DocuBot is a tool to fill in legal documents and other forms through an SMS or other chatbot-like experience. The bot asks questions to fill in the form.
Here is more information from its creator, 1Law.
1LAW is proud to announce the creation of Docubot™, a legal document generating artificial intelligence. In conjunction with some of the best lawyers in the United States, Docubot is drawing on form databases of 1000’s of legal documents. Docubot will assist individuals with legal queries as well as generate documents for them. To help serve Legal Aid, Docubot will allow users to interact via SMS text.
Tech specs:
Written in Go at the server
Powered by Watson – Watson rest API
Swift on the iOS side
Communication via Websocket protocol
Back and forth handled through Websocket
Output –Everything is encrypted
The document is generated using a headless webkit browser that takes an HTML document and outputs a .pdf which is stored in a private S3 bucket and then a short-lived url is generated and sent to a user and each time a user loads the thread they will be given a new url. Document is backed up on the S3 server.
Contact us at: info@www.1law.com for more information.
Last Friday was the final class in the Stanford Law School/d.school class Prototyping Access to Justice. Kursat Ozenc and I were teaching the course as a practical, service design effort.
The big question guiding the work: if hundreds of thousands of Californians go to the courts to deal with their divorce, child custody, debt, and housing problems — how can we make the courts work for them, on their own terms? We know that growing numbers of people are trying to use the courts without a lawyer, but that the courts have been designed for lawyers — with complex procedure and intimidating jargon is so complicated that only lawyers can really figure it out.
Students were given initial design briefs that we had crafted from our earlier research into California Courts’ Self Help Centers last year. In the first version of this class, we followed litigants through their court journeys and interviewed professionals to identify key opportunities and breakdown points.
This quarter’s classes aimed to use this groundwork to jump more quickly into prototyping and testing. Each of the design teams worked on site at San Mateo County or Santa Clara County courthouses, and at the Stanford d.school labs — going through 3 cycles of scoping out a concept design, making a prototype of it, and testing it with many different stakeholders.
We ended up with seven proposals for the courts to pilot. Two concerned how to remake the court building and design of physical space. Two were new modes of guides, to present better ways to guide litigants through complicated tasks. One was about better form completion. One was about new modes for court feedback. And one was about better preparing court users before they come to court for the first time.
The teams made videos, maps, and presentations to capture their proposals, and we present them here for you to review. We ask you for your feedback now — because we are vetting these seven proposals to decide which to continue working on and possibly pilot with the courts.
1. Redesign of the Court Building: Visual Lines + Signs for Empowered Wayfinding
Team Chuka Ryori were tasked with helping people just as they arrived at the court the first time. How could we make people feel more supported, less confused and intimidated, and more capable of getting through the process efficiently?
Their proposal is to launch a coordinated, color-coded, pictogram-based wayfinding system in the court building. There should be color paths on the floor for the most common user destinations, with pictograms and a palette that supports finding the right place.
They did guerrilla-design work, by “decorating” the actual court with new lines, signs and pictograms to test how users reacted. The results were overwhelmingly positive. Our next steps are to refine the color palette and pictograms, and then work with the court to implement the new lines and signs.
2. Redesign of the Court Building: Respectful, Transparent Line Waiting
Team Golden Design Warrior was focused on the next moment in the user’s journey, when a person found the Self Help Center, but now must deal with the long and confusing wait to get services. After several different ideas to change the layout of the space, the team moved to focus on how to set up lines that gave users greater transparency and more comfort while waiting to be served.
The team identified that people were rushing to wait in a confusing line. They were stressed out, and in turn stressing out the staff who felt as if they had to barricade themselves in against a huge amount of people who wanted things from them. The goal of the system is to give people a clear ticket that would give them an explicit place in line, and would let them relax, sit down, and see when they could expect to be served.
The first pilot is just with laminated cards and a person distributing them near the entrance. Then it can be scaled to an automated ticketing service.
This prototype has tested remarkably well with both litigants and professionals, reducing both stakeholders’ stress and giving them more of a sense of control. With the simple intake during the sign-up, the professionals can better prep for the clients’ cases. They also get insulated from the pressure of a huge group of people hovering around their doors.
