I’ve been preparing a presentation for the Canadian Institute for the Administration of Justice’s Legislative Drafting Conference on Rules as Code. In the process of trying to figure out how to introduce the topic of Rules as Code to legislative drafters, I’ve been thinking about the different meanings the term “Rules as Code” has been given.

我一直在为加拿大司法行政学院法典法规起草会议做演讲。 在试图弄清楚如何将“法规规则”引入立法起草者的过程中，我一直在思考“规则法规”的不同含义。

Some people use “Rules as Code” to refer to the larger public service delivery methodology, which includes a method of legislative drafting, where you write the legislation in natural languages and computer languages at the same time. Some people, I think more intuitively, use “Rules as Code” to refer to the product — a piece of legislation written in a programming language.

有些人使用“规则即代码”来指称较大的公共服务交付方法，其中包括立法起草方法，您可以同时用自然语言和计算机语言编写法律。 我更直观地认为，有些人使用“规则即代码”来指代产品，这是一种用编程语言编写的法规。

They aren’t the same thing, and the naming has been a source of some minor controversy among proponents. But what interests me is not what you call them. What interests me is what happens when you combine them, and what happens when you don’t.

它们不是一回事，命名一直是支持者之间一些次要争议的根源。 但是让我感兴趣的不是您所说的。 使我感兴趣的是，将它们组合在一起会发生什么，而当您不合并时会发生什么。

You can do both at the same time, clearly. You can use the Rules as Code methodology, and through that process produce Rules as Code products.

显然，您可以同时执行两项操作。 您可以使用“规则作为代码”方法，并通过该过程生成“规则作为代码”产品。

And you can also produce Rules as Code products without using the drafting and public service delivery methodology. That type of “Rules as Code” is more or less just how legal application software would be developed now, if developers compartmentalize out the legal reasoning. (I think if you don’t have the legal rules separate from the rest of the code, you’re talking about “Rules in the Code, there, somewhere”, which is different.)

而且，您也可以将规则作为代码产品生成，而无需使用起草和公共服务交付方法。 如果开发人员将法律推理划分开， 那么这种“规则即代码”或多或少就是现在将如何开发法律应用程序软件。 (我认为，如果您没有将法律规则与代码的其余部分分开，那么您所谈论的是“代码中的规则，在某处”，这是不同的。)

But surprisingly, there are also benefits to the legislative drafting methodology by itself, even if you never actually deploy any code.

但是令人惊讶的是，即使您从未实际部署任何代码，立法起草方法本身也有好处。

If you want to understand something, teach it.

如果您想了解一些东西，请教它。

Why is this true? Because the process of attempting to communicate something you know to a person who does not already know it forces you to engage with the difference in contexts. It’s a similar idea to the “curse of expertise”, where the expert forgets what it is like to be a person who does not already understand.

为什么会这样呢？ 因为尝试将您知道的东西传达给尚不了解的人的过程会迫使您参与不同的环境。 这与“专业知识的诅咒”相似，专家会忘记成为一个尚未理解的人的感觉。

Overcoming the curse of expertise requires you to put yourself in the shoes of a person who knows less, and rebuild your own understanding from that new starting point.

克服专业知识的诅咒要求您将自己置身于一个知识不足的人的脚下，并从新的起点重新建立自己的理解。

Similarly, when you are trying to understand a piece of legislation, a computer takes the role of the person who knows less. And the syntax and semantics of the programming language give you the entire context of what the computer already knows.

同样，当您尝试了解某项法规时，计算机会扮演了解情况较少的人的角色。 编程语言的语法和语义为您提供了计算机已经知道的整个上下文。

So attempting to put your legislation in code — even if you do it with pen and paper— forces you to understand it from a position of less context, and you learn it better. You catch mistakes you didn’t know you were making.

因此，尝试将法规纳入法规(即使您用笔和纸来做)，也会迫使您从上下文较少的角度理解法规，从而更好地学习。 您会发现自己不知道在犯的错误。

It is, in that sense, just an advanced technique of legislative drafting.

从这个意义上讲，这只是一种立法起草的先进技术。

For me personally, this is not a mere hypothesis. I have consistently experienced this effect. Every time I sit down to encode a ruleset, I end up with a deeper understanding of how those rules work long before I have asked the computer to do anything with the code.

