"Nota Bene" means "note this well" or "take particular notice." We at the O'Quinn Law Library will be posting tips on legal research techniques and resources, developments in the world of legal information, happenings at the Law Library, and legal news reports that deserve your particular attention. We look forward to sharing our thoughts and findings and to hearing from you.
N.B: Make a note to visit "Nota Bene" regularly.

-Spencer L. Simons, Director, O'Quinn Law Library and Associate Professor of Law


Friday, May 31, 2013

The Picture of Legal Research – Ravel



There is a new player in the electronic legal research world; its name is Ravel and it wants to do for legal research what television did for radio: make it visual.

Ravel is the brain-child of a group of recent Stanford Law graduates who wanted to come up with a new way of doing legal research that would be “radically easier, faster, and more intuitive.”  While the platform is more visually striking than the text heavy appearances of the more traditional platforms Ravel may not yet be a complete game-changer.  Let’s take a look.  

The site has the familiar Google minimalist aesthetic look and right away lets you know that the cool kids who use Google Chrome are preferred, although it seemed to work fine with my stodgy old Firefox browser worked fine. Coverage is limited to only U.S. Supreme Court and all Federal Circuit court cases included. There are no Federal District court or any state court cases.  The developers state that “we are aggressively expanding the coverage of our case opinion database,” but they don’t say what they are expanding to. In an interview with founder Daniel Lewis he states that statutes will be added, but says nothing about state court opinions or administrative law materials.  The product is still in beta mode so I’m willing to give them the benefit of the doubt. At present this is exclusively a federal case finding tool. Its advance search merely allows the researcher to limit the keyword search to particular federal jurisdictions. 

I did a few very simple and limited searches using Ravel that centered around the subject of my previous post, PGA Tour, Inc. v. Martin. I searched in the U.S. Supreme Court database using the term “golf”.  The results screen is divided into a narrow right-hand pane with the list of cases with keywords highlighted. The center pane consists of two graphs. A smaller graph at the bottom charts the mentions of the key term over time with hills and valleys indicating how much or how little the search term was used. The larger graph that occupies most of the screen consists of circles. Large circles represent “more important cases based on their citation network.” Each circle has a color: Red indicates a case has not been cited for at least 10 years, Blue lines link cases that a case is cited by, Green lines show cases that a case cites to. At the top of the main pane there are limiters titled; Ravel, Relevance, Court, and Cluster. Ravel is the default setting. Relevance raises more relevant cases to the top of the graph. Court indicates which court decided the case, and Cluster seems to do just that—cluster all the cases together, at least that is what it appeared to do. For what it’s worth Martin was the 13th case in the results list and by looking at its circle size it is not the most important golf case decided by the Supreme Court. 

I changed the search to encompass all the available jurisdictions and the search returned a total of 2085 cases with the top 75 cases being displayed. I changed the search again to “golf” and “disability” and the Martin case grew to a large circle (relevant cases have larger circles).  By narrowing my search with the additional term the significance of the Martin case grew. 

By clicking on the case in the right-hand pane you get a completely new screen with the text of the case in the center, page numbers on the far left, and another graph in the upper right corner. The opinion itself provides all the expected citations and can be printed in PDF format.  If you create an account you are allowed to make annotations on a case which will be saved, which is always a handy thing. The graph in the upper right corner is titled “Opinions Citing PGA Tour, Inc. v. Martin”. This graph has years along the bottom and numbers on the side and a slider that lets you move from year to year. As the slider passes over a year you get the number of cases that cited to your main case for that year along with links to those cases below the graph. 

What the Ravel group has come up with is a great beginning of a case research tool. Graphically the site is interesting; rather like Westlaw’s graphic view on steroids (and much more helpful), but it doesn’t possess the depth of Shepard’s.  The graphics are a welcome addition because they do provide the ability to see patterns in cases across courts and make easy connections between cases, but my fear is that students and newer attorneys will rely on the size of the circle rather than the actual language in the case. 

In the interview Lewis states that “in a number of other research spaces where people navigate large amounts of data, including scientific research, finance, and engineering” graphical interfaces are widespread. Perhaps I’m old fashioned but I don’t really consider law in general to be “data” given the nuance and uniqueness of each separate legal opinion.  All of the other fields mentioned strike me as relying on “raw data” that require organization into a discernible pattern. To me law is still, generally, a textual medium, and while graphics certainly has its place, and I for one would welcome it to electronic research, it is not more important than reading each actual relevant case to see what was said. The developers of Ravel will of course say that the prudent attorney will of course read every case, and yes, a prudent attorney will, but the graphics may act as a crutch for the less diligent attorney and that is my fear. In the interview Lewis states that the Ravel group had all become “intimately familiar with the existing legal research tools, and couldn’t help but notice that the interfaces and underlying search technology had failed to keep up not just with the times in general, but with ever increasing amounts of legal information in particular .”  I am sure the armies of programmers at Thomson-Reuters (Westlaw Next), LexisNexis (Lexis Advance), and Bloomberg Law will be quite happy to know that a group of summer associates “intimately familiar” with legal research are ready to tell them how to do their jobs. Those crazy Millennials (shakes head slowly).  

In a world where the big players, Westlaw Next, Lexis Advance, and Bloomberg Law, all trumpet their powerful search algorithms, I don’t think that graphics will be enough to make Ravel a real player in the long term. I hope that this ambitious law school project spurs the big boys to add the option of using more graphics in their products. While I am too practical to assume that there exists the “Platonic Ideal” of an electronic legal research platform, Ravel is a nice step forward.

No comments:

Post a Comment