The joy of this design is how a simple service intervention can have a huge experiential payoff — making the experience of visiting court or working there be less anxious, confusing, and stressful.
3. Visual Book Guide to Following a Legal Process
Instead of worksheets and forms, or instructions told out loud before a person leaves the Center, how do we convey instructions and guidance to them? How do we make it easier for them to follow the procedure, so they stay on track and get it all completed correctly and on time?
Team Jiffy Justice proposes a visual booklet, that gives people a step-by-step map of what the process will look like, what to do, and how exactly to finish the steps. It’s about envisioning, modeling, and taking legal actions out of abstract text language, and into clear, grounded situations.
The team made a map that can be printed as a poster, a handout, or part of the book. It gives the systems-level view of the case. People liked this as an orientation material, but still wanted more detail about exactly what each of these steps entails. A high-level view helps give a person the mental model of the system, but they want to dig into more specific instructions and strategies.
The team made the booklet to enhance the guide, to go from the map to the detailed instructions.
They built it specifically so it could be easily printed on common paper sizes by the Self Help Center. It incorporates the map, but then with details of the forms, the filing info, and common flags and warnings.
The next, scaled-up version of this would be a digital version (most likely on mobile) that has the step-by-step guidance and the map for the person to follow along as they go through the process.
4. Text-based coaching through complicated process
Team Exit took this same challenge — how to help people through complicated procedures that they often fail at? Their proposal is more tech-centered, harnessing the power of the mobile phone. They created a prototype of the RemindMe Text system, in which litigants would get coaching reminders, customized due dates, and clear blasts of instructions about what to do to serve process (a particularly thorny part of a process, that people often screw up).
The team embraced the principle of staging information and providing it at the right moment and context. Rather than give huge worksheets with general information all at once, segment it into specific messages and customize it with the user’s own information.
This program could later incorporate other kinds of messages, beyond reminders — including the maps and visuals that Team Jiffy Justice had in their booklets, or the wayfinding and prep materials that other teams proposed.
The great part of this proposal is that the text message channel, opened up between the courts and the litigants, can allow for a diversity of services to be provided in the future. As more technology is developed for court services, they can be integrated into this same channel.
5. Prep People Before Court with Warnings and Key Info
Even before people come to court, how do we make sure they come prepared to make the most of the day — and not waste it? Especially if it takes several hours to even get to speak to someone at court, how do we make sure people come with the paper, translators, and knowledge enough to get their tasks accomplished?
PrepMe is an idea to do better outreach around this Prep information, via websites, mobile apps, and other court materials. It should be in multiple languages, and show very prominently the most common prep information people don’t know: about translators, child care, and timings.
This information can be presented also in court correspondence, posters, fliers, and any other ‘touchpoint’ where people are thinking about using the court system and planning for how to do it.
It prioritizes language access as a fundamental principle of design of court information, rather than as an add-on afterthought.
6. Help People Fill in Forms Better
One of the big failpoints in the legal process is the correct completion of forms. Team Remind proposed two prototypes — one paper-based, the other tech-enabled for improving litigants’ ability to complete Service of Process forms.
The paper-based system involves tagging up and creating a model completed form, that would guide a person through exactly how to follow this model.
The tech-based guide uses a Google Doc form to let people enter in the key data points, and then uses Python to fill in the form with this data. The litigant (or the process-server) never needs to see the Judicial Council form except when they print and file it. The Python script does the completion for them.
The vision of this prototype is to have a 2-pronged tech/paper strategy, so that resources are allocated to different types of users in the system. It is also to come up with cheap hacks to use the power of technology. Rather than contract with an expensive, proprietary vendor to provide for form-filling, the goal here is to mash together existing, modern, mobile-friendly services (like Google Docs) to get a very cheap and quick working system of filling in forms.
The other big insight here was in the power of having an interdisciplinary team, with lawyers and computer scientists working together to find the most strategic uses of technology that would serve the legal system. Lawyers should know the power of Python — a major takeaway for our partners.
