NIJ Audio Transcript: Forensic Information Data Exchange: Improving Business Processes Through the Efficient Exchange of Information
- William A. Ford, Director, Information and Sensor Technologies Division, Office of Science and Technology, National Institute of Justice, U.S. Department of Justice, Washington, D.C.
- Aaron Gorrell, President and CEO, Waterhole Software Inc.
- Michael O'Berry, Operations Manager, National Missing and Unidentified Persons System, National Forensic Science Technology Center
- Kevin Kosiorek, Criminalist, Serology Section, Crime Laboratory Unit, Boston Police Department
- Jim Markey, Sergeant, Phoenix Police Department
- Aaron Gorrell — Part 2
William A. Ford: This project is getting better and better every year. This came out of an idea of a lack of content and a lack of information between the forensic community, law enforcement, courts, corrections and that whole information sharing.
My name is Bill Ford. I'm with the National Institute of Justice. We have two great officers or two great labs that are here who, as I call them, willing victims. The way this product has moved to where it is is because we've had buy-in from the community to make this better.
So, with that, this started off with an idea … let me backtrack a little second. No, that was perfect. That was perfect because the idea of the National Information Exchange Model, which is a joint DHS, Homeland Security and Department of Justice initiative, to come up with that standardized language to share information, that's great, big data model out there in the sky.
We work with actually another effort, our next slide, which is Global Justice Information Sharing System, which is a FACA program that reports to the Attorney General. The major associations and criminal justice have the voice to inform the Attorney General, going, “Hey, there's this gap. Please address it.”
I'm very fortunate with inside the Office of Science Technology that I can also assemble a group of individuals and go, “What are your challenges? What do we need to go after?” So we assembled, as I call them, 15, 20 guys and gals with guns or whose paycheck says “the city of” or “the state of” and asked that question, and with that, then we get to turn around and do targeted-type implementations such as this.
This is what FIDEX came from, the Forensic Information Data Exchange model. Aaron is a wonderful person. We've met many years ago in the sense of global, and when Mike and his team teamed up with Aaron, we went, “We got this done.” If you want to speak XML, this guy speaks XML.
So, with that, I'm going to turn it over to Aaron and let him give you an idea of where we are with FIDEX.
Aaron Gorrell: Thanks, Bill. I promise not to speak in XML today. I saw the sugar out there and I'd be fairly incomprehensible on a good day.
So just a little bit of background about FIDEX, or Forensic Data Exchange: What NIJ was looking at … as you know, NIJ has spent somewhere around $330 million in their ongoing effort to try to reduce case backlog in crime laboratories. What they wanted to do is take a look at new and innovative ways to reduce backlog instead of relying on the concept of providing money to labs to hire additional analysts or to outsource certain kinds of examinations.
We wanted to look at fundamentally changing the equation to figure out is there a way to reduce backlog without actually doing the examination, which seems like heresy until you start looking at it and doing some of the analysis that we've done with the IJIS Institute, and you start talking to crime laboratories and find out, and it varies from crime lab to crime lab, anywhere from 50 percent to 80 percent of the cases in their backlog have already been disposed.
And so we started taking a look at is there a possibility for getting information sharing going on between the courts and the district attorney and the crime labs in order to identify those cases and in order to be able to identify those cases to potentially cancel the examinations if there's no longer any need — if they've pled out or if the case has been adjudicated or whatever has happened.
So that's where it started off is as a standards development effort to develop standards for court software solution providers to be able to feed information into the crime lab or for law enforcement software providers or in-house providers to be able to share information with the forensic crime lab.
And it ended up culminating in a couple of pilot projects, one with Boston Police Department and the other with Phoenix Police Department. Now, as Bill said, in terms of how we kind of envision the format of this presentation, we do want to kind of make it informal, as informal as you can when you have 100 of your closest friends in a gigantic field-sized room here.
So at the conclusion of each speaker, please feel free to speak up and ask any questions, and then our plan is to allow 15 minutes or so at the end of the presentations to also discuss it.
So our first panelist, Mike O'Berry, is the National Forensic Science and Technology Center program manager, and essentially he's the one who kept us in alignment and frankly made sure I got paid. So Mike is going to kind of give you a little bit of background information on the project, how it initiated and some of the steps that the NFSTC did in order to get the project off the ground and continue the project.
Michael O'Berry: The National Forensic Science Technology Center worked with the National Institute of Justice through a cooperative agreement on FIDEX. So I have to say that right upfront or Bill will kick me in the shin.
The program funding came through, obviously, the Department of Justice, Office of Justice Programs, and the NFSTC was originally established through the American Society of Crime Lab Directors in 1995. So it's been around for quite a while, and we provide a lot of different services to the criminal justice community.
The NFSTC is located down in Largo, Florida. That's Tampa Bay, basically, and you can see that we have a number of services that we provide through the NFSTC: training, technical assistance and evaluation, obviously criminal justice, community support. The labs are one of our main focus, but we've obviously branched out into crime scene investigation training, firearms, impression evidence. There are a number of training projects that we support and deliver.
The vision is for the forensic science community and its users to have complete confidence in the quality of science provided by the justice community, so … I'm sorry, to the justice community. Our mission is that we are dedicated in assisting forensic scientists in achieving the highest level of quality services for the justice community. So you can see that we have a broad vision and a focus in terms of services for the justice community.
The advisory group for FIDEX and drilling down to the specific project was a national group established in 2007 that helped scope and helped us understand the needs, issues, and we've put together a pretty diverse group. The outcomes from the national working group were forensic case-related justice information would be captured and shared in an interoperable, uniform, scalable format, which is mumbo jumbo for case information being shared between crime labs, law enforcement and the courts.
