As a parent it’s been awful to avoid play dates. I’ve been brainstorming about an app called PlayDate or Contact Calculus or maybe Contact Planning. The idea would be to be able to share your profile of behaviors with other people so you can coordinate things like play dates. For example, we could explain that our kids have had access to a total of 4 people and only been inside two buildings, our home and my parents’ home, since February and share that with other parents so they could see our safety level. The idea would be they share too and we can all make informed consent decisions. It’s not only practical, it’s the morally right thing to do to enhance human agency, as a well designed version of it would expedite re-opening and reconnecting for mental health.
The core things that feel important are the ability to safely share information like exposure and processes, and the ability to talk about places and events. For example, hosting a birthday party with a bunch of country kids who haven’t left their ranches or woodland properties for a month would register as very low risk. So it’s the ability to do it and the feedback on how risky or not something should be considered.
Mathematically, it’s important to normalize these kind of micro-network pods to help people reconnect, and statistical risk prediction, opportunity cost analysis, and any number of libraries of ways to use the data collected from the user about their behaviors; it’s important because people who aren’t risking much more to have a more normal life absolutely ought to do so. i.e. Someone who is a frontline restaurant worker and someone who is a frontline grocery store worker are not personally risking a lot to allow their kids together, in terms of how risky the virus is to their family. Of course there is a macro risk of contagion across networks, which I get, but the macro risk of the collapse of mental health and child development networks profoundly outweighs it.
The platform - once up and running with a significant number of users - would also allow people to “calculate” different risk models, and use the latest science to analyze choices. A short case study on how that might work: in a month, we find out that wind speed correlates inversely to outdoor transmission, which probably means breezy days are better for outdoor events, not stagnant humid days. So a barbecue is rescheduled from Sunday to Saturday because Sunday will be 15 mph more breezy, which means there is a net risk reduction big enough and enough at risk people to merit the frustration of change.
A benefit about the IRL-UX would be that people can rationalize their decision making in a way that feels empowering, not prescriptive, but is still rationalized against a shared objective reality of analysis, not hearsay and media parroting of pseudoscience. A real life example is that I have a friend who came up for an outdoor visit who was worried we would transmit to her; she works in a frontline public-facing indoor food service facility. In the app it would be wildly clear that we were very safe and she was a risk to us, which we were happy to take to have dear friends visit.
I have a strong instinct that, with enough data, Bayes-like eigenvector prediction models will forecast transmission events without knowing whether or not anyone is infected by calculating - basically - net and adjusted risk scores. If the coronavirus is spreading like seeds in soil, the way events and interactions are handled is the soil. We may struggle to predict behavior of seeds, but we’re totally in control of the soil. We can make sure the coronavirus seeds don’t grow by making sure the hyper contagion events are seen as bad by people themselves, not top-down government rules.
In conversations with several very smart people, I’ve started to see some momentum around actually building something like this, but following a real product design and development process, not pontificating on what to do with late stage data wizardry. I’m hoping I can help and excited to work more towards that.
In closing: we’ve got to change the metaphor from Contact Tracing to Contact Planning. People deserve the capacity to plan not react.
For nerds: a rough model of data architecture in google sheets.