7. Gather user input and experiences to feed back to the Courts
Team Law4U drafted a prototype of a kiosk in the Self Help Center’s office, that would ask simple questions from people as they’re waiting to get service. They’d be able to rate the court’s quality of service and give ideas for improvements.
In the future, this program could also recruit litigants to join a Standing User Testing panel, in which they’d be compensated for reviewing new court efforts or giving more feedback to the courts. This would feed into a broader culture of testing and experimentation in the system.
—
These seven prototypes are the result of 9 weeks of hard, creative work by our Prototyping Access to Justice class. Many thanks to the wonderful students and coaches!
We are soliciting feedback now on these prototypes, so that we can then proceed to pilot implementations of some of them in the courts. Let us know what you think!
I took a photograph of this display in London Heathrow Airport, Terminal 5. It is a very public display of the customer feedback for the airport. It has the results of surveys for different factors of the airport experience, displayed right on the monitors that show flight times and other important information.
A sketch from my notebook, while I was observing a waiting room in a Court Service center in Boston, for people who were waiting for help with housing cases.
This change in the document accompanies more systemic changes.
THE PLAN
Steps to ensure that people who receive a summons appear in court include:
A redesigned summons form that makes the date of appearance easier to understand. The City and courts worked with ideas42, a non-profit behavioral design lab, to redesign the summons form making information easier to understand in order to better prompt people to return to court. Additionally, the new form will collect individuals’ phone numbers and include a phone number and website where recipients can access their cases, view when their court appearance is scheduled, and determine whether they have outstanding warrants. The website will also have translated copies of the summons form. The new form will be operational this summer. A comparison of the old form and the new form, and additional information about the science driving these changes, is available above. This effort was funded by the Laura and John Arnold Foundation.
A reminder system to ensure defendants appear in court. The courts will test a number of different reminders citywide during the fall of 2015, using both robocalls and text messages. The results of this pilot will be carefully examined and the most effective method will be scaled up citywide.
Flexible appearance date and night court. Sometimes people who want to make their court appearances and resolve their cases sometimes have work, family or other life conflicts with their court dates. Beginning with a pilot in Manhattan North, individuals who have received summonses will be permitted to appear any time a week in advance of their court appearance. The court will also be open until 8:00 p.m. on Tuesdays. If these pilots increase court appearances, the programs will be scaled up citywide.
Steps to enhance transparency and improve the quality of justice in summons court include:
Publicly available quarterly data. Four times a year, the City will post data showing summons activity broken down by charge and precinct, allowing access to enforcement trends on the neighborhood level. Going forward, the police department’s annual report will also include summons activity with details such as the location at which a summons is issued and the race of the recipient. Explore this data.
Real-time, electronic access to case files. Since April 2015, the Court has been providing defense attorneys with tablets that provide them with all of the factual allegations docketed for that day so they can better advise their clients and quicken the adjudication process.
Online payment of fines. Beginning this summer, the courts will implement a new process permitting people convicted of summons offenses to pay fines online.
Training on collateral consequences for court-appointed attorneys and Judicial Hearing Officers. Since late March 2015, court-appointed attorneys and Judicial Hearing Officers have been trained in how to better advise clients about the collateral consequences associated with summonses.
Ideas42, the Behavioral Design Lab, has more details about how they chose to redesign the form to better work with people’s behavior and biases.
One of the insights we discovered was related to New Yorkers’ mental models, or personal understandings and beliefs, based on impressions or anecdotal information, of what the court experience would be like and what it represented. In this case, attending court was strongly associated with lost work time and wages, unrealistically steep fines, and uncomfortable experiences. These perceptions were strong enough that some people who were fully aware of arrest warrants for FTA still chose to miss their court date. In doing so, they were avoiding a negative experience now while risking a far worse one—arrest— down the line. Furthermore, many people who received summonses for small offenses—like spitting or littering—felt that having to go to court was a penalty that far outweighed their mental model of the violation’s seriousness. In other words, the punishment didn’t fit the crime, so they chose not to appear.