The pre-login examination requests were the main focus. What we realized through surveys and focus group members was that there was a big need for the time spent in communication related to case submission, to reduce that and to provide multilevel communication, evidence-related information in the same format and make it accessible to all the different users.
The FIDEX partners for the working group included the Department of Justice, Office of Justice Programs, NIJ, NFSTC executive members and our staff, IJIS Institute, the Arizona Criminal Justice Commission, ACJC.
As you can see, Justice tracks LIMS. So, for those of you — how many people in here are familiar with Laboratory Information Management Systems? OK, great. So we tried to include vendors and people who are involved with the technology that would be linked to FIDEX — Waterhole Software, obviously, Evidence Tracker — and the pilot sites initially included Boston Police Department and Phoenix Police Department crime lab.
So the partners played a big role in what we're doing and how we drilled down to the Arizona subgroup. The Arizona Criminal Justice Commission was gracious enough to support a subgroup so that we could study work flow in Arizona between local, county, and state crime labs and law enforcement agencies, and that became a big part of what we did to make the FIDEX case exchanges practical for users.
So that's really all I have to say about the NFSTC and kind of in a very short story how we arrived here and supporting FIDEX with a cooperative agreement with the NIJ.
Kevin Kosiorek: Good afternoon. My name's Kevin Kosiorek, and I'm a criminalist with the Boston Police crime lab. And we are a relatively small forensic group; at least, I consider us a small agency, small forensic group when I look at labs like Michigan State Police and L.A. County.
But, about a year ago, that's when I met Aaron, and he was in our police department for a meeting with some of our command staff regarding the FIDEX system. And, you know, one of my coworkers and I, we were basically ordered to attend this meeting. I don't know how many of you out there are actively working cases in a crime lab, but wherever you're working, the caseload is typically pretty great, and, you know, we weren't too happy about having to go to this meeting. But we reluctantly went and we heard what Aaron had to say, and I immediately was very excited about the possibility of FIDEX and incorporating it into our lab.
We have a number of items of evidence that are handled by our forensic group each year. This is just, you know, some of the numbers. Just in the crime lab alone, over 3,600 items of evidence have already come in through our evidence receiving window in about five months. So we handle a lot of items. You know, there are labs out there that are much bigger, but just with the number of people that we have in the lab, 20 people handling about 3,000 items of evidence, it is a lot of evidence coming in each day.
Another thing about working in the Boston lab is that we do work with one agency. So we primarily work for the Boston Police Department as well as some of the colleges and universities, and also we work with one district attorney's office, which is the Suffolk County district attorney's office, which, basically, Boston is Suffolk County, if you're from Massachusetts.
But, again, going back to this meeting, we reluctantly went to this meeting and we heard what Aaron had to say about FIDEX, and we, you know, we, along with virtually every forensic lab in the country, there is that breakdown in communication between the forensic scientists, the DA's office and the detectives in charge of the case, and there usually isn't that free flow of information.
But one thing that we would probably all agree on is that the crime labs or the forensic units in general have a lot of information that both sides want, and really how do you get that information to everybody and make everyone happy?
Some of the things that we were concerned with when we met with Aaron for the first time is just the redundancy of having to enter in more information into another system. Our lab, currently you have to enter each case into an Excel spreadsheet. We also have an electronic evidence tracking system, which is not a LIM system, and then we also have a number of other different spreadsheets that are kind of optional — “optional” in quotes — for you to enter in your information. So we were really concerned with is this going to take us a lot of time to enter in this information again and again and again and is there any way that we can kind of minimize the amount of places where we have to put it in.
Again, the first meeting we had was just to hear what his proposition was and for the business model, but we were very excited at the end when he said that he could … pretty much, once you enter in the incident number, you could have your preferences set up, and it would be the same for every case.
Another thing about our meeting, our second meeting when Aaron came back to discuss just how FIDEX was going to work: At this point, we had a member of the command staff, a deputy superintendent and a veteran homicide investigator attend this meeting. This meeting was actually very interesting because you could see that there was immediately going to be a tug of war for information between the DA's office, the crime lab and the detectives in charge of the case. Pretty much, the detectives were concerned with, you know, if the DAs or ADAs, assistant district attorneys, have access to all of our evidence information, who is going to be the one to, you know, request evidence examinations, and if the assistant district attorneys are allowed to do that, will it cut them out of the loop, because although we do have a very good, strong working relationship together, you know, when a detective works nights, the district attorney works days, sometimes the communication, there's a breakdown and it falls apart, so we want to try and avoid that.
So what we agreed was that we would have FIDEX set up just internally where the detectives could be involved in viewing what FIDEX has to offer, and the crime lab would be basically in charge of who would be allowed to access FIDEX with each case.
When we're entering in evidence after each case comes in, if it's a big case, let's say a homicide where we have hundreds of items of evidence, the forensic group is broken down into the crime lab, the latent print section, the firearms section and the crime scene response unit. All of those units are entering in their evidence simultaneously, which can be very difficult. So you really don't know what evidence you have in the case until several weeks after, and you also don't know where it's going to be located.
You can go in and you can look through our evidence tracking system, but, again, there's just so much per volume of evidence per case that it can make it very difficult.
So what we wanted to avoid was basically, you know, if the latent fingerprint unit is generating items of evidence, how are we going to know what is going to be submitted to us later on, and if the crime scene response unit is doing the same and the firearms section, you know, we need a heads-up as to what is being generated and where it's going to be stored and also if it's going to be arriving in the lab at some point.
OK. So, when we're working with the detectives, too, and there's that general fear over who is going to have access, there are a couple of things that we were concerned with, working in the lab, that the detectives weren't necessarily thinking about, so we had concerns of our own, and some of those were the e-mails. If you're working in a lab and you communicate with the detectives or the district attorney's office through e-mails, you know, sometimes there's different subject matter in the e-mail that you would rather not be a part of your case file.
