Legal Innovation & Technology Lab@ Suffolk Law School
A reverse chronological list of talks and presentations from our team members.† When possible, we include video and a copy of the slide deck along with information on the venue. If you are looking for a talk given at an event hosted by the Lab, but not presented by a Lab member, you should consult the event's webpage or our collection of LIT Bits.
† Sessions are grouped by event (e.g., a conference) and presented chronologically within groups.
Data Science Demystified: A Crash Course for Attorneys, Boston Bar Association
May 23, 2018 (David Colarusso). Learn how to wrangle data, spot patterns, and predict the future with machine learning in this hands-on crash course. Working with data from a simulated law firm, data scientist and former public defender, David Colarusso, will share free tools and methods for the analysis of data, big and small. No prior experience required.
Before the session, you will be provided with a set of preparatory materials, including a data challenge. Should you choose to accept it, you’ll spend the session learning and building your own machine learning algorithm, with the option of entering it in a battle of the bots against your peers. Either way, you’ll leave with an introductory understanding of machine learning and a set of tools that will allow you to transcend the lawyer’s default use of Excel for data handling.
Attendees should bring a laptop with them as this program will be interactive.
Session Info |
What Every Clinician Should Know About Digital Security and Using Technology to Leverage your Impact, AALS 2018 Conference on Clinical Legal Education
March 23, 2018 (Kim McLaurin & David Colarusso). Presenters from Suffolk Law School discussed how Suffolk is integrating technological training across its clinical programs, as a possible model. The session included a description of Suffolk’s Accelerator-to-Practice Program and newly created Legal Innovation and Technology (LIT) Fellowship Program and Legal Innovation and Technology Law Lab (“LIT Lab”), in which students provide consultation and legal technology and/or process mapping services to organizational clients such as courts and non-profit organizations.
Coding for Lawyers (C4L) Summit, Suffolk University and Legal Hackers
April 10, 2018 (David Colarusso).
A number of law schools have begun to offer "coding for lawyers" (or CS for lawyers) courses to students. This invite-only summit convened C4L professors/instructors to explore the purposes, methods, and future of these courses, with an overarching goal of being descriptive, rather than prescriptive. Specifically, rather than ask "should lawyers learn to code?," this summit asks why instructors are teaching lawyers and law students how to code, and the pros/cons of various approaches. This summit was planned by Lab Director David Colarusso in collaboration with Legal Hackers.
LIT Lab (A Case Study in Clinnovation), Suffolk’s Clinnovation Conference: Where Legal Innovation & Technology Meet Clinical Pedagogy
April 9, 2018 (David Colarusso). The Director of Suffolk's LIT Lab, David Colarusso, introduces the Lab, followed by a student-led sharing of their work with real-world clients (videos below). Projects span the selection from data analytics to document automation.
Geography-based Youth Service Finder, Suffolk’s Clinnovation Conference: Where Legal Innovation & Technology Meet Clinical Pedagogy
April 9, 2018 (Nicole Siino).
When a juvenile is facing delinquency charges, courts often order youth to locate and participate in community programs as a condition of probation. Their ability to find and successfully participate in these programs can have long-lasting impacts on their sentences - and their futures. A constellation of social services exists for youths. These services, however, are offered by a variety of agencies, public and private, and make use of differing eligibility requirements. Additionally, access to such services is often dependent on physical proximity. Consequently, service discovery is a time-intensive process, and exhaustive exploration of services by youth and youth advocates is often impractical. For this reason, the lab is working on the creation of a tool that will allow users to filter service providers by location, age, and service type. Current thinking is focused on the creation of a map-based tool that allows users to search for and find relevant service providers as one might look for a restaurant an Yelp. Initially, this web application (website optimized to look good on a phone) will concern itself with service providers in the Boston metro area. However, the technology being used would allow for the expansion of this application to include expansion into more service types and areas. This will include documentation to help others create their own instances of the app.
Visit Working Web App. |
How to Make Your Own. |
Aiding Access to Justice through Letter Automation, Suffolk’s Clinnovation Conference: Where Legal Innovation & Technology Meet Clinical Pedagogy
April 9, 2018 (William Bean, Corby King & Anthony Metzler).
For years, HomeLine, a nonprofit Minnesota tenant advocacy organization has offered a number of legal form letters for download on their website (e.g. repair requests and security deposit demand letters), but many people in their service population don't have easy access to a printer or the internet outside of their phone. Consequently, they would like to produce a "chatbot" that helps users fill out and "mail" letters from their phones to their landlord addressing their housing issue. One of the board members likes to talk about "doing it all with your thumbs."
Visit a Beta Version of the Web App. |
Codification of Tacit Knowledge via Algorithm, Suffolk’s Clinnovation Conference: Where Legal Innovation & Technology Meet Clinical Pedagogy
April 9, 2018 (Jeff Price & Charmain Wood).
Palace Law is looking to encode its client selection criteria into an algorithm. Palace Law is a personal injury and workers’ compensation firm. Currently, when speaking with potential clients, they ask a number of questions about their case (e.g., are you still working, what type of injury do you have, etc.). An experienced employee then makes a determination as to whether or not something is a "good" case. Their goal is to teach an algorithm to assist in this classification by training it on historical data from potential clients. Consequently, they are in the process of establishing a data collection framework. Their ultimate goal is to discover predictors of a good case, where the measure of a good case takes into account a number of post-disposition metrics. In the near term, however, a replication and codification of existing decision criteria will work as a strong proof of principle. Their hope is to produce a classification that scores new calls with a quality ranking (e.g, 0-100).
