Home | Enter via LawMoose Subscriber Edition | Login locally

Comparing Laugment to Other Legal Intellectual Term Collections

An excerpt from the paper: LawMoose, LawSaurus and Laugment From Pritchard Law Webs - What Comes After the Search Engine? - Semantic Network-Based Lawyer Augmentation, distributed to accompany a presentation by LaVern Pritchard at the Stanford CodeX Center for Legal Informatics, Stanford Law School, Stanford, California (June 25, 2015)

We looked to other efforts within the legal world to publish organized sets of legal terms.

For example, the Practical Law Glossary, publicly visible on its website, had 2,076 terms as of November, 2014.

Cornell's Legal Information Institute publishes Wex, a "collaboratively-edited legal dictionary and encyclopedia." We found Wex to have 5,290 terms and noticed, on multiple visits, that it does not grow.

The 9th edition of Thomson Reuters' Black's Law Dictionary, published in 2009, held 45,000 terms.

The 10th edition, published five years later in 2014, contains more than 50,000 terms, including 7,500 new ones, according to the publisher. Referred to as "the most widely cited law book in the world," it is undoubtedly the standard reference dictionary in America.

Laugment has more than five times as many terms compared to Black's. That's because Laugment is about a lot more than words in a dictionary.

Nuria Casellas, a recognized European expert in legal semantic systems and a visiting researcher at Cornell Law School earlier in this decade, wrote in February 2010 in an article on VoxPopuLII.

"[I]n the current legal ontology literature there are few explicit accounts or insights into the methods researchers used to elicit legal knowledge, and the accounts that are available reflect a lack of consensus as to the most appropriate methodology."
See Semantic Enhancement of Legal Information… Are We Up for the Challenge?, VoxPopuLII (February 15, 2010).

She also wrote about techniques for legal ontology engineering:

"[I]n order to represent knowledge in practice, legal ontology engineering could benefit from the use of social science research methods for knowledge elicitation, institutional/organizational analysis (institutional ethnography), as well as close collaboration with legal practitioners, users, experts, and other stakeholders, in order to discover the relevant conceptual models that ought to be represented in the ontologies."

Our own methods, honed over years, represent a combination of approaches, essentially an "all of the above" strategy, for acquiring and building our own legal domain model and transactive memory system. Sources range from the analysis of primary law texts, both statutory and case law, to articles, blog posts, lawyer electronic discussion group postings, live term and relationships captured during continuing legal education course presentations and speeches, magazine and newspaper articles analysis, legal conference videos, and many other sources. We've even found a secondary use for CLE brochures, which often feature new or difficult issues to entice practitioners to register.

There is quite a spectrum in the range of number of relationships that individual terms have, with some terms at the intellectual core having hundreds or even more than a thousand direct relationships to other terms, and others having few to none (depending on work in progress and the evolutionary survival of the fittest environment characteristic of networks in which preferential attachment is the norm).

Not all Laugment relationship types are created equal. Some are primarily navigational or linguistic. Others are valuable for classification and modeling, and yet others may provide mentoring level insight.

That is why we rank our relationship type vocabulary on an ascending scale of intellectual value. Some relationship types matter more than others. When we display a particular term's network of related terms, they cluster according to the highest intellectual value relationship types first.

Laugment Relationship Language and the way we use and manage it is, of course, proprietary, and so we are unable to share details in this document. We started with a few relationship types more than a decade ago. We now have more than 250. A model doesn't have to be fully computable to be valuable to a professional who appreciates suggestions and reminders even when direction and formal inferencing is not possible due to the nature of the domain itself.

We understand the West Key Number System® to have approximately 100,000 topical points used for classifying appellate case law. The Key Number System is a classical taxonomic hierarchy of a defined textual domain. It's not intended to cover what lawyers do or clients want. It's intended to dissect and organize what appellate courts write about in their formal opinions.

Therefore, to compare Laugment to the West Key Number System is to compare apples and oranges. But it is about the only other organizing framework with which the legal profession is familiar.

Laugment does not have a hierarchical system, although it has many hierarchies within the system. Laugment is a polyhierarchical and indeed multidimensional network that constantly evolves and expands, one designed to stand on its own independently of the legal document universe. Laugment can do all these things because it is built with our LawSaurus knowledge authoring, editing and publishing application. See Marc Lauritsen, The Lawyer's Guide to Working Smarter with Knowledge Tools at 93 (ABA Law Practice Management Section 2010) (LawSaurus "offers a stunningly generic architecture for any information management problem that can be expressed relationally.")

The Key Number System, by contrast, is the structure by which documents are abstracted and assigned to consistent categories over generations.

Laugment is, in short, more natural; more bottom up than top down. It's descriptive without the intent to be prescriptive - provocative of thought; evolutionary; holistic; reshapeable.

Home | Enter via LawMoose Subscriber Edition | Login locally

Thanks for visiting us on July 20, 2024. This domain is currently a place for our occasional experimenting and sharing without any pretense of adding style. You're on the server.

Contact us for more information!

© 2024 Pritchard Law Webs, Minneapolis, Minnesota. Ph. 612-332-0102.

Laugment, Lawyer Augmentation System, LawSaurus and LawMoose are service marks of Pritchard Law Webs.