You know, for instance, we work with a very small agency. We generate some pretty good relationships, working relationships and personal relationships with the detectives. If you're out having a few cocktails one night, you don't really want that in the e-mail of “How is my case going? Oh, you know, by the way, it was a great time at the bar last night.” So we try to avoid those situations.
The detectives sometimes don't realize that we're documenting all that, even though we specifically say we keep a contact log of any time you request an examination that goes in the case file, but, again, that's something that we want to avoid. So we want to avoid having the e-mails being, you know, part of the case file. That's one excellent thing about FIDEX is that when you're using FIDEX — and you're going to have a demonstration later in this — Aaron is going is show you the simplicity of the e-mails is perfect. Basically, it shows you the case number, the incident number for your agency as well as a brief description of what the item of evidence is.
Now, if you're having an evidence request, the e-mail is also going out with just the general evidence request. So you're eliminating all that personal information. That way, if you do want to exchange e-mails with the detective and decide where you're going out for drinks that weekend, you don't have to have it be in part of the case file, which is something that I personally really like. It's just having the simplicity of the e-mails.
Now, with the district attorney's office, there was another concern that if the district attorney's office was requesting different examinations for our evidence, that what happens later on in the case if they request something, because we have a lot of new district attorneys or assistant district attorneys that will ask for specific items to be tested, not really knowing what type of testing we offer, not really having a forensic background whatsoever. So what if they're asking for something inappropriate and all of a sudden this is discoverable and, you know, we didn't perform the test that they asked? I don't know how it works with everybody else in the country, but I'm assuming it's pretty much the same. You know, when you go through discovery or discovery requests and the defense gets a copy of your case file, they do look at all your correspondence with the detectives and the district attorney's office, and they want to see are you following up on the different examinations that were requested. So that was one of the concerns.
I personally don't see that as being a concern because you can always explain that a case is … basically, it's a living thing, and you're discussing different types of examinations that are going to take place, and then you may have new information develop. And then, all of a sudden, those previous requests are not really relevant. But that was something that the detectives were concerned with.
So right now we use FIDEX pretty much internally for the e-mail notifications on evidence, but I would like to speak a little bit about that.
We get a lot of evidence in, and a lot of units within the police department have access to our evidence tracking system. Evidence may be split off an existing item where it's been … basically, it's called a “new item of evidence.”
When I send something to the latent fingerprint section, what happens is, you know, they may swab the mouth of a bottle area, and I want to know where are they in the process of examining that item of evidence. Well, I will get an e-mail notification — it's three lines long — that just says that a swab has been generated and now that is a part of the evidence.
Now, I personally really like this because it really has eliminated a lot of the confusion for me where I have to go and diligently check, you know, and try and find out what items of evidence had been created and when they were created, because, you know, you have a big caseload, you know, you have hundreds of cases that you're working on, from, you know, the current cases all the way to cold cases, and it's hard to keep track. You may have a thousand Post-It Notes on your desk. This is one way of eliminating that because you're just getting an e-mail notification, and it'll kind of bring it to the front of your mind, so that you're now going to work on that case.
We've had a lot of success stories, too, with using the FIDEX system. We've had a number of criminalists that have asked to be part of using FIDEX. Initially, there was a lot of hesitation because, you know, it's entering data into another system, but we've had one criminalist in particular where she was waiting for oral swabs for a breaking-and-entering case.
She was working a large number of these cases, and she was waiting weeks for this oral swab to come in. Little did she know that the swab had been submitted. It was just, you know, not a procedural error but kind of an oversight that the receipt wasn't placed in a place where she could see it on her desk, and a deadline was pretty much missed for court. So that has completely eliminated, you know, missing the deadlines by using FIDEX and using, you know, what it has to offer for the notifications.
Now, what we really don't use FIDEX for currently in the lab is for our statistics. You know, it seems to be a recurring theme throughout the conferences that statistics are, you know, more and more important each day. Your commissioners and your police chiefs want to know exactly what's going on in the lab. They want more accountability. You know, they want all of this information regarding what are you doing with your time and how are you processing evidence and reducing backlogs.
But you can use FIDEX for this purpose. We haven't used it yet for this purpose because it's really taken me a while to convince some of my coworkers to get into the FIDEX mindset, and now since there have been some issues with potentially missing deadlines, I've convinced them.
And actually, since going and preparing for this conference and preparing a little lecture, I went over to some of the other units, such as the latent fingerprint section and the firearms section, and I was just kind of telling them what I was gathering data for specifically with the amount of items we collect, and they were really upset that I haven't allowed them to use FIDEX yet, which I think they're forgetting that we had this conversation about a year ago and nobody really wanted to talk about it.
Kosiorek: But, really, when I go back now, now I have an audience that is going to listen to me, and they're going to be very interested in using FIDEX. So there isn't enough that I can say about FIDEX and how it could work. Having gone through a couple of the seminars here in the last couple of days, you know, talking about the backlogs in Michigan and then in L.A. County, this would be a perfect tool for those two agencies, you know, for looking at workload as well as just figuring out where cases are in your system.
And it's so simple to use. Really, once you have your preferences set, all you have to do is enter in your case number. It works by itself. It basically aggregates all of that information. So, you know, I really just can't say enough, but Aaron is going to show you a demonstration later.
Anybody have any questions first? Yes.
Audience Member 1: What is … how does it handle emergency or ASAP cases?
Kosiorek: Well, we haven't really used it for the request for examination as of right now, but, I mean, you'll have a request for an examination. I guess Aaron might be able to touch on that.
Aaron Gorrell: Sure, Dr. [inaudible]. You know, it really depends on what your business process is for those kinds of emergency examinations. The tool, when we developed the tool, and I'm getting ahead of myself a little bit, but one of the things that we discovered is every crime lab has a little bit different business process in terms of requesting examinations and so on and so forth, so one of the first requirements we had was whatever tool we built can't force any kind of business process down an agency's throat.
