Before it was HIFLD, it was briefly FGDWG. Before that, it was a nameless thing in its infancy. In those early days, it bounced around between conference rooms in Norfolk, Dahlgren, and Chantilly. I was fortunate enough to be in the rooms where it was born and took shape.
I was a contractor supporting an organization that had been doing critical infrastructure vulnerability assessments for over a decade at that point. I had already been doing geospatial development for our products. These were PDF documents ranging from forty to ninety pages describing how military bases relied on commercial infrastructure and potential vulnerabilities that could affect mission readiness at those installations.
On the foggy morning of November 1, 2001, our government branch head and I shook off our sleepiness from the previous night’s trick-or-treating with our kids and drove to Norfolk for a meeting that we’d been told to attend but really didn’t know much about. It turned out to be the meeting that initiated what would eventually become the HIFLD. The meeting included a lot of Federal civilian employees, contractors, active duty military, and recently-activated National Guard.
The significance of the meeting wasn’t really apparent to all of us at the time. Among other issues, it acknowledged the importance of understanding infrastructure and recognized the intensely geospatial nature of the homeland security problem that was before us. Eventually, our branch head passed around a copy of the report he’d brought. The Navy captain leading the meeting looked it over and said the information was exactly what was needed. Our branch head replied that we had over 90 more just like it. The reply to that was something like “Great! But a 90-page document is useless. Can you turn each one of these into a map that tells us what’s important and why? There’s a lot of great engineering information in here that I never want to see again.” Our branch head said we could.
When the meeting adjourned, everyone left thinking our product would be a picture – a CROP, or “Common Relevant Operating Picture.” The word “relevant” was meant to distinguish this picture from all of the existing COPs at the time that were perceived to be stuffed with information that simply got in the way of making a decision.
As we drove back to Dahlgren, the branch head looked at me and asked “Now, can we really do that with our reports?” I said yes and then broke it down for our team when we got back. The government tech lead was ecstatic about the challenge and we got to work. We cobbled a proof-of-concept together using a locally hosted ArcIMS server and showed it at the next meeting, which was in Dahlgren. There was a positive response to the progress we’d made.
By the next meeting in Norfolk, talk was already shifting to the fact that the product wasn’t a picture – the product was really a common set of data that would go into a picture to ensure that everyone saw the same thing. At this point, the thinking still centered around analysts working in desktop GIS using local file systems. DHS was still a couple of years away from existing and DoD IT policy hadn’t caught up to the kind of data sharing we were going to need, so configuration management of data was the key issue. HIFLD had yet to be born or named, but it had just been conceived.
The operative word of that meeting was “if.” If we could dynamically layer data from authoritative sources. If we could visualize it. If we could make it available for download. If, if, if…
Those of us who’d traveled down from Dahlgren looked at each other and agreed that we hadn’t heard anything we couldn’t do with technology we already had. We also agreed the discussion would be stuck at “if” unless they could see something. So we knew what we needed to do with our proof-of-concept. We had a month until the next meeting in Dahlgren so we set about wrapping our ArcIMS site with what passed for web development in 2002. We still didn’t have authority to connect to outside servers – and they didn’t really exist anyway – so we set up separate servers hosting different data to simulate accessing external feature services. Our Esri rep at the time gave us top cover to over-install ArcIMS because he saw the value of what we were doing.
The next meeting started off with a lot of “ifs” and you could see the uniformed staff growing impatient. The word had begun to spread about the meetings and attendance from other organizations was growing. Our tech lead stood up, asked for the floor, and proceeded to demo the results of our work. He concluded by saying that the technology already existed to build the picture we needed. We needed to focus on policy to enable the necessary sharing and we needed to come to agreement on the data that would go into the picture. That aligned well with what those leading the meeting were already thinking and helped corral the others who were getting sidetracked. We didn’t hear the word “if” again that day.
The following meeting was also in Dahlgren and it finally got a name – the FGDWG….
I was reminded of this period because of the recent announcement that HIFLD Open was being discontinued. I haven’t been involved with HIFLD in a very long time, but it still holds a special place for me – particularly because of the intensity of the work and the significance of the problems we were trying to solve.
I will miss HIFLD Open. *It was convenient, but most of the data is available elsewhere. The idea of providing a subset of HIFLD data that could be accessed by the public was always a “nice to have.” Its core mission was, and remains, supporting homeland security and critical infrastructure protection use cases. HIFLD Secure will continue to support that mission with high-quality, relevant (there it is, again) data. And it will continue to evolve.
It would be impossible to name all of the people – many of whom have since retired or moved on to solve other problems, or all of the organizations – many of which no longer exist – involved in those early days forming what would eventually become HIFLD. The work we did in Dahlgren was one small part of a larger whole. It could have been unwieldy, but somehow it wasn’t. There was great leadership in those early days as well as great motivation from everyone involved. Despite the horrific circumstances that led to the genesis of HIFLD and everything that followed, there was an overall positive and collegial atmosphere to that early work.
For me, this was the beginning of my pivot toward data, rather than software, as the more challenging problem to solve. Data is the evidence and the exhaust of human activity and is as messy, inconsistent, and complicated as are we and our daily endeavors. Collaborative data efforts such as HIFLD, OSM, and many others demonstrate this as they push forward through negotiation and contention to produce pictures that are never quite perfect but limitless in value.
Header Image: Neotigen, CC BY-SA 4.0 https://creativecommons.org/licenses/by-sa/4.0, via Wikimedia Commons