就我个人而言，这不仅仅是一个假设。 我一直都经历过这种效果。 每次我坐下来对规则集进行编码时，在我要求计算机对代码执行任何操作之前，我都会对这些规则的工作方式有更深入的了解。

Most recently I have been encoding a set of rules regarding Covid-19 in Alberta. I read them a few times and saw nothing wrong with them. Then, I tried to put them in code, and realized that there is a scenario in which a person who is known not to have the disease can nevertheless be forced to continue to isolate.

最近，我在艾伯塔省编码了一组有关Covid-19的规则。 我读了他们几次，没发现他们有什么毛病。 然后，我尝试将它们放入代码中，并意识到在一种情况下，仍然可以迫使已知没有这种疾病的人继续隔离。

Professor Scott Brewer at Harvard Law has a similar argument, which relies not on programming languages but on formal logic. He asserts that lawyers should learn to analyze, for example, appellate court decisions by making formal logic representations of the arguments made by the court, and analysing them using the methods of formal logic. Doing so reveals things that a less disciplined analysis doesn’t.

哈佛大学法学院的斯科特·布鲁尔(Scott Brewer)教授也有类似的观点，它不依赖于编程语言，而是依赖形式逻辑。 他主张律师应该学会分析，例如，通过对法院提出的论点进行形式上的逻辑表示，并使用形式逻辑的方法对其进行分析，来对法院的判决进行上诉。 这样做可以揭示纪律较轻的分析所没有的事情。

The programming languages with which I am most enthusiastic about for use in Rules as Code are built on a foundation of formal logic. Perhaps at some point something along the lines of Blawx will be used in law schools to teach legal analysis.

我最热衷于将规则作为代码使用的编程语言是建立在形式逻辑的基础上的。 也许在某些时候， 法学院将使用类似于Blawx的方法来教授法律分析。

Wouldn’t that be something?

那不是吗？

Speaking of Blawx, my paper on Rules as Code and Blawx just went live as a part of release 1.2 of MIT’s Computational Law Report. I’d love your feedback.

说到Blawx， 我关于CODE和Blawx的规则的论文 刚刚成为麻省理工学院计算法报告1.2版的一部分。 我希望收到您的反馈。

计算法工作机会 (Computational Law Job Opportunity)

• I just heard about a wonderful opportunity for a 4-year PhD with Dr. Prakken at Ultrecht University, as part of a project to generate explanations for the conclusions of “black box” artificial intelligence algorithms. Dr. Prakken is a leading expert on modelling legal arguments, and an excellent teacher.

  我刚刚听说与Ultrecht大学的Prakken博士一起获得为期4年的博士学位的绝佳机会 ，这是为“黑匣子”人工智能算法的结论做出解释的项目的一部分。 Prakken博士是法律论证建模的领先专家，还是一位出色的老师。

接下来… (Coming Up…)

• A piece I wrote for Slaw.ca about the SMU Centre for Computational Law will be coming out in late August.

  我为Slaw.ca写的关于SMU计算法中心的文章将于8月下旬发布。

• I will also be participating in a Rules as Code panel on the first day (September 10) of the 2020 Legislative Drafting conference held (virtually) by the Canadian Institute for the Administration of Justice. Check their website for registration details. There are sponsored tickets available for students, I understand.

  我还将参加由加拿大司法行政学院举办(虚拟)的2020年立法起草会议的第一天(9月10日)参加“规则作为规则”小组。 检查他们的网站以获取注册详细信息。 据我了解，有赞助给学生的门票。

I am a lawyer at Round Table Law, I teach “Coding the Law” at the University of Alberta Faculty of Law, and I’m a senior researcher at the Singapore Management University Centre for Computational Law. Computational Law Diary is a series of posts on what I’m thinking about in the computational law world. They are my own opinion, and do not reflect the opinions of the Centre, the Research Programme, SMU, U of A, or anyone else.

我是圆桌会议法的律师，我在艾伯塔大学法律系教授“编码法”，并且是新加坡管理大学计算法中心的高级研究员。 《计算法日记》是有关我在计算法领域中所思考问题的一系列文章。 它们是我自己的观点，并不反映中心，研究计划，SMU，A或任何其他人的观点。