But what we did find is that there's a high degree of consistency with the kind of information that they're capturing. So, you know, I don't know your specific process for requiring the emergency requests. I would say because everything is real time with the software, it should be able to be configured to meet business requirements.
Kosiorek: I just want to say one more thing, because I had a note and I forgot to say it. In September of last year, our latent fingerprint section was accredited by ASCLD and the DA's office. Well, the district attorney, Dan Conley, he came in and gave a little speech, you know, for the accreditation ceremony. And he basically put out a number. He said I just want to congratulate the latent fingerprint unit in their accreditation because we've had successful prosecution in 85 percent of our gun cases.
And that's one thing that I would like to see happen in the future, is to have the court system, because that's really what it's ideally designed for, is to get everybody involved — the courts, the detectives, as well as the crime lab. You know, we don't know the disposition of a lot of our cases and that's something that we really need to have the courts help us out with by allowing us to have access to their system.
I know that in our court system they're, I believe, going through a process in the next few months where they're going to be implementing new technology there, or a new system, so I think we have to wait for that to take place before we can go back and ask them to allow us to have access.
But, you know, we're sitting there in disbelief that, wow, 85 percent of our cases are ending up in a successful prosecution. That would be really nice to know as it's happening and being able to look at these items and say, “OK, fine, we don't have to examine this because it's already gone forward and the person took a plea,” or, you know, “Yes, it is moving forward, and they need our results in order to, you know, maybe reach a plea deal or go forward with trial.”
So, any other questions?
Audience Member 2: I wanted to know … two questions, actually. Is this available for all forensic units in full-service forensic labs, not just DNA, I would assume?
Gorrell: Yes. When I show you a demonstration it's going to show how it can be used across any kind of examination types with any types of evidence, quite frankly.
Audience Member 2: And then to take it one step further, and perhaps you'll touch on it. Is it also available for the DA's office or I guess information to be inputted on their end as far as offender and arrestee collections from various sites to ensure that the sample has been provided and then submitted to the forensic lab?
Gorrell: Absolutely. The application is a Web-based application, so as long as you can get to the server, the physical server through your network to access the Web page, and you've been given authority to be able to view the case, you being the specific ADA or detective or criminalist, you can add it, you can update it, we can set up rules about what fields you can update based on your role.
Audience Member 3: I was wondering, do you guys assign cases to your specific analysts so the e-mails get sent to you directly? Or do they get sent to everybody?
Kosiorek: That's a good question. Actually, you set up your rules. You'll have an administrator that can assign cases and once they list you as the case analyst, you'll get a notification — again, three lines long — saying that this case has been assigned to you. And then, as the analyst, you'll have the authority to then include detectives that are involved in the case, so you may want to include specific investigators, maybe an investigator and their supervisor depending on how your agency works. And then you can also include people from the DA's office, so you can actually input various e-mail addresses and include whoever you'd like. And that's the beauty of it, is that, whoever is involved in the case, but you as the crime lab analyst will have the authority to do that.
Audience Member 3: OK. So do you find that you're inundated with e-mails? It sounds like a lot of e-mailing. I mean, do you have to sort through them? I mean, how does that…
Kosiorek: It depends on the case, you know. If I have a case where there are a couple hundred items, I will notice that I will get flooded with e-mails when latent fingerprints starts examining, like, a case of beer bottles and they start swabbing all the bottles and splitting those items. But it's still very good to know because I know that all of those items are going to eventually come across my desk.
But I'm not really flooded right now because I'm using it in kind of a small population. When I get back, and now that I have more interest, it's going to be much bigger. But I don't foresee it as being a burden, it's really more of a heads-up, you know. I may have a list of 10 e-mails and I'll flip through them just to see what is being processed and just to see where each item is in the chain, but it's really not bad at all.
Audience Member 4: My question has to do with the storage of electronic data and the discovery issues that come along with that. So how are discovery issues handled with this data?
Kosiorek: Well, we haven't really had a discovery issue yet, but if there was a question, I mean, I could always go back and print out the e-mail notification. The requests for analysis … if a detective is going to request an analysis, usually what happens is they're already submitting the item to us, so we're having an evidence submission form. So usually that information is the same. I guess if it were different, then I would have to print it out, but really we haven't had that issue.
Gorrell: Yeah, and one of the reasons we actually developed the tool so that you can actually take an electronic version of the report or pdf or whatever and attach it to that record within FIDEX. And Boston PD for a lot of reasons, just like what I think you're talking about, decided they didn't want to do that and so for the specific results of the examination, you can just not use that feature, that capability, and actually disable it.
Jim Markey: Thank you very much for this opportunity. I'd like to thank NIJ, NFSTC, Waterhole Software for giving me this chance to come up here and speak.
I'm an investigator. I use information and exchange it with other units as a team all the time, and so I've never been accused of looking backwards. I'm always looking forwards. So, when I saw the initial presentation of FIDEX, probably about a year ago, I was thinking, “Hey, you know, that's great for the lab; that's wonderful; they're going to be able to organize themselves with the prosecutor, hopefully, but what about me? What's that going to do for my investigations?” And I'm going to go through … I'm going to outline some major cases that we worked where information exchange was good and some things weren't so good, and then I'm going to talk a little bit about cold case and where I think a project like this would fit into cold case as well.
I'd like to talk about … or I should say the information that I'm going to share about two of the cases that I'm going to talk about, they're public record. One of them, the individual is still going to trial at the end of this year for about eight murders. So it is public information, and so anything that I share with you is already out there in the public media as well.
So, if you want to play that first clip.
[Video clip presentation begins.]
Eight homicides, five sexual assaults, 10 kidnappings.
