There have been many thoughtful remarks on Westlaw Next (“WN”) already (for a cursory example: http://outofthejungle.blogspot.com/2010/01/westlaw-next.html and http://www.legaline.com/2010/01/first-look-at-westlawnext.html). I intend to limit my review to my personal interactions with it from trial experience. With a mixture of real-time reference questions, hypotheticals from my legal research class, and research for my own publication, I made ample use of my 6 week trial. I also saw two live demonstrations: the first at a sponsored breakfast in New York; the second with Yale’s representative. I found the product to be a vast improvement over the previous model, but still flawed. I hope some of the major problems are corrected long before students receive passwords, and I am very concerned about the likelihood that legal instructors will need to work with this system before it is ready.
The Interface
First, I am happy to report that the interface is fantastic. You have clean lines, pleasant colors, and an uncluttered work space. Veterans of Westlaw may forget how overwhelming the original search screen is for a new user; and I am certain that the clutter creates a stressful experience. The WN screen looks fantastic when it presents the user with search results, and it looks even better when you view a document.
I was thrilled with the presentation of a statute. The table of contents button appears; you can still scroll to previous or subsequent sections; you can identify the notes of the decisions, and you can key cite. Despite the new locations of these functions, I immediately recognized the majority of them. Public laws, despite their hyperlinks, don’t currently exist in WN, which brings us to a problem—not all Westlaw content is available on WN. In this limited instance, you are invited to suspend WN and log into regular Westlaw to see the document. (More on this problem below). I was also excited about the “versions” function. It was exactly what you want and expect versions to be: the P.L. number, the effective date, the duration, etc (in other words, not the search box that appears on Westlaw). The search box was one of my great complaints with Westlaw (to which any of my former students can attest). Here is an example of a previous version of 29 USCA 2611:
§ 2611. Definitions
29 USCA § 2611
Effective July 7, 2004 to January 27, 2008
Legislative Action Pub.L. 108-271, § 8(b), July 7, 2004, 118 Stat. 814
The Search
If the “versions” function made me happy, imagine my glee when I searched for Roe v. Wade and got… Roe v. Wade! On WN you get the real one, not a denial of a petition for rehearing or McTaggart v. Roe v. Wade. You get the one you want to see. This, the simplest of searches, is groundbreaking. WN appears to finally search by relevance, instead of reverse chronological order. It will revolutionize legal research to be able to immediately pull up the document you want, and the document everyone in the universe knows you want!
The accuracy of WN (regarding primary law) is a huge selling point. Despite the flaws, and whatever negative tone you gather from this review, the arrangement of the results, and the ability to manipulate them, are things I will use religiously. When I was asked to find a judge’s “major opinions,” I turned immediately to WN. I searched for the judge, looked at his writings, focused on cases he authored, and then displayed “most cited” first. I selected the first few pages of documents, downloaded them into a nifty folder, and emailed them in a zip file to a student. This was one tool I was able to use to get more useful results than Westlaw’s reverse chronological order.
Those of you who attended a breakfast may remember the following example. CEPA is the Conscientious Employee Protection Act. At the breakfast we learned most researchers think about the Act as a “Whistleblower Act,” which is a term not used in New Jersey. The demonstration got to the heart of a current legal research limitation: vocabulary. WN tries to allow you to search by concept. In other words, when you run a search for “whistleblower” in New Jersey, the system should be able to decipher what that means, and put you into an appropriate legal classification. However, Mark Estes sent me an interesting tidbit about the headnotes. When he ran a Westlaw search in New Jersey for: sy,di(whistleblower) he retrieved 13 cases; whereas searching sy,di(CEPA) retrieved 100. This was a classic Westlaw problem, and precisely the type of problem they hoped to remedy with WN. Unfortunately, I duplicated his search on WN and got the exact same results. My guess is that the algorithm does not work within fields.
I don’t want to be too hard on them, however. I will note that when you search “whistleblower” within New Jersey, the first statute to appear is the Conscientious Employee Protection Act (this is definitely not the case in Westlaw), so WN has absolutely corrected something.
Speaking of headnotes, they are… quite unfortunately… not good. For several days during my trial I couldn’t even access them. I could read the headnotes in a case, but when I clicked on a key number, an error appeared. When I finally got in, I found the notes very confusing. For example, I was unable to change the jurisdiction from the Supreme Court to the Second Circuit. The cases were presented in reverse chronological order (WHY is Westlaw so enthusiastic about this?) I could only narrow the search with a focus term. This was one of the few instances where I reverted back to my 1L days—was the problem with Westlaw or was I missing something? In short, it doesn’t really matter: headnotes are what Westlaw really offers; if I can’t use them adequately, where does that leave me?
Returning to the discussion about search and accuracy, Mark and I both had a problem with the secondary sources. He selected the database “All States and Federal” and searched for “Berkeley, CA eviction” but could not find the Rutter Landlord Tenant practice guide. “Berkeley eviction” in California and Federal likewise did not turn up the Rutter guide, although it did turn up many secondary sources with “Berkeley” in the title.
Similarly, I looked for a treatise that I knew was available on Westlaw: Bittker’s Federal Income Taxation of Individuals. A simple database query of “Federal Income Taxation of Individuals” in Westlaw would have yielded the result I wanted. In WN it did not. It retrieves primarily law review articles. This led me to problem referenced above—could Rutter and Bittker simply not be here? I limited my search to “secondary sources” and still only found articles. I was able to finally find the resource; however, I was forced to do precisely what WN claims you shouldn’t have to do: I figured out where it would be. I went into Topics, then Tax, then Treatises.
KeyCite and Filtering
KeyCiting posed serious problems for me. One big problem was that you can only view documents by document type. If you keycite a statute, you cannot simultaneously look at case and regulations that cite that statute. You must first select a type, such as cases, narrow the search as you normally would (focus terms, jurisdiction, etc). After you have completed that search you will need to return to the beginning, select regulations, and run a subsequent search. Try as I might, I was unable to figure a way around this.
This same problem appears, and more problematically, when using KeyCite to find secondary sources. You cannot, for example, search the ALR and law reviews simultaneously. Other problems: 1. WN lists all of the journal titles to the left and allows you to select them—if you know the title of the publication you want to search; 2. WN only displays results by “newest first” and “oldest first,” (i.e. not by times cited or times viewed or downloaded); 3. the author’s name will never appear unless you open the document.
I also had a problem with filters when I KeyCited a document. Each time you select a filter, the search runs immediately and the screen refreshes. So when I choose “cases” as the type of document I was interested in researching, the screen refreshed. When I limited cases to “Federal,” the screen refreshed. When I limited Federal Cases to “Reported,” the screen refreshed, and so on. As you may infer, the screen refreshing was not quite as rapid as I would have wanted. This resulted in an extraordinary waste of time while I waited for the screen to reload. I may be more impatient than most, but I fear this particular nuisance will only get worse as more users enter the system.
In sum, the platform is better and the experience is certainly better, but I would only favor this system for specific tasks that rely on primary law. While the design and interface are sleek, I think too little attention was paid to KeyCite, Headnotes, and Secondary Sources.
Reviewed by Ryan Harrington, Reference Librarian, Yale Law School, Lillian Goldman Law Library