# An Institute for Creative Technology - Riffing on ARPA
## Meta
- Could you keep track of a history of sections they looked at so they can easily jump around
- Could you leave stub links to note that these will be filled in later?
- Principle: assume people are jumping around
- Remember to pay attention to :: and <> in final piece
- Could you have an epistemological certainty indicator?
- Questions for editors
	- Where would an illustration or diagram help the point
	- Where do you want more expansion? Where is there too little expansion?
	- Should I used “R+D org” or “Innovation org?”

## Writing
### Preliminaries
- [[Purpose behind ICT Writeup]]
	- [[Creating an organizational structure around a project prematurely can have many downsides]]
- [[Assumptions of the ICT Writeup]]
- [[Who is the ICT Writeup For]]
	- The piece should be able to be read differently depending on who is reading it and what they want to get out of it. 
		- Choose your own adventure but you get to choose one of the premade characters or design your own. 
			- Maybe a colored trace depending on which you choose?
- [[Methodology for ICT Writeup]]
	- [[Thinking about PARPA requires unorthodox ways of thinking about institutions and incentives]]
	- [[Counterfactuals are hard]]
	- [[Incentives are the fundamental forces of peopleland]]
- [[Glossary]]
 
There is a lot of preliminaries or postliminaries that are important to building up the piece. However, it’s important that it stay tightly focused. To that end it’s important to note where there are clear holes and plan for them to be filled. 

- ::separate:: Why is a DARPA Riff Necessary - the full constraints probably want to be a separate thing. Maybe explicitly say that this is *not* to convince you of the thesis - this comes later.  See [[Assumptions of the ICT Writeup]]


- What problems is the ICT addressing (contra many other “research is broken”)
	- [[Overlapping institutional constraints rule out several classes of creative work]] 
		- [[Einstein would have been stuck in a patent office - failures of paradigm-shifting science]]
		- [[We have no flying cars - failures of paradigm-shifting engineering]]
	-  [[In the past, corporate labs filled a particular niche in the innovation ecosystem]]
		-  [[Ideally corporate R&D enables work on a general purpose technology before it’s specialized]]
		- [[Ideally corporate R+D enables targeted dicking around]]
		- [[Corporate labs no longer fill the niche they did in the early-mid twentieth century]]
		-  [[Healthy corporate R+D requires a trifecta of conditions]] 
		- [[AI research is an exception to the decline of corporate R+D]]
		-  [[Why try to emulate DARPA and not Bell Labs]]
		- [[Healthy R&D orgs feel like they have very smooth project rampups with high ceilings]]
AT THIS POINT I think is the correct place to introduce the fact that we’re going to be feeling around the edges for this so-far imaginary organization
		- [[One function of a private ARPA is to fill in the gap once filled by corporate R+D]]
	-  [[Things that are not paper-worthy enough for academia and not product-focused enough for a startup]]
	- NOT [[Einstein would have been stuck in a patent office - failures of paradigm-shifting science]]
	- Going after [[We have no flying cars - failures of paradigm-shifting engineering]]

	- ::Separate:: General Constraints of existing systems
		- [[§Philanthropy and Non-Profit Constraints]]
		- [[§Corporate R+D Constraints]]
		- [[§Academia Constraints]]
- Specific Takeaway around Innovation Orgs. ::Does this want to be a separate section or woven into the different pieces?::
	- On Profit and Value Capture
		- [[tensions]]
		- [[Profit makes organizations auto-catalyzing]]
		- [[There is a significant class of innovations that would create drastically less value for the world if their value had been captured by their creators]].
		- [[Value capture does not always need to happen through IP]] ::Stub::
		- [[It is hard to capture value from research]]
			- [[It is especially hard to capture value from externalized research]]
		- [[Decoupling from market discipline is like cave diving]]
		- [[Someone needs to buy manufactured technology eventually in a monetized economy]]
	- On legal structures
		-  [[Legal structures are intimately tied to the question of how money works in an organization]]
		- [[There are no first principles for legal structures]]
		-  [[The traditional non-profit structure is a bad idea for an organization that requires a significant chunk of capital and produces uncertain results on a long timescale]]
	- On Building Organizations
		- [[It is important for big things to start small]]
		- [[Research orgs don’t scale]]
	- [[Precise visions are more likely to happen]]
	- [[Prototyping needs manufacturing in the room]]
	-  [[Principles for Innovation Orgs]]* This may want to be an appendix 

	- It is important to get people on board not necessarily with the constraints of existing systems but with the constraints that ICT would be heading into - specifically, the nature of profit around research and the nature of fairness around research. 
		- This may want to be a series of separate articles about the nature of the problems. However, it also directly affects how to think about profit. 