Tonight, Phoenix Police are finally saying “case solved.”
This man never again should be able to walk the streets of this valley or anywhere else in freedom.
The evidence that we used to review these cases consists of forensic evidence, physical evidence and circumstantial evidence. All of this evidence points directly to Mark Goudeau as the one who's committed these crimes.
Police say they received 8,000 tips, interviewed 3,500 people and followed up on 2,500 pieces of information. Not once did they come up with anyone but Goudeau. Still, his attorney says his client didn't do it.
[Video clip presentation ends.]
Markey: I don't know if you caught those numbers that were put out there, and that's one of many investigations that we had going at the time that became an information and management nightmare.
A good friend of mine, Ritchie Martinez, who is the president of the International Association of Law Enforcement Analysts, told me one time, “You know what, Jim, failing to manage your information will fail your investigation every time,” and he's absolutely right.
And so I learned that lesson a long time ago. Some of it was the hard way and some of it, I think, as we look towards the future, we can manage this information much better.
You know, I did a couple presentations and I talked about all the crimes, serial crimes that we've had in Phoenix, and somebody came up after the presentations, says, “I don't really want to come to Phoenix.”
Markey: And I thought it was because of Senate Bill 1070 that they were boycotting, but, apparently it's because we do have some issues.
Since I've been in the sex crimes unit since 1998, we have logged well over 60 serial rape suspects that we've investigated and arrested and prosecuted most of those.
These three folks right here … and I'm going to talk about David Wilson at the top, Mark Goudeau on the left and the serial shooter case. You can Google any of those, and there's so much information on the Internet, but, unfortunately for us, those three cases were occurring all at the same time. They totaled about right around 20 murders and probably somewhere in the vicinity of 20 rapes between those four individuals that we were trying to manage with three different investigative units. And guess where all that evidence and all that information was getting funneled to. The crime lab. So you talk about resources and you talk about having to manage your resources. I'll talk a little bit later on about middle management and my term that I use for that.
Some things, you just … Hollywood can't write a better script. When we finally arrested Dave Wilson, we found out a bunch of things that we knew and really didn't know and then came to find out later on he was the son of a sergeant on our police department, and the unit that picked him up the night that we were doing surveillance, because we were managing our information in our crime analysis unit, was the same unit that his dad had just transferred out of as a supervisor.
So, about a month before he had left this unit, we picked him up that night on a trespassing, and then when we went to do the search warrant on his car, we looked at his license plate. And what does that say to you? That's not Dr. Jekyll's. He laughed about that. He goes, “Oh, you think that says 'Dr. Jekyll's.'” That's D.J. and Kelsey, his two kids. So, as he would go out as Dr. Jekyll and Mr. Hyde and commit these crimes, he would go back to his two children. So Hollywood can't write these things sometimes any better.
So we had to manage a lot of information, and our resources were very limited. It took a long time to get the wheels of the investigative process moving to where I knew we needed assistance, but I had to be able to go to these other units and convince them, with this information, we have some solid ideas of what we want to do.
We started to see patterns in our crime analysis of where he had tendencies of areas that he liked to commit his sexual assaults in, and so we were able to manage our resources, and we had — at this point, we had five different surveillance teams in five different apartment complexes, including the apartment complex that he lived in, which … managing just that part of the investigation, you have to understand the apartment complex he lived in has its own zip code. That's how big this place was.
And so we used information. We used crime analysis, and we used just some good old-fashioned gumshoe work and ended up with Mr. Wilson one night in one of our surveillance areas. But this is the old school of how we would manage all of these sexual assault examination kits and any other evidence that was funneling through the lab. We didn't have a communication system that we could access in real time. We were having meetings every day. We would include everybody in these meetings, and I'll talk about working one of these major cases and the amount of information that gets shared or doesn't get shared and how that impacts your case.
But, as you can see, on the right-hand side, we were getting a lot of elimination samples from a lot of folks, and so we wanted to have that information back in real time, because, if we have a surveillance team following one of our leads around, we want to know as quickly as we could can we pull those resources away from that individual because we know that's not him. We were very fortunate in this case because we were getting DNA samples from those sexual assault kits that were early on linking these cases together, so they weren't hunches. It was actual solid information from the cases that was allowing us to link this together.
And then we wanted to know who would screen it, and we would be able to go to that person and say, you know, “What items were screened? What items weren't screened?” We actually went line by line through these sexual assault kits. You know, they show up as one item usually when they get impounded and go to property, but when you open them up at the lab, I've seen as many — and I am sure you have, too — as many as 30, 35 different items inside the kit, depending on the circumstances of the case and how the nurse did the examination.
So it became tracking this information nightmare, and then we had to work — when we finally arrested him, we had to exchange and we had to work information with our Maricopa County attorney's office. So we had to sit down with them, and we went through basically department report by department report, and we identified every count of sexual assault, kidnapping, everything we could on each one of those cases. And this is how we tracked it. This is how we exchanged the information. This is actually — with my notes, this is actually one of the original pieces of paper they used for charging when it went to grand jury on this, and as you can see, we ended up with about 60-some counts on this individual.
And then we had Mark Goudeau. It was interesting. The night that we arrested David Wilson, Mark Goudeau was on the west side of Phoenix raping two sisters at a park. We didn't know that at the time because all of our resources were concentrating on this big arrest, with the media and everybody getting their ties on and their badges polished so they could go in front of the cameras.
But he was out doing his second rape, and we were able to link those fairly early on basically through his MO. What we didn't know in the end is, retrospect, looking back at the crimes that he was committing became a very daunting task of trying to manage the information from not only homicides, armed robberies, sexual assaults, but a homicide that was outside our jurisdiction that happened in Tempe. It was one of his first homicides.