Some of these negative perceptions were further compounded because many New Yorkers affected by summonses are lower-income and experiencing some form of time and/or resource scarcity, making juggling workday court appointments and far off dates even more difficult. For those experiencing scarcity, it can be challenging to change work schedules or make other necessary arrangements to appear in court. Many don’t have steady work shifts, making the 2-3 month delay between summons issue and court dates an additional hassle. During this lag time, some people we interviewed reported experiencing unpredictable changes in job or housing, making it even more difficult to plan ahead for a court date.
With these and other insights from behavioral science in mind, we redesigned the physical layout of the pink summons ticket. The new form prominently features the appearance date and court location at the top of the ticket, where people are more likely to see it (the old version had this information at the very bottom where it was easily overlooked). The new form also clearly states in bold typeface that missing the assigned court date will lead to a warrant (important information that was completely absent from the previous form). Behavioral science tells us that simple tweaks like these can have an outsized impact on how people act.
CAIR Chicago has sponsored a new initiative to mobilize legal help and interpreters (as well as knowledge) for people at risk of civil rights violation or immigration problems. Their Travelers Assistance Project was launched to give travelers alerts, assemble an Attorney Corps, and an Interpreters Corps.
T.A.P CAIR-Chicago’s Traveler’s Assistance Project, a first of its kind nationally, created to help travelers secure legal help in light of the #MuslimBan. T.A.P is ONE stop shop. It includes an online system that collects at-risk traveler information and puts it in a queue for our in-house Attorney Corps and legal teams of our partner organizations to streamline on our ability to assist on the ground. It provides important traveler “know your rights” updates. Lastly, you can sign up to join us as a volunteer attorney or interpreter.
Dulles Justice Coalition is a grassroots organization in the DC area, in which lawyers have come together to provide help to immigrants. Specifically, it arose after the January Executive Order that upset the travel plans and border-crossings of refugees and people from Muslim-majority countries.
They formed as an impromptu group and also stood up a website to take in requests for legal help.
They also embedded legal precedents and other research as resources for lawyers to use. They created a lengthy Resources for Lawyers google doc, with definitions, case precedents, and other files and references for pro bono lawyers to use.
Today we held our Prototyping Access to Justice class on-site at San Mateo County court house, specifically in and around the Self-Help Center and Family Law Facilitator.
The six student teams are all at the point where they have working prototypes that they want to test. They each have hypotheses about how they can make the legal system better for people without lawyers, and have embodied these hypotheses into a new tool — digital- or paper-based.
Instead of our usual class setting at a design studio at Stanford’s d.school, we created an impromptu class space in the Waiting Area on the 2nd floor of the Superior Court, where people are lining up to see people at the Self- Help Center, or are waiting to be called for an appointment. Some of the teams also set up testing spaces inside the Self Help Center, for when people had down-time after they had filled in forms or were waiting for next-steps.
The teams sought out people to give quick feedback, as well as longer experiential testing. They had interactive click-through prototypes of digital tools, paper mockups of new tools, posters and floor pathways for navigation, and tablets with new feedback forms. They had gift cards to give to user testers, to compensate for their time.
They tested their prototypes in small groups — with some taking notes (or translating into Spanish) and others leading the questions. They also had designer and developer coaches with them, to help them spot new opportunities and to run the testings.
Takewaways
So what were the takeaways? I was able to pull out some high-level insights during my debriefs with each team, as well as some specific points for improvement.
1. The forms are too many and too complex. This was a refrain that each team heard from users, no matter if their questions and prototype revolved around forms or not. If there is one big message that family law litigants have for courts, it is: make your forms easier to understand, and easier to complete.
There is an overload of paperwork, that is laid out in a way that does not make sense to people, and overwhelms them.
2. Little things about court — like parking, way finding, and security checks — have a big influence on people’s experience. Though we as lawyers might think about the legal procedure, forms, and hearings as the main determinants of people’s procedural justice and sense of fairness about court, there are other more pedestrian factors that shape their time with the legal system. If parking is difficult, expensive, or with a ticking timer, this puts an extra layer of pressure and confusion. If the security guards doing initial checks at the door are adversarial or cold, this raises the stress level of people and sets them off on a bad foot. If there is confusion about where they are going or how to get there, people lose confidence in themselves and feel that they are wasting time and not being strategic.