- [[Organizations with structures worth learning from]]. Academic papers traditionally do a review of prior work. Since we’re talking about institutions, it’s important to do an institutional review instead of a literature review. It shows that I’ve done my research and gives people frameworks to think about innovation organizations. At the same time I want to avoid couching this idea completely in the context of “what about other organizations?” It’s worthwhile to pay attention to whenever an organization is particularly worth paying attention to it and hyperlinking to the part 
	- [[Venture Research, Bell Labs, and DARPA form non-orthogonal basis vectors over the space of innovation organizations]] ::needs work::
	- [[Comparing and contrasting Bell labs, DARPA, and Venture Research]]
	- Non-academic Research Organizations
		- [[Ink and Switch]]
		- [[OpenAI]]
		- [[Willow Garage]]
	- Corporate R&D Labs
		- [[DeepMind combines bottom-up research work with top-down strike teams]]
		- [[Xerox PARC]]
		- [[Bell Labs]]
		- [[Google X]]
	- Funding Orgs
		- [[Prime Coalition]]
	- Startup Studios
	- ::separate:: Why did Bell Labs Work
	- DARPA obviously
	- ::The full list of these should be an appendix::
		-

- Definitions and core ideas. These are probably worth defining lightly in the piece itself and heavily elsewhere. 
	- Reaper Functions
		- [[Reaper functions beat culture decks]]
		- [[Reaper functions dominate other]]
	- Money factories
		- [[Innovation orgs need a money factory]]
		- [[Innovation orgs need to be aligned with their money factory]]
			- [[Alignment requires existential threats]]
	- Innovation organizations ::needs work::
	- Research
		- [[As a platonic ideal, research is an unscheduled quest for new knowledge and inventions with hard-to-predict outcomes]]
	- Paradigm Shifting Work
		- [[Changing paradigms is hard]]
		- [[DARPA works on programs that go against the established paradigm]]
		- [[We have no flying cars - failures of paradigm-shifting engineering]] ::needs work::
	- Organizational Structures
		-  [[Nobody has any idea what the affordances of human systems are]]
		- 	[[There are a number of experiments that seem like they can only be done in the context of an organization]]
		- [[Running multiple organizational experiments flies in the face of common wisdom]]
	- Innovation Ecosystems
		- [[From the outside the innovation ecosystem looks complete and healthy]]

- Need to establish the **demons** up front - these are the inevitable tradeoffs or intractable places where you will always be able to accuse inconsistency

[[Institutional structures can be analogized to anatomy]]
- Skeleton 
	- 
	- [[It is worthwhile and possible to riff on ARPA]] ::This probably wants to be early as an assertion and then filled in throughout::

	- [[There are DARPA-style ideas that DARPA doesn’t pursue]]
	- [[A private ARPA requires new legal organization structures]]
	-  [[Can the ARPA model exist outside of the government?]] (yes)
	- [[21st century riffs on ARPA should be private]]
	-  [[PARPA should create a replicable institutional model]] 
	-  [[PARPA needs to be obsessed with PMs]]
	- [[A private DARPA riff both can and must do multiple experiments at once]]
	- [[Inverting the normal relationship between an R+D organization and is money factory may enable new classes of activities]]
	- 	Goal: enable more awesome sci fi shit
		- [[Arpa-riff Mission]]
	- Long-term organization that sprouts others
		- Think coral reef or mushroom instead of octopus or slime mold
		- [[PARPA will eventually spin out companies]]
	- Program focused
	- PM centric
	- Trust
	- Important to emphasize what it’s *not*
	- 	- - [[ARPA Model Conclusions]]
		-  [[Anybody attempting to emulate DARPA needs to be willing to sit with a lot of discomfort for a long time]]
		- [[You need to minimize actions that lead to short term glory if you want to emulate DARPA]]
		- [[A private ARPA requires new legal organization structures]]
		- [[It is hard for a private ARPA to capture value without hamstringing impact]]