We kind of got caught in the middle on this one. The information started flowing in as far as leads and tips and other types of communication that we weren't prepared to handle at the time. We didn't have a database. We didn't have a tracking system. It was one of these on the fly, can we develop something, go down to our IT people, can you give us something, because we're already behind the curve. And so it became very important for us to try to manage this to see where we were, what we had and where we were going to go with this investigation.
Again, we started using the crime mapping, the crime analysis to try to identify the areas in which he would have a tendency to maybe go back and offend. This is the first time in my 28 years … my sister-in-law ended up living a quarter mile from where he lived, and he had committed … up in the upper right-hand corner, off of 32nd Street and Thomas. She lived at 33rd Street and Thomas. And it was the first time in my 28-year career, I told my family, “Do not go out of your house between the hours of 7 p.m. and 10 p.m.,” because that was his prime time. That's what he liked. That's when he had some free time to do whatever he needed to do.
We were 15 minutes away of preventing probably seven murders. We had one surveillance guy that, for whatever reason, he took an interest in our unit when we were trying to get resources and we couldn't. And he said, “I'll tell you what, what are you guys thinking?” We said, “We think he's going to be up around 32nd Street and Thomas.” He sat there. He was part of the SWAT team. He sat there and he got a call out. He had to go to a barricade. Ten minutes later, our suspect commits an armed robbery shooting and attempt kidnapping a female.
I can guarantee you that if Billy Clark, who's the surveillance officer who was there, he would have killed him, and we would have ended anything because he has a tendency to be a really, really good shot.
Markey: That's why we put him there because that was the only way … you know, he was either going to be dead or he was going to kill somebody, and, obviously, he decided to start killing people.
The other thing about information is the night that he raped the two girls — and we'll talk about the two sisters in the park — patrol ended up taking disposition on that, and they thought they had found the crime scene. Well, we brought them back in the next day for a comprehensive interview with one of our sex crimes investigators, and we realized they probably didn't have the right location. And so we went back out and we found the location, and we actually identified additional evidence at that location. So now we have more evidence that we have to try to track, and that was very critical in the prosecution's case that they wanted to show to the jury. And, you know, we had to say, hey, you know, we missed it the first time, but we went out and got it the second time. So, when you're talking about these major cases, you know, you're talking about managing resources. You're talking about managing people, and, really, you know, managing people is more like leading people, I would say.
You know, you manage things, production, output, but really team building, communication, that's what managing people is all about. But you can't do any of that unless you manage your information and everybody is on the same page and has access to the same information and everybody is heading in the right direction, and that becomes difficult at times.
So, when you're managing information, especially in one of these cases, you know, tips, leads, evidence, communication becomes so critical, and so how do you overcome those types of issues? I don't know if there is the pie in the sky that's going to solve every communication issue, every management issue of information out there, but I can tell you at least we're heading in the right direction. At least we're looking at solutions that I think can be applied to a variety of situations that we all get involved in.
So this would be a typical communication breakdown or a communication task force of how it would be developed. Can you imagine trying to manage all the individuals that are assigned to each one of those little boxes as well as the information that's being exchanged? I mean, somebody would tell me something at a meeting, and the next day I'd never remember it. You know, it wasn't getting documented, and it wasn't getting exchanged. It wasn't getting massaged. You know, we weren't mining the data like we should have been at that stage.
And this is my futile attempt at PowerPoint stuff. You know, so you set up a war room, and now you've got a major case and you've got all this information, this data, and you've got all this evidence and the stuff flowing in, but what actually happens to it? When you break it down, where does that information go? Someone has to deal with it, right? Eight thousand tips. So how do you deal with 8,000 tips with the fear that the one tip that's in there that you don't do appropriate follow-up on is your suspect, and he goes out and commits an additional crime?
So, you know, when you think about, well, managing this information, well, there's got to be a process to do it, and we set up a little process basically when the information came in, no matter how it came in, this is how it worked, this is how it flowed, and, hopefully, we were able to identify, assess the good information and filter out the bad information that was not for us. Of course, sometimes you just don't know what happened in a case, and that's why we have homicide detectives.
I'll talk a little bit about cold case. We started in 2000 with a VAWA grant, and we have developed for the last 10 years a database, internal database to track this information.
I'd throw him in there because his name is Jack Russell. I know that seems ironic, but he committed several sexual assaults in Phoenix and then fled to El Paso, Texas.
But here you have a team, and how can you be a team if you don't communicate and understand the information that each other has and how it's important to share that with other folks? You cannot be successful in a team without the exchange of data, information, talking to each other, being able to look at archived data and massage it and pull out the stuff that's appropriate and that you need in your investigation or your prosecution or in the crime lab to test.
And so, you know, you're a team, and you may not meet as a team, per se, but you're all supposed to be heading in the same direction. You're supposed to be moving towards the goal, and the only way to do that is to have this information available to everybody.
So, when you're talking about cold case information management … you know, we talk about how we set up our team. We just didn't shotgun everything to the lab … well, we kind of did, and then they panicked, and we gave them a body from our unit and said, here, use this person to screen some of these kits that we had back in 2000 and 2001. But we went through our case. We reviewed our cases. We put them into a database, and we were able to track our leads, timelines, milestone dates on evidence, on talking to leads. I'll show you one of the screens we had.
This is a very simple database that really is outdated at this point, but we continue to use it because, at this point, we're looking for options. We're looking for something in the future. You know, you need to track and review evidence and identify where your backlogs are, and as that evidence moves through your system, one of the questions that my detectives get asked when they go to grand jury on these cold cases is the grand jury wants to know timelines, OK, when was that kit obtained, when was it sent to the lab, when was it examined, when did you get the CODIS hit. They want dates, and then, of course, the last question is “Why did it take so long?” And then my detectives have to sit there and try to explain the process to the grand jury, but they're asking these questions. We wouldn't have that information unless we were able to track it effectively.