Navigating the World of Public Benefits, Suffolk’s Clinnovation Conference: Where Legal Innovation & Technology Meet Clinical Pedagogy
April 9, 2018 (Alessandra Ambrogio & Michael DiFilippo).
There are a constellation of social services available to low- and moderate- income individuals and families in Massachusetts. These services range from food stamps, to fuel assistance, to free transportation. These services are offered by a variety of agencies, public and private, and make use of differing eligibility requirements. Consequently, service discovery is a decentralized and time-intensive process, and most people never avail themselves of the full spectrum of services to which they may be entitled. As such, an exhaustive exploration of services by potentially eligible parties and their advocates is often impractical. Unlike the youth service finder tool, this tool will focus on surfacing relevant services for low- to moderate-income individuals with moderately complex eligibility requirements (e.g., multiple interacting criteria and levels of service such as TANF). That is, unlike the youth service finder, which acted like a directory for service providers, this tool is focused more on finding specific programs tailored to a person’s eligibility and needs. This will be accomplished by first eliciting replies to a series of guided interview questions - all geared toward identifying client need. These replies will be taken as inputs and used to determine the likelihood that one would see a benefit from a given program based on a set of predefined rules and machine learning algorithms. A set of services will then be presented and explored based on an attempt to prioritize those with the highest potential benefit for a given person seeking services.
Visit a Beta Version of the Web App. |
LIT Lab: What Next?, Suffolk’s Clinnovation Conference: Where Legal Innovation & Technology Meet Clinical Pedagogy
April 9, 2018 (David Colarusso). The Director of Suffolk's LIT Lab, David Colarusso, comments on the above presentations and provides his vision for the future of the Lab.
Legal Innovation Lightning Round I, Stanford’s CodeX FutureLaw 2018 Conference
April 5, 2018 (David Colarusso). Margaret Hagan, David Colarusso, and Metin Eskili discussed their work leveraging Machine Learning and Design Thinking to help address access to justice issues, including the development of a tool for the crowd-sourced labeling of legal data—Learned Hands. The talk spans roughly the first 14 min of the embedded video.
Are You Competent to Practice Law + Q&A, Vermont Law School's Annual Solutions Conference
March 23, 2018 (David Colarusso). Colarusso participated as panelist, sharing his experience working on the Dookhan matter and discussing the importance of technical literacy.
Legal Hackathon Lab, Vermont Law School's Annual Solutions Conference
March 23, 2018 (David Colarusso). Colarusso facilitated a #FlashHack (mini hackathon) in which he taught attendees QnA Markup, helped them brainstorm project ideas, and watched as attendees created working prototypes, all in 45 min.
Speaking the Same Language: Data Standards in the Administration of Justice, LSC's 2018 Innovations in Technology Conference
January 11, 2018 (David Colarusso). If an argument over the use of one or two spaces post-punctuation has ever concluded with blows, you can be sure a lawyer threw the first punch. Form matters. Deciphering an author’s meaning is a tax on their argument, one the legal community has done its best to abolish. Form matters in part because it should not matter. When the final product is a printed brief, no self-respecting court would consider forcing parties to adopt a monolithic technical solution, for example, requiring that all filings be typed on a Underwood Universal. However, in the digital realm, many courts are happy to do just that. This talk explores how data standards can work to abolish the decipherment tax in the digital realm. It is based largely on Data Standards and Disruptive Technologies in the Administration of Justice, 50 Suffolk U. L. Rev. 387 (2017)
| Session Info
| Slide Deck (Image Credits in Speaker's Notes)
Averting the Law's Robopocalypse, LSC's 2018 Innovations in Technology Conference
January 11, 2018 (David Colarusso). Human decision making is flawed. The promise of machines making data-driven decisions might seem like a welcome innovation. However, such systems are themselves subject to bias. Created by people and trained on historical data, algorithms can bake in existing prejudices behind a gloss of mathematics. You can prevent the Law’s Robopocalypse by helping to shape our digital future. Lawyers must echo the words of Justice Warren and insist on an answer to the question “is it fair?” while learning to partner with technology to augment, not replace, the practice of lawyering. To do this, they will need an understanding of the justice system informed by randomized control trials and rigorous evaluation, and powerful open source tools that work to address market failures in legal tech. Such a future, however, depends in large part on access to accurate, standardized court data and an understanding of its importance. In this session, participants will learn specific steps to help create such a future while gaining an introductory understanding of algorithmic bias, data standards, open source software, and randomized control trials.
| Slide Deck
How the Criminal Justice System Should Think About Thinking Machines: Avoiding Algorithmic Bias et al., Indigent Defense Research Association meeting (part of the Annual Meeting of the American Society of Criminology)
November 15, 2017 (David Colarusso). Human decision making is flawed. Evidence suggest that judges impose harsher sentences on less sleep, avoid changing the status quo on low blood sugar, and unduly punish defendants after their football team suffers a loss. Consequently, the promise of dispassionate untiring machines making data-driven decisions seems a welcome innovation. However, such systems are themselves subject to bias. Created by people and trained on historical data, poorly-crafted algorithms can bake in existing biases, obscuring them behind a gloss of mathematics. It is necessary to carefully consider how such models are used and constructed. An error-prone risk model may be welcome when used to help triage scarce social aid but morally unconscionable when used to determine sentencing. This presentation will explore the question of how one should best approach the construction and use of such models, including a focus on understanding what it is a model is optimizing (its definition of success) and how this constrains its use.
Session Info |