Last Friday, it was reported by the Washington Post and Ars Technica that Chairman of the Judiciary Committee, Senator Patrick Leahy, had sent a letter to Judge Bates, the head of the Administrative Office of the Courts, urging the AOC to put back online the recently-removed PACER documents from five courts. I had not seen the full letter posted anywhere yet, so present it here:
Free Law Project agrees with Senator Leahy that taking these documents offline represents “a dramatic step backwards” and that the Courts’ currently proposed work-around represents “a troubling increase in costs…” We hope the AOC will be open to restoring online access to these documents and stand ready to help make these documents freely available online for the public were that agreeable to the AOC.
Brian and I were guests this week on the live Internet show, This Week in Law on the Twit Network. In the show we cover a number of topics ranging from the history of Free Law Project and the need for innovation in the legal arena to the copyright and trademark issues in the latest Deadmou5 brouhaha.
Also available on YouTube. We hope you’ll enjoy watching.
At least since the destruction of the Ancient Library of Alexandria, the world has known the importance of having a backup. The RECAP archive of documents from PACER is a partial backup of documents taken offline by five federal courts. It is impossible to determine how complete a backup we have, because the problem with missing documents is that you cannot even determine that they are missing without a complete list of what used to be available. No such lists exist for the documents from these five courts.
But as coverage of this surprising and unprecedented action by PACER officials continues (see techdirt), the BBC has an article that takes an interesting approach by pointing out some of the landmark civil rights cases taken off PACER through this action.
The BBC mentions the case Ricci v. DeStefano which was decided at the Second Circuit while Sonia Sotomayor was a Circuit Judge. Sotomayor, now a Supreme Court Justice, had her role in deciding the case closely scrutinized during her Supreme Court confirmation hearings. Many who dug in to Sotomayor’s background during those hearings likely relied on the documents available via PACER related to this and other cases she decided. What will happen when the next individual from the Second, Seventh, Eleventh, or Federal Circuit, is nominated for higher office? Unless PACER officials make some accommodation, such as the one we asked for yesterday, researchers and the public will face the daunting task of requesting these materials individually, possibly in person, at excessive expense and inconvenience for everyone involved.
And as I told Common Dreams, the Administrative Office of the Courts (“AOC”) cannot argue that some of these cases have been closed for “13 years.” Some were closed this year! They also cannot argue that “many had not been accessed in several years” because as we demonstrated in a recent video, it can be very difficult and expensive to use PACER. Thus, PACER usage statistics should tell us nothing about the importance of these files. And often, as was the situation with Judge Sotomayor’s historical record, some cases only become particularly interesting years later and for reasons possibly unrelated to the case itself.
Put another way, libraries should hang on to their copy of Plato’s Crito, even if its been a few years since anyone checked it out.
This PACER situation presses home more than ever the importance of having a backup. Many critics of the PACER system were dissatisfied with many aspects of the AOC’s custodianship of America’s judicial record, but no critic ever thought that after so much work to move to online electronic records, we would return to an offline system of access to court records. If we now cannot even count on the courts to keep the electronic records online, even if behind a byzantine paywall, then we really must insist on a backup.
A recent announcement on the federal PACER website indicated that PACER documents from five courts prior to certain dates (pre-2010 for two courts, pre-2012 for one court, etc.) would no longer be available on PACER. The announcement was reported widely by news organizations, including the Washington Post and Ars Technica. The announcement has now been changed to explain, “As a result of these architectural changes, the locally developed legacy case management systems in the five courts listed below are now incompatible with PACER; therefore, the judiciary is no longer able to provide electronic access to the closed cases on those systems.” See a screenshot of the earlier announcement without this explanation:
This morning, Free Law Project signed on to five letters from the non-profit, Public.Resource.Org, headed by Carl Malamud, asking the Chief Judge of each of these five courts to provide us with access to these newly offline documents. The letter proposes that we be provided access in order to conduct privacy research, particularly with respect to the presence of social security numbers in court records, as Public.Resource.Org has done previously in several contexts. In addition we offer to host all the documents in a free public archive, at the Internet Archive, as we do now with RECAP documents. Free Law Project would also plan, ultimately, to incorporate the documents into our CourtListener platform, for even easier full-text searching and public accessibility.