- Muscles
	- Externalized research
		- [[R&D orgs lie on a spectrum between externalized and internalized execution and benefits]]
	- Hypothesis 
		-  [[You can create a program design discipline that enables better research and development]]::needs splitting::
	- Funding sources
		- [[Could DARPA have a positive return?]]
		- [[How does money work in PARPA?]]
		- [[Is there a minimum effective budget to generate outlier results?]]
	- Structure
		-  [[Focused Program Organization]]::needs splitting::
		- [[An ARPA riff will need at least seven programs to demonstrate viability]]
	
- Skin
	- Focuses
	- Core theses 
	-  [[Employee-ownership could enable PARPA to align long-term incentives]]
	- [[PARPA needs a constitution with checks and balances]]
	-   [[A 1B dollar PARPA budget could come from a 25B dollar endowment at 4 percent interest]]
	-  [[Quantized Heilmeier Catechism]] 
	-  [[An ARPA-Riff could establish sales channels for frontier tech]]
	- [[Workshops could be a way to start small building a private ARPA]]
	- [[Is there room for an expanded TRL framework?]]
	- [[Good Simulations could be the ‘why now’ of roadmapping tools]]
- Clothes 
	- Specific programs

- [[Reasons why an arpa riff would succeed or fail]]

### References
- [[Financing Advanced Manufacturing: Why VCs Aren’t the Answer]]



[Web URL for this note](http://notes.benjaminreinhardt.com/An_Institute_for_Creative_Technology_-_Riffing_on_ARPA)

[Comment on this note](http://via.hypothes.is/http://notes.benjaminreinhardt.com/An_Institute_for_Creative_Technology_-_Riffing_on_ARPA)




Fuse Program

Organizations with structures worth learning from

DARPA Program managers have a tenure of four to five years

Forecasting Science & Technology (ForeST)

Forecasting Counterfactuals in Uncontrolled Settings (FOCUS)

Seth Goldstein

A private DARPA riff both can and must do multiple experiments at once

ACE Program

DARPA falls into many classic government organization traps

Jason Matheny

# IARPA
Started in 2006
Reports into the Office of the Director of National intelligence
Structure explicitly based on [[DARPA does multiple levels of top-down problem generation and bottom-up solution generation]]
### IARPA Program Managers
Explcit Role
* identifying goals and objectives
* developing rigorous test and evaluation criteria
* specifying short- and long-term milestones to gauge the health of the program
* managing all technical, procurement, and financial aspects of the program
Serve for 3-5 Years
IARPA issues both RFIs [Example RFI](https://www.iarpa.gov/index.php/working-with-iarpa/requests-for-information/low-cost-mirrors-for-interferometers) and Solicitations for Research
IARPA issues prize challenges [Prize Challenges](https://www.iarpa.gov/index.php/working-with-iarpa/prize-challenges) - looks like purse is $50,000
Iarpa seems to have its shit together way more than DARPA
[[What are the incentives to fill out RFIs?]]
[[How can I get access to previous RFIs?]]
## People
As of Jan 2020:
Director is [[Catherine Marsh]]
Deputy Director for Research is [[Catherine Cotell]]
Deputy Director for Operations is [[Cheri Benedict]]
Chief of Technology Transition is [[Marianne Kramer]]
Hee hee they have a program called [“Amon Hen”](https://www.iarpa.gov/index.php/research-programs/amon-hen)
They are all about the LoTR references:  [SILMARILS](https://www.iarpa.gov/index.php/research-programs/silmarils) 
[MORGOTH’S CROWN Challenge](https://www.iarpa.gov/index.php/working-with-iarpa/prize-challenges/966-morgoth-s-crown-challenge-4)
Every program page has a Program information identifier 
## Questions
[[What are IARPA Seedlings?]]
## References
[About IARPA](https://www.iarpa.gov/index.php/about-iarpa)
           

