers, using automatic translation tools. Machine translation has advanced to the point that this can be done accurately and quickly. That would enable Canada deliver on the promise of a bilingual country, while also providing access to the millions of Canadians who can't read French or English fluently. This would be impractical to do with human translators, but can be done efficiently at scale for dozens of languages with great quality, and for even more at lower fidelity. Standard formats are the first step towards unlocking laws using other types of technologies, whether that's automated translation or screen readers for people with sensory disabilities.

References are essential

How laws relate to each other is a critical feature of our legal system. Judges interpret them and elaborate upon them, referencing existing laws. We need to standardize references to allow computers to understand the various relationships between laws — critical because of the interdependencies that exist in Canadian law.

It doesn't matter what the format is, it just needs to be consistent.

Helping computers talk to each other 

APIs are standard interfaces for computers to access data. Most laws today are published in ways that humans can understand but don't provide any programmatic way to access laws (or take active steps to block programmatic access). The best example of what this could look like is the United States Federal Register API. Although there have been some efforts in Canada like the BCLaws API, APIs are not typically available for Canadian law. This needs to change.

Municipal laws (and everything else)

Municipal laws are important but rarely available to the public. The City of Toronto published well over 1000 by-laws last year and there are millions of by-laws across Canada that can't be searched or discovered without making a trip to a research library. These laws should be registered in a central location and made available.

The broader point, though, is that all laws need to be made available to the public, no matter what form they take. There's a lot more to law than just provincial or federal statutes and regulations.

Plain language summaries

Most laws are written in legalese that the public can’t understand. Where possible, we need summaries, organized in a way to deliver what's most important to the readers. A good example of this is the format adopted by the Nova Scotia Court of Appeal in explaining its decisions (e.g. this recent murder case). Although computers can summarize information, today's technology isn't capable of providing concise summaries that legally- trained people who work with the subject matter can produce.

Presumption of openness

The key to Digital Law is the presumption that laws are open to all and ought to be accessible. Many stakeholders in the legal system resist the notion, in part because they're comfortable with the status quo and don't see what advantages openness has to offer to them (or you).

Lawyers primarily access law through subscription services like LexisNexis, Westlaw, textbooks, and other services that aggregate legal information and publish it in formats that can be used. It is a multi-billion business globally and lawyers pay a hefty price for those services.

There are also contractual restrictions that prevent third parties from using digitally published laws to develop tools to help the public. For example, the Ontario Superior Court outsources publishing to an organization funded by the Federation of Law Societies of Canada, which then bans automated collection of the cases through their terms of use agreement (in part required by upstream licensors, the courts and boards) and technical countermeasures.

The Canadian Judicial Council (which oversees federal judges in Canada) even discourages indexing by search engines so that court information won’t show up in Google searches. But that is at odds with the growing movement for open data in Canada and abroad, where governments have recognized that freeing data is the right thing to do.

The type of businesses that could be built around Digital Law are fundamentally different than the current structure in which a small number of companies dominate the industry through private deals to gain access to court decisions. Free, public, standardized data is the raw material needed for creating accessible legal products to serve Canadians.

What about cost?

Overall, Digital Law would likely end up being cheaper than the current system of ad-hoc publishing where thousands of different organizations are expected to figure out how to make their legal documents available. A standard system would streamline this and greatly reduce the IT costs involved in maintaining these systems.

A relatively modest team of people could support the entire country's infrastructure for laws. Modern software delivery methods allow very small teams of developers to deliver hugely impactful systems.

Even if Digital Law doesn’t turn out to be cheaper than the current system, the consumer advantages would more than make up for it. I can't imagine a cost-benefit study wouldn't conclude this to be a more cost-effective approach (and that would be a great first step). The current model relies heavily on non-government actors supplying the appropriate metadata and formats as to make the law useful, but this comes at a heavy cost in the form of legal product subscriptions and inconvenience for the public.

Access to justice through Digital Law

Open access to Digital Law is an absolute must if we want our laws to be understood by people. There are simply far too many laws to work with on a manual basis.

We need to upgrade our infrastructure to start delivering cheaper legal services to the people who need it (which is all of us). If Digital Law became the standard in Canada then incubators like the Legal Innovation Zone will quickly cultivate a wide range of tools for Canadians. Some of the ecosystem already exists to take advantage of Digital Law, but the government must do its part. Private actors can't force boards to make their decisions available online.

Without Digital Law, we will never have affordable, good quality digital legal solutions. It’s an important first step toward solving the A2J crisis.

This is the second in a series of posts about solving the A2J crisis. The first article explains the larger problem and provides context for this piece. This is an edited version of a longer article.