The National Law Journal collected various tweets from lawyers and academics upset over the PACER announcement, and the general reaction to the surprise decision has been dismay, as many of these practitioners and professors rely on these documents for their work. We hope the Chief Judges we have reached out to today will agree to our request and allow us to get these documents back online for the public as soon as possible.
If you only watch one video about using the federal Public Access to Court Electronic Records (PACER) system, make it this video by Free Law Project’s Brian Carver: “Using PACER: What Could Possibly Go Wrong?”
The video provides a demonstration of what a regular member of the public might experience trying to find a copy of a recent newsworthy federal district court opinion on the court’s website and through the federal PACER system. This example was genuinely chosen because Brian himself had heard about a recent newsworthy case out of the District Court for the District of Maryland. In fact, we’re fairly sure that other examples might cast these sites in an even worse light.
Free Law Project believes that Congress should provide adequate funding to the federal courts so that the financial argument for PACER’s fees would be moot and everyone could agree that public access to court records should be free. But even in the absence of that, we conclude from this demonstration that the non-document related fees in PACER for search results and reports that are charged without an interstitial warning of their magnitude are particularly onerous and should be abolished. The courts could take this step immediately, while continuing to charge for actual court documents by the page, and a major obstacle to public access would be removed. Of course the video demo also illustrates a number of usability concerns that we think are best resolved by allowing third parties, such as Free Law Project, to build better search and access tools on top of these documents, but this is only feasible if at least these third parties are given free or very low cost bulk access to the underlying documents. We hope the Administrative Office of the Courts will be open to such options.
We recommend watching the video Full Screen and/or in High Definition so that the slides are readable.
In 2010, we wrote our first scraper. It was a nasty affair that could do nothing more than download PDFs from a webpage — any webpage. Since that point, we’ve come a long way and today we’re extremely excited to announce that the Juriscraper library now supports every state court of last resort. In most states this is the “Supreme Court”, though some states call their highest court the “Court of Appeals” or similar.
This means that no matter what jurisdiction interests you, no matter what area of the law you work in, you can follow the works of your state’s highest court in real time, getting emails or RSS feeds of the latest cases that interest you. And, of course, as new opinions are issued by these courts, we will keep adding them to our system, continuing to build the biggest repository of cases we can. As of today, Juriscraper has collected more than 400,000 opinions, and we expect that number to grow and grow.
On top of this, we also support dozens of intermediate State appellate courts and all of the federal courts of appeals. Check our list to see if a specific court is supported and let us know if there’s one we should add — user requests are often how we prioritize our work! There are already a few administrative bodies and other specialized courts that we plan to add soon.
Completing our appellate court coverage is a milestone we’ve been working towards for a very long time, and we could never have gotten here without lots of support, both financial and social. The following people have contributed code, presentations and time to help get Juriscraper where it is today (let me know if I forgot you!):
- Asadullah Baig
- Brian Carver
- Ben Cassedy
- Andrei Chelaru
- Alan deLevie
- Bo Jin (Krist)
- Deb Linton
- Andrew McConachie
- Matt Meiske
- Taliah B. Mirmalek
- Polya Pelova
- Raymond Yee
- David Zvenyach
And the following organizations have provided generous support for this project:
Thank you all very much for the great work you’ve done and the support you’ve provided. We now have a complete and free system for efficiently gathering opinions from state courts of last resort.
This is truly a momentous day and you might be wondering what’s next for Juriscraper. The first thing we’ve begun doing already is to collect oral argument audio. An important feature of Juriscraper’s design is modularity and after a few minor re-workings, the code is now well-positioned to archive more than just court opinions. We are currently working on expanding Juriscraper’s reach to audio files and we soon expect to build an archive of oral arguments.
Beyond this, Juriscraper’s design allows it to be easily turned to other document types or jurisdictions entirely. We have long thought that we’d also like to archive open access law journal articles and incorporate those into CourtListener. (Get in touch if you’d like to help!) Or you might have some entirely different archiving project in mind, the Juriscraper code can take much of the heavy lifting out of your way and allow you to focus on the unique aspects of the documents you’re hoping to archive. If you put the code to some such alternative use, please get in touch to let us know!