[Web URL for this note](http://notes.benjaminreinhardt.com/IARPA)

[Comment on this note](http://via.hypothes.is/http://notes.benjaminreinhardt.com/IARPA)




Prime Coalition

Build a map of how innovation organizations are structured now and in the past

Build a map of how innovations are funded now and in the past

ICT Distillation

Transition from exterior of the maze to the layout and why of the ARPA Model

An Institute for Creative Technology - Riffing on ARPA

ICT Outline

# Organizations with structures worth learning from
## Standard questions 

### Research Orgs
* [[SRI International]]
* [[Willow Garage]]
* [[Otherlab]]
* [[DeepMind]]
	* How did they get started 
* [[OpenAI]]
* [[Intellectual Ventures]]
* [[Xerox PARC]]
* [[Google X]]
* [[Bell Labs]]
* [[Lockheed Skunkworks]]
* [[D.E. Shaw Research]]
* [[Creavis]]
* [[Dynamicland]]
* [[Allen AI Institute]]
* [[Ink and Switch]]
* [[Microsoft Research]]
* [[Santa Fe Institute]]
* [[YC Research]] 
	* Why did they shut down?
* [[Media Lab]]
* [[MRC Laboratory of Molecular Biology]]
* [[Flatiron Institute]]
* [[Actuate]]
* [[Activate]]/[[Cyclotron Road]]
* [[Rad Lab]]

### Funding Orgs
* [[DARPA and corporate R&D filled overlapping but different niches]]
* [[ARPA-E]]
* [[IARPA]]
* [[X-Prize]]
* [[Research Bridge Partners]]
* [[Gates Foundation]]
* [[Howard Hughes Medical Institute (HHMI)]]
* [[BP Venture Research]]
* [[Prime Coalition]]

### Startup Studios
* [[3rd Rock Ventures]]
* [[Flagship Pioneering]]
* [[Idealab]]
* [[Deep Science Ventures]]
* [[Science (Startup Studio]]
* [[Pioneer Square Labs]]
* [[Singularity University]]
* [[Entrepreneur First]]
* [[YCombinator]]
* [[SRI Ventures]]
* [[TandemLaunch]]

### Whitespace Research
* [[Lux Research]]
* [[JASON group]]
* [[Polyplexus]]

### Pioneers of new models
* [[ARDC]]
* [[Sequoia]]
* [[Venrock]]

### Random
- [[Research Seminars dot Org]]


[Web URL for this note](http://notes.benjaminreinhardt.com/Organizations_with_structures_worth_learning_from)

[Comment on this note](http://via.hypothes.is/http://notes.benjaminreinhardt.com/Organizations_with_structures_worth_learning_from)




# Transition from exterior of the maze to the layout and why of the ARPA Model
While many people can inspect the exterior of the maze and the stories are pretty precise on the relevant features of the minotaur (bull head, human body, kills (and probably eats) people, superhuman strength, etc.)  the layout of the labyrinth could conceivably be in any configuration that conforms to the exterior walls. 

When faced with [a situation where you cannot reason from first principles]<[[There are no first principles for legal structures]]> and slow iteration loops there are are limited options for thinking rigorously about it. Neither scientific inquiry nor engineering design can function so we have to look at historical narrative.

Imagine we’re in some parallel universe where Daedalus was a prolific labyrinth builder and while nobody had ever gone through the modern Labyrinth we’ve described, there were others who had successfully navigated other labyrinths. The best strategy is probably to learn as much as we can from them and try to adopt their tactics to our own situation. In our case, there is a pantheon we could dig into[^1] but the two mythical organizations that stand out the farthest are Bell Labs and DARPA. Of these two, I would argue that DARPA’s labyrinth looks much more like the one we face today. 

[^1]: There are obviously many [[Organizations with structures worth learning from]]  - and a deep discussions of what to learn from each of them is important but beyond the scope here.




Ben's working notes

Organizations with structures worth learning from

Standard questions

Research Orgs

Funding Orgs

Startup Studios

Whitespace Research

Pioneers of new models

Random

Links to this note