You know, the other thing is statistics. You know, every time we would get a grant, we would have new requirements to provide statistical recap of what we accomplished. You know, people are giving us money. They have a right to know what we did with that money, and we should be able to give them numbers and explain exactly how that money was used and what it was used for, and, hopefully, they're successful numbers.
You know, we're looking for a process to link cases, to look at series and analyze those series for information that can be shared by a team and accessed by a team. You have to trust each other on that team that that information is going to be used as part of that investigation. You know, these cases are cold for a reason, and now we have to go back and bring them alive again.
So that's just one of our basic admin sheets. You know, I can look at this and I can tell you that we have available evidence, and I can also tell you if there's DNA on a suspect. There's a typical identified suspect offender in one of our cases.
So, I guess, you know, I'm saying we need to be prepared. We need to look to the future because I'm not going to be here in 10 years doing this, but I want to leave it for somebody, that it's not a mess that they have to clean up, that they have to start all over from scratch. We're here doing the job now. We need to ensure that we're doing it effectively and efficiently and managing this information for really the future.
Don't become a middle manager. You know, when you're at 30,000 feet in an airplane, that's not the time to think about “We should've built a better airplane.” It's too late; you're up in the air. And that usually is what happens. We manage by crisis. The crisis is going to be here.
There was a presentation last week at the death symposium in Scottsdale, and they were talking about 9/11, you know, how much information, pieces of data, millions and millions of pieces of data that they had to track, everything from forensic information to lead information to dealing with families. They had to build something on the fly. Don't put yourself in that position. It's very hard to get your administration to look at this as a real issue.
I'm sure you have been in a jurisdiction where they've had some sort of plane crash, you know, with mass fatalities, where now you've got to manage all that, not just the crime scene but manage all that information that's occurring on the fly.
Know where you're going in order to get there.
[Video clip presentation begins.]
According to police, he terrorized the valley for months. Tonight, the accused baseline killer, Mark Goudeau, knows how long he'll spend behind bars for the sexual assault of two sisters.
Mark Goudeau in court, as he's sentenced to 438 years in prison. Here's an idea of how long that is: Four hundred thirty-eight years ago, in 1569, William Shakespeare was just 5 years old, and the first colony in the United States would not be settled for another 38 years.
[Video clip presentation ends.]
Markey: Amazing what the media will come up with.
Markey: What does William Shakespeare have to do with a serial rapist and murderer?
You know, I'm really proud of the teamwork and the cooperation. We made it through this crisis, and the case, the information, the lab work, it withstood a very vigorous defense in trial and convicted him 438 years, and that's really what we're working for.
And I thank you very much.
Aaron Gorrell: One of the times that Jim and I have talked, one of the things that he talked about specific to information management is when you look at your relationship with the community and the relationship with the media, one of the things that he stressed for me is that relationship is not going to be based on the work that you do on a regular basis, all the successes that you have on a day-to-day basis, the, quote/unquote, “simpler cases.”
It ends up being judged by those major cases, those ones that have thousands of pieces of evidence, and whether or not you manage that effectively and efficiently goes in large part to determining how the public and how the media, how successful they view you.
We view FIDEX and the version that you're going to see and the version that Kevin has been using and that Mike has been using, Jim has been using — we view it as a stepping stone, and that's the idea of these presentations is we have certain capabilities now, the capability which is focused on making sure that information is shared between organizations in a very fluid way. There is also — and where we're hoping that FIDEX can evolve to, a FIDEX 2.0, if you will — is to start looking at managing some of that information where forensic examination becomes a portion of that, but you start looking at integrating information in from multiple other data sources.
So, to that end, when we started looking at it technically, from a technical perspective, how we wanted to design the tool, what functionality and capabilities we wanted it to have, one of the baseline requirements that we identified is we had to have multiple configurations. Every agency is in a different place, frankly, from a technical capability standpoint.
You have some organizations where you have 10 police officers and five of those might be part time, and we wanted FIDEX to be accessible to them as well as the major metropolitan police area. So we designed it with multiple configurations in mind.
In its simplest format, it's a web-based application with a database, that anyone that has access to that server and access to that particular case can receive information, all the way to the point where we implemented it with Boston Police Department, where it's actually, once you enter that incident number, it's going out, it's grabbing that information from the evidence management system. It's grabbing that information from a records management system or a booking system and creating these links electronically and everywhere in between.
And then, as I mentioned earlier, we also wanted to make sure that this did not compel any particular work flow. When we take a look from a data exchange standpoint at the kind of information that agencies wanted to collect, there's a lot of consistency between agencies. There were wide variations in terms of the specific work flows. We wanted to make this thing work flow agnostic, if you will. And we also wanted to build an open system. So, if you have developers, if you have those internal IT resources that are capable of expanding and extending the functionality of the tool, they're able to.
So what I would like to do — and this is always a mildly risky proposition — is actually give you an online demonstration of the FIDEX tool based on one particular scenario of usage.
So the concept here is kind of to follow BPD's, Boston Police Department's, way of processing a case, which essentially starts off with a detective or the crime scene team enters information to the evidence tracking tool and then the detective may or may not take a look at that and fill out a long form by hand and say, “For this piece of evidence, I'd like these examinations,” and so on and so forth.
Now, the way that we ended up adjusting that business process is we still assumed that the crime scene team or the detective is going to enter information into their evidence tracker tool. It's a legacy tool set, and what we want to do is actually import information. So what they do is they enter an incident number, hit query, and what the tool does at this point is it goes out to all interconnected systems, says, “Here's my incident number, give me any and all information that you have about this particular case.”
Now, in full disclosure, I'm running this off of my laptop, no Internet or anything like that. So I'm actually pulling information just from my laptop, but the way we have BPD configured is to go out to these various systems.
