flâneur

How I Learned to Stop Worrying and Love Intelligible Failure

issues.org · 4,134 words · saved by 2 readers

With the growth of “ARPA-everything,” it’s time to start testing predictions about what makes the model succeed. That means being able to understand when and why it doesn’t. When I was appointed by the US health secretary to serve as the very temporary, de facto leader of the newly created Advanced Research Projects Agency for Health (ARPA-H) in May 2022, I knew I didn’t know enough. Or at least not enough to ensure that my small team and I could set up an ARPA to actually do what it was supposed to, rather than end up as a cargo cult with all the ARPA window dressing and none of the substance. Allow me to explain. ARPAs are built to produce innovation. They are charged to bring about world-changing advances through technologies that do not yet exist. The examples everyone trots out are ARPAnet (the predecessor to the internet), GPS, and self-driving vehicles. Delivering such technologies requires taking risky bets on ideas that could be big—if they work. The question, as I see it, is

How I Learned to Stop Worrying and Love Intelligible Failure By Adam Russell With the growth of “ARPA-everything,” it’s time to start testing predictions about what makes the model succeed. That means being able to understand when and why it doesn’t. When I was appointed by the US health secretary to serve as the very temporary, de facto leader of the newly created Advanced Research Projects Agency for Health (ARPA-H) in May 2022, I knew I didn’t know enough. Or at least not enough to ensure that my small team and I could set up an ARPA to actually do what it was supposed to, rather than end u

saved by

related reading