Version 1.0: You Have to Start Somewhere
I spoke earlier this on a panel about Text-to-911 at 9-1-1 Goes to Washington. As you probably know, the FCC has been very active in promoting Text-to-911 in order to enable both equal access for the deaf and hard of hearing communities and to support emergency communications via one of the most common forms of communication in use today.
David Furth, Deputy Chief Public Safety and Homeland Security Bureau of the Federal Communications Commission, made a comment that really resonated with me and, based on the comments I received after the panel, many of the attendees. He stated that we have to think about Text-to-911 as a version 1.0 of NG9-1-1. I couldn’t agree more.
The cost of failure in public safety is astronomical. This isn’t a video game where the only damage is disappointed teenagers if it flops. Every decision, whether technology, process or regulatory, must weigh the benefit against the risk-adjusted cost. Benefits and costs are often not measured in dollars and calculating the risk is not easy. In this case few argue against ultimately enabling Text-to-911, the discussion really centers around the risk associated with enabling a less than perfect solution. Will we have lost messages, operational issues, training failures? Yes. And yes, the cost of those types failures are high. The question is at what point do you say it is acceptable to take on those issues because the overall benefit outweighs those costs. Does one failed attempt to communicate via text message negate 10 lives that were saved because someone was able to communicate effectively? We can’t really say that today’s system is completely bulletproof can we? Outages, mis-directed calls, and training failure still occur on occassion.
To be clear, I’m not arguing that we should jump willy-nilly into a solution. Actually, quite the contrary. What I’m saying is that you have to start somewhere, and be willing to learn and adapt. If we don’t, we’ll either never get started or we’ll end up building something in a vacuum that doesn’t actually solve the need. In software development, the trick is always to balance the baseline requirements against the bells and whistles you’d like to have. Version 1.0 needs to solve the core problem you are addressing, not every problem you can envision. The only thing I’m ever sure of in new product development to write my essay for me is that you can never envision every usage scenario that will arise – and you if you try to solve all the ones that you do envision you’ll end up chasing your tail and never getting a solution deployed.
Think about the evolution of E9-1-1 as an example. We all knew that Phase 1 was not the end game, but it was a natural Version 1.0. Locate the cell sector that the call came from so that you can at least route it to a nearby PSAP. Phase 2 comes along as the second version and provides more accurate location, but even those accuracy requirements morph as we realize where the gaps are and what is possible. Processes supporting the technology evolve, and the technology then evolves more. Then new challenges like non-initialized phones come along that throw a wrinkle into the works at https://grademiners.com/case-study-help and we start to look at how to address that.
Let’s look back at Text-to-911 and some logical assumptions based on observed facts:
1) We aren’t going to get suddenly flooded with text-to-911 calls if the FCC decides on a preferred interim solution. Text-to-112 has been enabled in some European countries for several years and well marketed. The rates of text calls are extremely low (sub 0.1% of calls). We will have time to roll something out, learn and adapt. We can also easily phase the roll-out starting with those in most need like the deaf and hard of hearing communities.
2) Text calls take longer than voice calls. The same European countries mentioned above see text calls lasting 11 minutes on average. Does that mean we need to avoid it? I don’t think so, I think that means we need to build some processes around shortening those calls and enabling call takers to effectively manage multiple calls at once. Much of that time gap is inherent in SMS – maybe RTT is a better method. Maybe automated scripts should be enabled to identify non-emergency calls or at least pre-triage the calls? We won’t know until we get some experience.
3) Text (in all its forms) is pretty reliable. Is it perfect? No. But, then neither is my phone’s voice connection. Average dropped call rates in the United States are between 2.4% and 4% depending on the carrier.
I often hear “good enough isn’t good enough in public safety”. I disagree. Sometimes good enough is better than the alternative of having nothing and we have to start somewhere. The FCC Emergency Access Advisory Committee sponsored an Exhibition Fair to Text-to-911 Technologies and Applications (March 28 and 29) that showcased the spectrum of interim solutions available. Each has its pluses and minuses but clearly showed what is possible and what innovative companies can dream up to solve a challenge.
Whether it be texting, IP-based call handling, location assignment service, or any of the future communications technologies enable by NG9-1-1, we need to take the approach that version 1.0 is a good starting point and make the commitment to keep improving.