By Sharon Nelson and John Simek
In the early days of electronic filing, the skeptics doubted that we would ever see the day when e-filing would be standardized throughout the federal courts. Clearly, however, the naysayers were wrong, for that day is almost upon us.
The Administrative Office of the U.S. Courts (AO) is now in the "mopping up" stage of bringing the last few bankruptcy and district courts online to join their sister courts in the Case Management/Electronic Case Filing (CM/ECF) system. Although previously scheduled to be completed at the end of 2005, the new target goal for completion of this massive project is now the end of 2006. As it turned out, there were a few stragglers—some courts that already had a homegrown system, which made data transfer complex, and others that simply were late "getting on the train."
Adoption of the CM/ECF program remains discretionary rather than mandatory, but the standardization of the process within the federal bankruptcy and district courts has become a phenomenon that no court could continue to resist over time. The only courts remaining to join are the bankruptcy courts of the Northern Marianas Islands and the district courts of Southern California, Southern Florida, New Mexico, the Virgin Islands and the Western District of Wisconsin.
As a final motivating factor, the AO will no longer support any of the old systems—any court that chooses to keep an old system is "on its own." As you might imagine, it's pretty cold in Siberia without support, so the move to the federal system has been accelerated by the thought of trying to survive without technical support from the federal government.
To help trace the progress of electronic case filing and the CM/ECF's current status, we interviewed Gary Bockweg, who has been shepherding this process over many years. He currently holds the title of Division Chief, Office of Court Administration, Technology Division within the AO.
Bockweg is a veritable fountain of statistics, some of which are profoundly impressive. If you haven't been keeping track, 93 of 94 federal bankruptcy courts have now gone "live," as have 89 of the 94 federal district courts. Also in the system now are the Court of International Trade and the Federal Court of Claims. More than 26 million cases are currently in the system, and more than 240,000 lawyers have now filed electronically using it.
Progress and Interesting Wrinkles
One interesting wrinkle has been the impact of the Bankruptcy Reform Act, which became effective on October 17, 2005. In May 2005, more than 136,000 new bankruptcy cases were filed. In May 2006, the number of new cases filed was down to about 53,000, undoubtedly as a reaction to the more stringent requirements of the new law—which seem to have emptied out the pipelines by prompting everyone considering bankruptcy to file before those new provisions went into effect.
In fact, in the first two weeks of October 2005, just before the law became effective, more than 600,000 new bankruptcy cases were filed. The system slowed under the weight of the onslaught—but it never shut down. This live and unrehearsed "stress test" was not anticipated, but the result was a tribute to a well-designed system.
In terms of design, curiously (at least to many observers), the AO is sticking with a user ID and password system. Although digital signatures are now far more reliable and standardized, Bockweg believes that it will likely take several years before they are implemented as part of the system. The code on which the system is run is now nearly 10 years old, though, of course, many adjustments have been made. Within the next few months, Bockweg says, the courts will complete their migration to the LINUX platform.
So how many problems have been reported by the system's users? Bockweg says there are not many complaints. One problem that crops up from time to time involves e-mail notifications that are delayed or just do not get delivered. Often, this is the result of a change of lawyer address that hasn't been recorded in the system. Sometimes, spam filters trap messages. Sometimes, firms or lawyers get so many incoming e-mails that messages simply "go missing" at the firm. From time to time, the AO has managed to get itself on one of the blacklists and has had to scramble to get itself delisted.
Many firms are now being proactive and using the system's capability to notify multiple addresses—this allows them to have a backup in the event that one of their providers goes down or one of their lawyers goes on vacation. Many elect to send not only to the individual lawyer but to a main firm e-mail address as well, so that all case developments are logged and monitored.
The AO is also being proactive, adopting an Ironport e-mail management system. If the AO receives bounce-back messages from an e-mail address, the process of tracking down the problem falls to court staff members, who generally call the lawyer to verify the correct e-mail address.
Data Harvesting: The Gold in the Records
The ka-ching, ka-ching, ka-ching of the federal courts' PACER system has continued to be music to the AO's ears, making the CM/ECF project totally self-supporting and with no additional fees charged for e-filing. Last year, revenues from PACER exceeded $40 million. At 8 cents a page, a lot of folks are choosing to access a lot of pages.
Does this mean the federal court data is being harvested by parties outside of the profession? You bet, and for all sorts of purposes. Banks, credit card companies, mortgage firms, stockbrokers, real estate agents—the list goes on and on. The data contained in federal courts' records—which are, of course public records—constitutes black gold for businesses.