So once a detective opens up or anyone else involved in the case opens up, they can see five or six tabs across the top of the screen. One of the very fundamental requirements, quite frankly, that Sergeant Markey and other police officers drilled into me is this thing has to be simple. You've got to make it simple to interface with. You do not want to overwhelm people with information. So we created … we identified the six kind of areas of information that seemed to be the most relevant, such as the request, the date and time of the incident, some basic incident location information, basic information about the offense, including offense code, and if we had a court case management system integrated into this and it had gone to that point of actually initiating a court case, we can pull in the docket number and the current status of the court case.
But, probably, the most important tab is the second one, the evidence tab, which is pulled in from that evidence tracking system, and one of the things — and Kevin touched on this briefly — one of the things the tool does is every night at about one o'clock in the morning, it goes out, asks the evidence tracking system, “Do you have any additional evidence associated with this case?” and it's going to do that on all active cases. And if it does, it adds it to this list and, maybe even more importantly, notifies anyone associated with the case that that additional evidence has been added.
So, again, taking on the role of an investigator, I've got these 11 pieces of evidence and let's say I want to add an examination, and I'm going to add it to items number two and three, and I'm going to request that the lab compare the casings to the projectiles. The discipline requested is firearms, and the service requested is … we'll just pick this one. These dropdown lists are configurable. So whatever your lab has for these different areas, discipline and services, you can certainly standardize on those and add the examination. So now the detective and anyone else associated with the case can go into the tool and see all of the examinations that have been requested on any particular piece of evidence.
Now, the BPD, as it is with Phoenix Police Department, one of the next steps is, again, they've got to include that long form with the physical evidence, and because we have all of the information that we need for that long form, what we did is we re-created that submission form so that they could electronically generate that straight from the tool.
Now, generating this form does a couple of things and potentially three things. Number one, the detective would take this long form, print it out and attach it to the physical evidence when they're transferring it over to the property and evidence room of the crime lab. They still want that piece of paper.
The second thing it does is it notifies a supervisor in the crime lab that additional evidence is coming your way, giving them kind of a heads-up. And if we had this hooked into a Laboratory Information Management System, what it would also do is essentially pre-populate that management system. All the information has already entered about the evidence. All the examination requests are there, and the idea is to avoid this need for redundant data entry where your front office clerks accept the evidence, have to re-key in all that same exact kind of information.
Then from there, the idea is that the supervisor assigns a criminalist who takes over the examination. Assigning it is pretty straightforward. We pre-populate a list of people that have been assigned to it. Now Claudia has been assigned to this thing, and now she has authority to see the case. She also will receive all e-mail notifications associated with that case.
The other tabs accumulate information from other external systems. For example, the suspect in this case, Gregory Knight, was booked in on this case, and we actually grabbed this information in from the booking system, including the booking number.
Case parties is where you start managing who receives notification. So if you assign an analyst, they're automatically added as an interested party for this case, but you can also add any other number of people, whether it be detectives, whether it … it might include supervisors and so on and so forth. You can add other people and assign a role to these folks to make sure that they receive these automated notifications.
Then, finally, we also talked about needing to know about these court hearings that are coming up. So, if you're interfacing with a court case management system and they're associating or you have a way of linking it to the incident number, which most case management systems do, you can start receiving electronic feeds of upcoming court dates for this particular case. And, again, it actually has a piece of functionality in there, configurable, that if you've got open examinations and you've got an upcoming court event within a week, it will send everyone who's interested in that notification out an alert to say, “Hey, you've got something going on here. You may need to either continue the case or maybe the data has been updated or whatever,” but it just gives everyone a little bit of heads-up using existing systems, using your existing e-mail networks that you already have.
Now, one of the functionalities that, quite frankly, we almost discovered by accident by talking with Boston Police Department is this need for statistical information. Even agencies that have fairly robust and sophisticated evidence tracking systems, they seem to be kind of short. These systems don't seem to be shipped with great reporting mechanisms, but we've got a lot of that information in the FIDEX tool.
So a supervisor or anyone given the authority to can click on things and view in real time what is your current backlog, what kind of cases are being requested. This information has actually been accumulated primarily from Phoenix Police Department, and so you can see for the last six months, IAFIS requests constitute about 50 percent of their examination requests, which seems about right.
You can also start looking at it from a managerial standpoint — where are my backlogs, what examinations have been submitted to particular analysts and what is that status of this. And we ended up removing the full name of the analyst, but you can see they can see what examinations are currently open based on who it's been assigned to, and even the color indicates the kind of examination that has been requested of that person.
And I will tell you when we ended up showing this chart to a crime lab director, they were surprised by the result. They had no idea that these kind of examinations were open, and so they found it extremely valuable.
And then the last one I want to talk about is very much … it's going to take a little bit. It's doing a lot of work behind the scenes. It's very much a theoretical analysis. What we wanted to do is take a look at if we can get a list of all the evidence that is currently, quote/unquote, “in the property room,” which assumes that it's been put into the evidence tracking system, and compare it to examinations that have been made against that evidence, can we take a look and identify situations where you have biological evidence that have no corresponding examinations on it, which is, as you know, one of the big issues.
And what we ended up finding, using a subset of data, is that there are 29 cases where it seems like evidence had been collected, a rape kit had been collected, but no examination had been performed on this thing.
Now, this, as with a lot of these kind of tools, it ends up giving you hints. It gives you an area to look at for potential follow-up, and, in fact, we did end up following up on these, and, in most cases, it was a data entry issue. It was examinations were outsourced to an external crime lab, and the results of that aren't put back into the one location that we're pulling information from.
But the idea of the notion is that we might even be able to start identifying pieces of evidence that had been overlooked for whatever reason, whether it's because the victim wasn't willing to testify or the technology wasn't there, because this is the case that's 25 years old.