Last week legal publisher Fastcase included Free Law Project co-founders, Brian Carver and Michael Lissner on the company’s annual list of “Fastcase 50” award recipients. As their press release explains, “The Fastcase 50 award recognizes 50 of the smartest, most courageous innovators, techies, visionaries, and leaders in the law.”
Michael and I are humbled by and grateful for this recognition. We’re especially thrilled to see individuals we have worked with included on this year’s list, such as:
- Jake Heller, CEO of CaseText, whose team there has frequently been a helpful sounding board when Mike and I are thinking through the interesting questions that arise when trying to put useful legal research tools on the web.
- Colin Starger, of the University of Baltimore School of Law, with whom we’ve had great conversations about citations, metadata, and bulk downloads, not topics of conversation that everyone has as much experience with as Colin!
- David Zvenyach, General Counsel to the Council of the District of Columbia, who saw that we hadn’t gotten around to adding scrapers for the D.C. Court of Appeals or the District Court for the District of Columbia to juriscraper and wrote them for us.
These individuals were obviously recognized for their broader work, but we were excited to see those with whom we’ve already made a connection getting much-deserved recognition. Especially since we all join a distinguished group of past winners, many of whom we’ve also received advice or assistance from, including: 2013 awardees Bob Berring, Josh Blackman, Sara Frug, Tim Hwang, Dan Katz, Dan Lewis, Eric Mill, Josh Tauberer, and 2012 awardees Jerry Goldman, Waldo Jaquith, Peter Martin, Elmer Masters, and 2011 awardees Monica Bay, Tom Bruce, Sarah Glassmeyer, John Joergensen, Carl Malamud, Rob Richards, and Tim Stanley. At various times we’ve been inspired by, gotten advice from, or gotten help from all these folks, among others, so we think they deserve some credit for any headway we’ve made on the problems we’re tackling, while we’re happy to take all the blame for any missteps or not making more progress faster.
The only problem with these lists is that Fastcase CEO, Ed Walters, should be on there too. Thanks, Ed!
We’re happy to share today that we’ve completed a revamp of the CourtListener website to make it more polished, easier to use and easier to learn. There are a handful of changes we’re really happy about and that we’ve wanted to do for a long time.
First, of course, is our new homepage. The new homepage is designed to showcase our latest material, to make new opinions easy to find, and to better introduce CourtListener to new users. The most striking change in the homepage is that at its center it now has a huge search box where you can place queries, and if you’re an advanced user, you can press the “Advanced” button, and it will show you all of the search facets that we support, from Case Name to Citation Count.
The homepage also ushers in a change that’s been in the works for some time – we’re finally ordering our results by “Relevance” instead of “Newest First”. This change was made possible by the improvements we’ve made to our relevance engine over the past year, and is one we’re really excited about. We expect that as you use the new homepage, you’ll find the relevance engine to surprise and delight you.
The next major change that has come with the revamp is a new tour of the website that you can take to learn more about its features. In the footer there’s a link to kick off the tour and we’ll be experimenting with different ways to promote it to new users over the next couple of weeks. We are also working on a video that we expect will help people learn how to use the site.
The final change that we’re rolling out (aside from our new FAQ and About pages) are social links at the bottom of every page. We know people use different tools to stay up to date, so in addition to this blog, we also want to make sure that you can follow us on LinkedIn, Facebook, Twitter and Google+.
Today Free Law Project announced that it is partnering with Princeton University’s Center for Information Technology Policy to manage the operation and development of the RECAP platform. Most readers here will know that the RECAP platform utilizes free browser extensions to improve the experience of using PACER, the electronic public access system for U.S. federal courts, and crowdsources the creation of a free and open archive of public court records.