In fact, the data harvesting is so intense that court personnel have had to request some of the major players to do their data harvesting at night to keep from bogging down the system during the business day.
Thus far, there has been little outcry from privacy advocates, though with the recent publicity surrounding data breaches and their consequences, it may be that the privacy groups will eventually take a look at the notion that the federal government is selling data (albeit public records) to anyone willing to cough up the 8-cents-per-page fee.
In case you are wondering: Yes, some identifying information is now redacted under law, including Social Security numbers, credit card numbers and the like. But much is not redacted. Worse still, the burden of properly redacting the documents falls on the lawyers filing the documents, with somewhat mixed results.
Lessons Learned During Disaster Recovery
One obvious question we put to Bockweg: How did Hurricane Katrina impact the system? With justifiable pride, Bockweg responded that most courts were back up the day after Katrina. There were a couple of exceptions in Louisiana and Alabama, but it won't surprise technologists to hear that the technology was ready to go before the people were. As disaster recovery experts point out all the time, disasters continually demonstrate that the first and foremost problem is communications.
The AO was ready to get some courts back up with replicated data, but it couldn't contact the necessary court officials to secure their authorization. Of course, with the power out and the ongoing battle against the waters, it was something of a moot point since there was no one at the courts to conduct business.
There was also a curious twist that provided a learning experience for the future: Sprint had a major communications hub in a low-lying area of New Orleans. When it went out, the court had one of two choices: It could run CM/ECF as a stand-alone system that only court staff had access to, or it could put the system up on the Internet—but then the court staff would have no access to the system because they had no Internet. Immediately after Katrina, some creative courts were able to quickly drop in DSL lines as a backup measure for Internet connectivity. The AO is now considering this new redundant measure to prevent such problems in the future.
Another lesson, from both Katrina and 9/11, is the critical nature of the replicated data. Currently, the western courts have replicated data held in Missouri, while the eastern courts have replicated data held in Reston, Virginia. As an added precaution, courts in the immediate vicinity of the two data storage areas will have their replications moved to the opposite part of the country.
While security remains a headache, Bockweg notes happily that—despite several PACER intrusions—there have been no intrusions of the "clean" servers, which hold the original court data behind a firewall. It's not that they aren't knocking to come in. There are tens of thousands of "door knockings" each month, but so far security measures have worked. However, it is an ever-present concern, so the AO works closely with the National Security Agency (which actually performs some "white hat" hacking to evaluate security measures) to ensure that court data remains secure. Every so often, viruses, worms, Trojans and the like slip through the system and wind up on the "dirty servers" outside the firewall, but they are quickly quashed.
But there's another security worry for the future. Bockweg advises that the AO will begin an experiment in releasing its code to states and other federal agencies, which have long pleaded to have the code to incorporate into their own systems. While the AO is anxious to be of help, Bockweg notes wryly that "obscurity is security." The more the code is out in the open and available to would-be hackers and cyber-terrorists, the greater the vulnerability. This will mean that the AO will have to pull the security net still tighter.
You don't need a crystal ball to see the next goal for the AO. Bockweg reports that the first of the 12 appellate courts will go live early next year, though there is no current indication of which court will go first. This presents a brand-new challenge for the AO's programming folks. As Bockweg points out, the appellate courts have a completely different set of requirements—this project has been in development for a couple of years already, simply laying the groundwork for a very different kind of court. He also notes with a smile that "the appellate courts have been less eager to embrace this technology—they are not all clamoring for e-filing!" All of them, however, are very interested in the case management aspects of the new system.
Beyond that, Bockweg is eager to make the programming more service-oriented and to update the software tools used to develop the system. Unsurprisingly, there is a list of enhancement requests from lawyers and the courts that goes on and on and on. To the extent possible, the AO will be trying to fulfill those requests. Although just in the planning stages, there will probably be a multiyear technological upgrade of the entire system.
And in addition to working with the states and sister agencies, the AO also hopes to work with the federal Sentencing Commission and the Bureau of Prisons to develop a data-sharing capacity so that data can be exported and imported, allowing different parts of the federal government to become more interconnected and to exploit the efficiencies of data sharing.
As of July 2006, there is a new director of the AO, Jim Duff. What he will bring as a vision is anyone's guess at this point. He comes aboard the train at an interesting moment, when the bankruptcy and district courts are almost done and the challenge of the appellate courts lies ahead. But all in all, it has been a heck of a ride to date. The federal government is often lambasted for getting things wrong, but this is one project that has come up a winner.