3. Pathways on the Floor should be implemented immediately. Our team Chukka-Ryorui, who are focused on improving navigation, put down a dotted red line from the building’s entrance to the Self Help Center on the 2nd floor. They used masking tape to make the line — and it took less than half an hour to implement. The feedback was universally positive. People were able to follow it and understand it without any complicated explanation. Users reported that they already are familiar with this pattern from hospitals, and appreciate having it here. They want bold color lines that they can follow easily, along with complementary signage.
We recommend that courts implement colored floor pathways for their most popular routes: Self Help Center, County Clerk, and Jury Services primarily. This is a relatively cheap intervention (vinyl floor paths are not that expensive) that can have a major impact.
4. “Out-of-Court Homework” Tasks must be Modeled + with reminders. As we heard from litigants and from staff, the most common fail points are around all the tasks that the litigant cannot do at the court, but must do outside. Getting service of process done, and done correctly, with the paperwork noted correctly with address, date, and other pieces of data is a very common failure. Also, remembering to get this done at the right time is also a fail point.
Some of the recommendations in this space is to have more reminder services that proactively reach out to the litigant to tell them they have to do this task before their service.
Also, the demand is for models of forms that have been done correctly, with annotations about why it is correct and how to do it right.
5. Be Mobile First, with guides and tools for the phone. The overwhelming majority of the people we spoke with have mobile phones, and are willing to use them to get legal tasks done. Tools must be built for phones, not desktops.
We are setting a bounty for the best new product that lets people understand processes and fill out forms using the phone. Even if this is not ideal — even if we wish that people would have the big screens of a desktop computer when they’re doing complex processes, they will be using mobile phones and paper handouts most of the time in practice.
Six. Maps are key. The team that was testing out a giant, slightly comic-based map of child custody process got great reviews. People responded well to their characters and to the map-based view.
They are thinking in terms of both a paper-based map (that could be a wall poster of a general ‘happy path’ of how the process works in the ideal, combined with a booklet of in-detail maps that include detours). And in terms of a digital map, that could be zoomed in.
People were able to instantly figure out the paper based map. They know how to use it. The digital map was harder — people were more hesitant to use it, and to know what to touch on the screen in order to see what would happen.
—
More insights to come as our class proceeds — stay tuned for more of our design work and proposals for making Self Help Centers and the legal system more user-friendly.
I am also quite excited about setting up a more regular pop-up design lab on site at courts and other points in the legal system. To create more relevant and interactive designs, having input directly from litigants and court professionals is highly valuable. And doing the prototyping in the environment also helps the designs better mesh with this particular context, and what affordances and opportunities already exist there.
One of my Brazilian students in my Prototyping Access to Justice class alerted me to a very cool app in Brazil, all about empowering people about their legal rights.
It is an in-your-pocket tool for a citizen, to know what their legal rights are in a given situation. They can open the app, pick which domain their scenario is in (medical, at work, at a store, etc.). Then they can find the specific situation they’re in, and when they click on that — they get a quick, easy summary of what their rights are, and what ‘magic words’ they can say to assert the law.
Here’s a video of it in action.
And another video explaining it (for you Portugese speakers!).
The name of the app is a play on ‘badge for good’. Just like a police man can flash their badge to assert their legal authority, a person can pull out this app and flash the words and legal citation to assert their rights.
You can save common scenarios, to have them at the ready.
It’s a great model — one that I am exploring replicating for here in the US. Putting the essential law in people’s hands — in easy to use and easy to understand modes. Let me know what you think!
In our Prototyping Access to Justice class, Kursat Ozenc and I are leading student teams to get quickly from speculating about how the courts could be improved to implementing new concepts.
In our class today, in week 3 of the course, we had the students make some more progress along the Journey of Prototypes.
