TSI Engineering Strategy & Research Agenda

[Patri Friedman]

Hey all!  We have found a Director of Engineering, George Petrie, who will be working part-time & long-term on TSI's research agenda.  (We'll put out a press release later this month).  We'll be meeting w/ George next week to map out & kick off his research, so if you have time, I'd love any feedback you have this week on TSI's engineering strategy, key open questions, and research agenda.

Here's the link, or see the doc below:

-----

TSI Engineering Strategy

Current as of Q3 2010.  See also: TSI Strategy

Goals

TSI's mission is to further the establishment and growth of permanent, autonomous ocean communities, enabling innovation with new political and social systems.  By opening a new frontier, we intend to revolutionize humanity's capacity to improve quality-of-life worldwide by creating experimentation and competition among governments.

Because the ocean is a harsh and unusual environment, engineering is one of the biggest challenges facing seasteading.  While ships & oil rigs demonstrate some of the possibilities, they have significant deficiencies for our purposes.  Even small seasteads will require lower cost, less maintenance, and dealing with more severe weather, and large seastead agglomerations will require solving brand new problems in modularity and anchoring.

Seastead Engeering Challenges

Some of the unique challenges are:

1. Engineering cost.  Offshore platforms are a tens of billions/year industry, individual platforms can cost over $1B.  Seasteads must not only eventually be much cheaper per unit area or topside load, they much be designed with far less engineering resources and expense.  
2. Organizational budget.  TSI’s engineering budget is relatively tiny, so we must use it very effectively through techniques like using volunteers, low hourly-rate students, prizes, and obtaining grants.
3. Scale.  The problem of seastead engineering is actually a set of problems ranging from how best to let 1 family live comfortably at sea up to how to create a floating city of a million people, which will likely have very different answers.  While our most urgent need is to solve the near-term problems for small seasteads, it is also important for the long-term viability of seasteading to show that we have good potential solutions to the problems of a large city, and ways to bring costs down over time.
4. Modularity.  It is important to TSI’s mission that floating cities be modular: that there is some unit size (whether a single apartment or a whole city block) that can be physically rearranged, both to change the internal structure of the city, or to move a module to a different city.  This will be required for scaling as well - we will need to build up our city incrementally, there is no way we can get the funding or population to bring even 10,000 people at once.
5. Ocean Environment - Waves.  One of the largest challenges is waves, a “continuous earthquake”.  Waves are difficult because:
1. They have lots of energy - especially big waves.
2. Unless your structure is mobile (difficult when large), it must be engineered for the worst-case waves (ie rogue wave in 300-year storm), which means most of the time it is wildly overengineered for its actual environment, with lots of capital locked up in concrete or steel for the occasional big storm.
3. They come in as a spectrum of periods/frequencies which change over time - you are never dealing with a single period, which makes it difficult to tune structures since they must perform well for a range of periods.
6. Ocean Environment - Dynamism.  The fluid nature of the ocean makes movement easy. This can be beneficial (ie by enabling rearrangability), but it means that holding position is non-trivial.  It will be especially difficult for large agglomerations, where (for example) one can no longer assume wave transparency, must somehow parallelize anchoring, etc.  It may create location dependence (seamounts for anchoring, which drastically cut down on possible locations).
7. Ocean Environment - Corrosion.  The corrosive saltwater generally means short lifetimes and high maintenance costs.  Ships are drydocked regularly, while steel platforms corrode over a few decades.  Seasteads aim to be low cost, long-term, and independent of drydocks.

Potential Areas of Research

There are a variety of options to consider, which can be sorted a few ways:

• General
• Overview (ie Eelco’s work)
• Criteria (ie comfort, how may days, wave criteria)
• By Scale
• Single Family Seastead
• Poseidon - What is the best structure for 50-100 people (our currently-anticipated first seastead)?  Are we right to focus on ships?
• Poseidon+ - How can Poseidon scale to 1,000?  (For ships - multiple or one big)
• 10,000 - What would be the best way to build at this size?
• 100,000 - What would be the best way to build at this size?
• What does the cost/size curve look like?  Relatedly, what’s the structure/size curve - what designs work at what size ranges?
• By Structure
• Ships/barges - What wave conditions do they require?  How can we modify them for stability (comfort)?  There is good anti-roll tech, but what about anti-heave?  How can they be made modular?  What innovative subtypes might help us (SWATH, etc.)?  How can they be made cheaper?
• Platforms (spars, semisubs) - How do they compare to ships?  In what conditions/at what sizes are they superior?  How can they be made cheaper?  Modularity (ie wave blanket)?
• Breakwaters - Do they work?  Do they require anchoring?  What are their cost characteristics, scaling laws?  At what size are they superior?  Can they include energy generation, and at what marginal cost per unit energy?
• What other structures should we be considering?  What has been considered/tried for VLFS - Mega-Float, Mobile Offshore Base, PSP, etc.
• By Challenge
• Understanding scale - where are the phase transitions, ie between ships & platforms, or non-breakwaters & breakwaters?  This will likely focus on the low end (1 - 100,000 people), as it seems likely that a tech that can scale to 100,000 can then scale arbitrarily up from there.
• Wave Mitigation
• Location Control (anchoring and mooring, dynamic positioning, etc.)  What tech is available?  How does it scale?  What is cheapest?  At what depths?
• Corrosion (ie can concrete eliminate need for drydocking and extend structure lifespan?  Special paints/coatings for steel?)
• Motion Characteristics - What are the requirements for comfort, how does adaptation affect them (how rapid is it, what %age adapt, what are the adapted requirements).
• Modularity - How can seasteads be connected?  What are the motion implications?  What about transfer of people & goods between modules?  In what seasteads can transfer take place?

