Darpa seems to stand out among organizations tasked with turning science fiction into the mundane. Having innovated in many fields, robotics, the internet, GPS, it serves as an organization worth observing if one desires to replicate the prolificity of their R&D.
Darpa, however, is largely limited to military work, and can't be applied to pure research in other domains.
If we are to model the success of Darpa in other places, there are some questions which should be posed:
Which emulatable attributes contributed to DARPA’s outlier results?
What does a domain-independent “ARPA Model” look like?
Is it possible to build other organizations that generate equally huge results in other domains by riffing on that model?
Darpa is largely ran by 100Â Program Managers, who hold 5 year appointments, to create and run research programmes that pursue ambitious goals.
These goals contain different projects, and the groups working on these projects are called performers, and the highest level authority figure is a Director.
Darpa sports an extremely effective framework, the Arpa Model, which could be fruitful if applied to domains outside of the defense industry.
Distillation
Metadiscussion: Replicating outliers and the DARPA production function.
Due to the nature of outliers, data-based analysis, and pattern identification may be ineffective for assessing the reason for DARPA's success: all outliers are outliers in different ways.
Three mistakes when replicating outliers:
Cherry picking characteristics you don't understand.
"If you believe in things, that you don't understand, you'll suffer"
Copying at a high/abstract level.
Trying to clone the whole organization.
The better method is to gain a wholistic understanding of the concept, and see what applies to your specific circumstance: copying outliers has not been very successful historically, but apprenticing to outliers has.
Thoughts: This is the reason why Radian Labs has to develop an educational curriculum that allows for people to learn not just via studying and doing projects, but being able to do so under the wing of established professionals.
You can see the effect of this whenever you look at the genealogy of the Titans of mathematics and physics, for instance, when you look at von Neumann, Wigner, and Szilárd, you see they were all taught by a master, Lászlo Rátz.
Part of the reason this works is because outliers are all outliers in their own way, and an outlier (or master), will be able to assess the way in which an individual can become an outlier, nurturing their unique abilities as opposed to fitting them to a mold or providing them a template.