# PARPA should create a replicable institutional model The combination of an inherently unscalable model<[[Research orgs don’t scale]]> and relative robustness of institutional models compared to individual organizations suggests that it’s important to explicitly think about how to create a replicable institutional model while building PARPA. While a single organization riffing on ARPA will hopefully be able to accomplish many important things, it will be structurally unable to do *all* the important things. Unlike a company that can conceivably monopolize an entire market and provide all the search, or videos, or diapers, DARPA’s small, flat structure is probably essential to their success ([[DARPA is relatively tiny and flat]]) and any organization following a similar model will see diminishing returns with size. This effect is similar to what you see in venture capital firms and many private equity firms - small firms are able to get crazy returns on outlier results while large ones revert to the mean. Additionally, blazing the trail for new institutional structures is far more impactful than any given institution. The impressive thing about Sequoia and Don Draper is partially the investments they made, but more impressive is that they created a template for an entire institutional model used by hundreds of venture capital firms. In the same way that canonical media set can set a new standard for a medium — *The Godfather* for movies, *The Feynman Lectures on Physics* for textbooks[^1] — canonical organizations can set a standard for a type of institution. Bell Labs is the canonical corporate R&D org, Apple is the canonical electronics company, The Rockefeller foundation (or perhaps now the Gates foundation) is the canonical large philanthropy, the Rutherford lab was perhaps the canonical early 20th century physics laboratory. Aspiring to add to this list is admittedly ambitious and perhaps the sort of thing you can’t set out to do, but worth attempting regardless as long as you don’t let delusions of grandeur hamstring getting things done. Creating a replicable model is also a way of hedging your bets. Contingent factors can always kill a given organization even if it gets the model correct. Or perhaps like [[ARDC]], arguably the first venture capital firm, you get the idea and some pieces correct but screw up others (like legal structures) in an ultimately fatal way. What would you differently if you’re trying to create a replicable model vs. just building a one-off? One piece is to consciously document decisions - why you do one thing over another - and failures - things you tried that didn’t work. [[Written decisions are especially important to PARPA]]. Additionally, it means forgoing a level of secrecy and being open to helping people who want to do something similar. ### Related * [[Can you use transparency to earn enough trust to be opaque?]] [^1]:<Footnote>The concept of canonical media and these examples are from [How can we develop transformative tools for thought?](https://numinous.productions/ttft/#serious-work). <!-- #evergreen #project/spectech/org --> [Web URL for this note](http://notes.benjaminreinhardt.com/PARPA+should+create+a+replicable+institutional+model) [Comment on this note](http://via.hypothes.is/http://notes.benjaminreinhardt.com/PARPA+should+create+a+replicable+institutional+model)