How To Do Research

Research must be done cost-efficiently, ie students guided by faculty.  Ideally it will be “open-source”, with data (such as hydrodyamic models) made available so that we build up a body of knowledge over time.  TSI will generally strive to do work that would not be suitable for IP protection, although if it stumbles across a commercially valuable technology, particularly for non-seasteading uses, it may patent and license the tech in order to fund its mission and conduct further research.

TSI is a 501(c)3 non-profit, which imposes certain limitations on its activities, especially since seasteading technology will likely be commercialized and implemented by for-profit firms.  TSI should focus on creating public benefit through long-term, general research which it makes freely available to the community and the public.  It cannot do research which directly & immediately benefits for-profit firms (especially a specific for-profit firm) unless it charges those firms a market rate in an arms-length transaction.  So for example, while it is appropriate for TSI to research the general motion characteristics of ships and determine whether to recommend ships as a short-term seasteading solution to the community (“ShipSteading”), creating detailed blueprints for the first ShipStead which could be taken to a shipyard for construction would be inappropriate, as that would primarily benefit the first company which created the first ShipStead.  TSI could sell or license the blueprints to such a company, but it would be better to let the company create them in the first place.

Work To Date

See the TSI Research Page (Engineering section) and Engineering Blog.

Goals For The Next Year

• Knowledge
• Deepen our understanding of the space of seastead designs & challenges by continuing Eelco’s research agenda and completing / polishing his overview.  Understand key questions where further research is needed, outline research agenda.  Publish it as the first seastead engineering whitepaper.
• Know the best structure for near-term seasteading, probably ships, and its performance characteristics (in likely locations), cost characteristics (in likely sizes), and location requirements (current, waves, etc.)  Include as recommendation in whitepaper.
• Stretch: Begin to explore some of the longer-term scaling issues & solutions (ie get a feel for whether breakwaters are feasible)
• Community
• Begin to build a global team of interested researchers - students, faculty, and individuals, who are strategically coordinated (working on the right things).
• Hold the first seastead engineering summit, possibly as a more general VLFS workshop, drawing together top VLFS researchers as well as seastead enthusiasts.  Ideally be financially independent through sponsorships or grants.
• Establish a team and reputation that give TSI a solid chance to get grants.

Goals For The Next 5 Years

• Knowledge
• Substantial progress an all key questions - understand scaling issues, wave mitigation, location control, cost reduction.
• Near-term technologies deployed in-use by seasteading communities.
• Long-term technologies understood and ready for detailed design / prototyping when community size justifies.  Perhaps small prototypes or wave studies already.
• Community
• Have a global team of researchers, mostly with their own funding.
• Regular engineering summits which are well-known in the Ocean Engineering world (perhaps as a seastead/VLFS track at OMAE?)
• Good relationships with public & private grantmakers which substantially supplement TSI’s funding.

Next Steps (Q3 2010)

• Review & Revise TSI Engineering Strategy, high-level plan for research program, both “Knowledge” and “Community”
• Knowledge - Specific Research Projects (will be based on strategy, here are some guesses)
• Definitely: Start Location study (help recruit/supervise oceanographer/boundaries researcher)
• Probably: Help Eelco complete/revise/publish Seastead Engineering Overview
• Probably: ShipStead research w/ Miguel: Begin work on performance & cost characteristics & wave requirements for various sized ships, and how to improve them w/ the stabilization technologies Miguel has studied
• Community - (will be based on strategy, here are some guesses)
• Create proposal for DARPA grant.
• Begin scaling global seasteading engineering community, using George’s contacts, our list of engineers interested in seasteading, students, etc.  Crowdsourcing plan, central infrastructure (mailing list?  journal?  repository of hydrodynamic models?)
• Stretch: Start thinking about first eng summit/VLFS workshop