The journey metaphor is useful to hep anyone who has an idea for how a system (like courts or self-help centers) could be better — but doesn’t know how to make forward progress on it. We are creating a staged set of prototypes to develop, to go from the spark of an idea to a vetted, feasible concept.
Last week, the students had digested their research into how the Self-Help centers currently function. From this, they made journey maps and spotted opportunity points for intervention. Then, they brainstormed 30 new concepts for each intervention-point.
For example, what are 30 new ways to make sure a person can find their way from the entrance of the court to the right office? Or, what are 30 new ways we can help people get service of process right, and not fail here?
This was the first stage of their Prototype Journey — one line ideas on a post-it.
This is the most basic capture of the idea — hard to even call it a prototype. But it’s the seed that starts to lay out what the ‘it’ is, that will guide the rest of the design cycle.
From the post-it, the next stop is the Concept Poster — a slightly more robust prototype. We had the students sort out all of their post-it ideas into a hierarchy, and choose the top 6 ideas they thought had promise to be breakthrough and feasible.
For each of these top 6, they drafted a Concept Poster version of the post-it’s concept, that added more details on functions, target users, high-level value, and implementation requirements. The Concept Posters were then compiled into a catalogue, of the Ideabook.
The value of the Idea Poster as a prototype is that it can let your team quickly, visually present a range of ideas to experts and users. They can almost go ‘shopping’ through the ideas and figure out which ones are most exciting, do-able, and worth pursuing. It can help the team whittle down further to the best one to pursue first, or to reorient based on this initial input.
The Concept Poster/Ideabook prototype can also be a check-in, with the testers letting the team know if similar ideas have been tried before, could be replicated from elsewhere, or if the team should be talking to certain experts.
Once the team gets Ideabook feedback, then they can focus in even further to the first idea they want to pursue. For this single idea, they will flesh it out to the next stage of prototype: the Storyboard.
They go from one sketch of the idea to a visual flow of the entire, ideal journey the user has with the new concept. It lays out the key steps of how the concept will work, what the user will have to do, what the system/service-provider will have to do, and what the resolution looks like. All of this is the ‘happy path’ version of an implementation, where everything goes right and works out.
Once the team documents this ideal flow, they then use the prototype to identify the key assumptions they’re making about the user, the service-provider, and the system. By laying out the step-by-step of the ideal, they can see what needs to happen or exist in order for this ideal to be realized. The prototype helps the team spot what they need to test.
From there, the team quickly moves to the next prototype mode: The Wizard of Oz version. For each of those assumptions they’ve discovered in their storyboard flow, they have to make a way to behaviorally test whether it will really work the way they expect it to.
A Wizard of Oz prototype is an interactive realization of the idea. The user is not just reacting to a presentation of an idea — she is playing around with the ‘app’, or moving around in the new ‘space’, or considering whether to visit a ‘website’. The design team does not actually build a functional new app, space or website — but they use paper, sketches, enactment, and other hacks to simulate the concept.
The value of this prototype is getting more realistic feedback, about whether people will really engage with the thing you’re creating, if they’ll be understand it, and if they’ll get value from it. You can test the many assumptions baked into it.
Some common Wizard of Oz versions of prototypes are making paper-sketches of apps and then assembling them into a click-through using a tool like Marvel. Or putting a sign-up sheet for a new service or organization, and seeing how many people sign up or take the ad. Or, manually acting out a new tech-based service.
Each of these prototypes will let the team count behavior: how many people sign up? How many people are able to use the app you’re envisioning? How many people click on an advertisement for a new service? How many people stop to look and use a new electronic sign you want to install?
It also is valuable for more formal organizations like the court, to do something temporary and non-permanent — to demonstrate what is possible without asking for a large commitment immediately. Rapid prototypes of new signs, worksheets, space layout, or tech can show value, get feedback, and build momentum without any budget requests or committee approvals.
After the Wizard of Oz prototype, the teams will continue on their journey to more refined, content-rich, and tech-enabled versions of their ideas. But these first four steps of the journey help them get quick input, vetting, and refinements to make sure they’re on a promising track and doing something that will resonate with their various stakeholders.