I have been frustrated with PACER for a long time: as a member of the public, as a law student, as a litigator, as an academic, and as one trying to build systems for public access to court documents. I’ve been frustrated by the price per page, by the price for searches with no results, by the shocking price for inadvertent searches with thousands of results, by the occasional price for judicial opinions that are supposed to be free, by the price in light of the fact that Congress made clear that the Judicial Conference “may, only to the extent necessary, prescribe reasonable fees… for access to information available through automatic data processing equipment” when it has been demonstrated time and again that PACER revenues grossly exceed the “extent necessary,” and by the 2011 increase in those prices. I’ve been frustrated by the search interface or lack thereof. I’ve wanted, for a very long time, to be able to conduct free full-text searches across the entirety of the PACER database. The research questions I could ask and answer with such a system. Sigh… But most of all, I’ve been frustrated with PACER for a long time because it creates a wall between the public and the law.
So when CITP announced the launch of RECAP in August 2009 I was overjoyed and I became an early adopter. I reported bugs and suggested enhancements on RECAP’s now-defunct “getsatisfaction” community. I told anyone that would listen what a brilliant idea it was to involve everyone in the effort to create a free public archive of these public court records. I felt empowered by the idea that my own browser could become a tool in the effort for greater public access to the law. I got something approximating a crush (techno-crush?) on Steve Schultze, Harlan Yu, Timothy Lee, and Ed Felten, for the brilliance behind this creation. (Please go install the browser extensions if you haven’t already! For Firefox; For Chrome.)
Steve, Harlan, Tim, and others that have worked on RECAP have now all moved on from Princeton to fabulous careers. When Steve and Harlan began discussing with us the idea of Free Law Project taking up the slack created by their busy new adventures, Michael and I were immediately enthusiastic. RECAP is a perfect complement to Free Law Project’s existing efforts. The first version of CourtListener covered only the federal circuit courts and the Supreme Court of the United States. We’ve been working recently to roll out coverage of the courts of last resort in all the states (not quite there yet…) but have largely held off on tackling “the PACER problem” presented by the federal district courts. In part, we knew we could wait on this because RECAP was already addressing it about as well as anyone could. We have long assumed that we would at some point look to merge the RECAP documents with our existing documents, and knew because we focused on different courts, these would be largely complementary sets of documents. Well, the time to tackle that merger has arrived, and in the process we hope to provide a long-term home for RECAP maintenance and improvement.
However, to do everything we envision with RECAP and its existing document archive is going to require additional funding. (You can always donate!) Therefore, we expect in the coming weeks to approach anyone who is known to write large checks with this idea: Support us in creating the largest freely-downloadable collection of court documents that has ever been assembled on the internet, and in the process, help ensure the future of public access to the law’s killer app: RECAP.
We’re looking forward to it.
P.S. Please follow @RECAPtheLaw. We intend to start tweeting there.
See also, Steve Schultze’s post at Freedom to Tinker.
One of the goals of the CourtListener platform is to enable others to analyze judicial opinions. To that end we provide all of our data as bulk downloads and try to archive any opinion that a court publishes.
On some occasions, this results in a slightly confusing search result like the following:
We realize having an opinion in the system two times can be a bit confusing, but the reason this happens is because courts will sometimes make corrections to a slip opinion after its initial release. Sometimes the new version of the slip opinion will make note of the change, other times the court makes this change silently, perhaps hoping that the public either doesn’t notice or doesn’t care.
Usually these are minor corrections, but occasionally not. For example, in the case above, Justice Scalia made a mistake in his dissent, completely flipping the position of the EPA in the case he references. Such significant alterations are a rare occurrence and it has been widely highlighted in the press. Many systems will silently remove opinions that have such errors, but we have found that even studying the corrections made by courts is something that interests some academic researchers.
For example, Peter W. Martin, the Jane M.G. Foster Professor of Law, Emeritus, former dean of Cornell Law School, and co-founder of Cornell’s Legal Information Institute (LII), has written on this topic more than once and noted CourtListener’s usefulness for detecting these sorts of judicial corrections. Based on his analysis of this practice he also provides suggestions to the courts on how to handle revisions in a world of digital publication.
As our platform progresses, a goal is to begin automatically identifying items such as the above and presenting the differences intelligently. While this sort of detail is often inconsequential, both this recent instance and some that Professor Martin describes show that at times they can be notable and even newsworthy, changing the meaning of a portion of the court’s opinion completely.