Virginia Court Counsel and Fee Waiver Bot Journal

By Loren Chen

My name is Loren Chen, and I am 2L at Suffolk University Law School. I wrote a bot for the class Coding the Law, where each of us chose a jurisdiction (I chose Virginia) to research and develop a chatbot that would determine if an individual would be able to receive court counsel or fee waivers. You can find the bot here. To write my bot, we used a tool called QnA Markup. Basically, you create an interactive decision tree by nesting questions and answers. Here's my code.

User Testing

In this process, I did peer testing with my partner, collected feedback anonymously, and reached out to a lawyer in Virginia.

My partner provided me with this feedback:

Feedback from Jeffrey Price

I felt that this was very useful and helpful in starting my refining of the bot.

I collected feedback anonymously using Google Forms and obtained these results

While most of the feedback was primarily used to gauge the progress of my bot, some of the comments were very well thought-out and furthered my refinement of the bot to not only do its job, but also be user-friendly.

Finally, I reached out to a lawyer in Virginia for feedback. Here was his feedback:

Lawyer Feedback

I feel like his feedback was insightful and coupled with the earlier feedback I received allowed me to craft a bot I am proud of.

Research and Refinement

I went through several refinement steps.

In the first draft, there was the rough ideas of what the jurisdiction required and creating templates for both the civil and criminal aspects of the bot. After a bit of refinement, I sent a copy of the bot to my partner, in which I added in basic calculations and javascript functions. Following the feedback from my partner, I refined the bot to include more cues and also included the more automated calculations. For the copy I sent out for anonymous feedback, I refined the bot to include links to clarify and to cite appropriate law.

For my research, I looked into the relevant laws in Virginia. The two main relevant pieces of law are § 17.1-606 and § 19.2-159. I also looked at this for clarification on how to classify the various crimes in terms of determination of counsel (which was based off of § 18.2-11. Based on the feedback I received, I also researched on the actual determination of the equivalent of Miranda rights in Virginia. In Virginia, the prerequisite of the appointment of counsel is § 19.2-160.

The most important refinements needed were making the bot more accessible to users. This included adding more clarifying information and fixing up the interface (to compensate for the GOTO tags). In addition, I tweaked the calculations to work properly (as I had set it up improperly).

The only bugs/improvements I have not been able to address were getting all the fields to be auto-populated and how to let the bot record the data when going through the process again without quitting the bot.

I also added in this photo to replace the default one. The photo is licensed under CC0 Creative Commons (Free for commercial use and no attribution required).

Real-World Viability

This bot is likely viable as-is. However, there is always room for improvement. If I would like to make it more polished, I would have a more robust set of values that are auto-filled. Currently it only has a small list of values that are recorded, and to make it more viable, it would provide the option to transfer that data to the pdfs of the fillable forms. In addition, I would collect data from both the criminal and civil sides for that data transfer. In addition, I would continue to accept feedback to tweak any bugs that are found.