Patri / Vince meeting, Puerto Rico, Feb 20th, 2009
Discussion of TSI’s engineering strategy, key points. (wish I’d written this up sooner, my notes are a bit cryptic)
Vince’s version of his key points is on this wiki page. Discussion on this forum post. Wayne & I have a number of comments here on Vince’s ideas, as you will see. Note that since Wayne & I have gotten to add comments on Vince’s points, but Vince has not gotten to add replies, the discussion is somewhat one-sided, you should assume that in many cases there are good answers to our questions/concerns :).
How to do engineering
Our current deal with MI&T is a bad incentive system. Hourly payments to a consultant incent them to put in more hours than we want. They profit most by dragging things out / doing too much detail work.
P: Not saying they are doing so deliberately, but it will always “feel” right to them to go to a higher level of detail than we would prefer, because of the incentives. W: I think a milestone based deal with fixed prices would go a long way towards improving incentives.
Prizes are a good incentive system. You only pay when someone gives you what you want! Often far more effort is put in, than the cost of the prize.
W: Prizes can be a good system, but they are far from a panecea, they tend to cause needless replication of effort
Having an employee is a pretty good system too. They will tend to see things more from the point of view of the organization, and act in its best interests, than a consultant. “Employee = One Of Us. Consultant = More Hours Feeds My Kids”.
Small floating structures are very safe – they move up and down with the waves because they have a high waterline area (also uncomfortable). Worst wave problems are phase problems – bad things happen when waves crash on your structure because it is moving out of phase with the waves. This happens when you have a low waterline area. This gets you more stability, but also more danger.
W: This is an over simplification on Vince’s part. In short, many of the naval structures are bi-stable. They can float “up right” or “upside down”. If your small structure hits a big wave, it can flip. The flipped structure can be quite unsafe. By the way, when Vince hangs a big kite on his WaterWalker, he is inviting a flip.
Wave tank test is the gold standard for offshore engineering. Ocean is a big wave tank. The scaling laws are pretty simple: http://wiki.seasteading.org/index.php/Scale_models. So “amateurs” who build and test scale models can actually give as reliable results as professionals. Hydrodynamic modeling software is less accurate than tank tests! Vince may not be a professional, but in some sense he has higher-quality test results. (See wiki page on making our own wave tanks, mainly written by Wayne)
P: There are square/cube issues – structure movement may scale easily, but structural strength doesn’t. Also, wavelength matters, not just height. Vince is only able to test at a single wave spectrum, but you care about performance for the worst wavelength. Now, for models which ride the waves, there shouldn’t be much of a resonance effect, so this may not affect his type of design. But it would matter for larger structures like ClubStead.
W: You can control input, center of mass, and take measurements. It is just that Vince has not done so in any way that I can figure out. I do not want to completely dismiss what he has done, but I also do not put a super high value on it either. It is what it is; a relatively uncontrolled simulation.
I think we can do better if we work on it. My current thoughts are to use a pond instead of the ocean. That gives us relatively calm water. An inexpensive wave generator gives controlled waves (both height an wave length.) A stereo video can be taken with splitting mirrors. By painting the structure one color and placing three other colors on it, we can reverse compute the structure kinematics..
There are no seastead engineering experts (except the waves 🙂 ). It lies somewhere in between civil engineering, naval architecture, and offshore platform engineering. If you ask a specialist in any of those disciplines, they will build you something in their area. If we had asked a naval architect, they would have made us a ship. We asked oil platform engineers, so they made us an oil platform. Given that there are no seastead engineers, we aren’t comparing amateurs vs. professionals. We are comparing amateurs vs. “professionals in a related field”.
Given that we have wave testing available, it is fine to use amateurs to generate ideas. The waves will show us that the bad ideas are bad. We can use professionals to test material strength before building a full-size seastead. With a good evaluation process, it is OK if we get crazy ideas. There is nothing unsafe about a seastead designed by amateurs, if tested by professionals and gold-standard techniques like wave tanks. Yes, the ocean is deadly, but we only need experts at the end of the engineering process.
W: I am totally in favor of evaluating crazy ideas. Indeed an idea that starts out crazy could incrementally be improved to not be so crazy.
Secrecy is bad for the community. Inspiring people to think about and make their own designs has value even if the designs are bad. We want to encourage community amateur engineers, not shoot them down. Until there is engineering, there can be no politics.
W: Indeed. That is why I want to foster a cooperative community of engineers. I think we should figure out some way for people to run simulations using of the standard naval architect CFD codes.
We need to learn some of the engineering basics (or hire in-house knowledge) “If you want to make seasteading a reality, you need to understand the engineering, not just defer to experts”. It isn’t rocket science, it is something we can learn.
ClubStead Problems
Vince believes worst case for ClubStead is 45deg wave heading, so that when the crest is under opposite corners, you have 2 corners hanging on the other two.
It has tiling problems. The wall height of the platform is less than the vertical travel. The walls aren’t big enough. This means you can’t just put out bumpers and led them ride up and down (which is what boats do, as they have giant walls)
Patri: Is this true? The relative vertical travel of adjacent platforms should be fairly low, as both are in similar waves and moving similarly. If they were 180deg out of phase, then yeah, it wouldn’t work, but with a platform length of 200 feet and typical wavelengths of thousands of feet, adjacent platforms will be very close to in phase.
W: Yup. ClubSteads are going to be hard to tie together. I feel the issue is kind of a non-issue, since I do not think even one is going to get built any time soon.
Vince believes ClubStead’s wave motion is worse than his Water Walker, based on his wave tests and the reports numbers.
P: I have trouble believing this, given that CS moves less than the wave, and WW moves with the wave, but Vince is a smart guy so we should look into this.
W: I think Vince is dead wrong on this one, but until we push Water Walker through the same simulation codes as ClubStead, we will not know.
SFS:
Individual SFS are small, not seen as competing with a local industry, and less likely to get shut down. Also, can travel to many jurisdictions (ie countries in the Caribbean), not be a fixed target for any one of them to notice and complain about. (Same goes for a condo cruise ship).
We don’t need a skyscraper early. Even if the goal is a city, start with a village. Single Family Seasteads – especially if they tile – are good enough for the beginning. Once we have the village, we can try to fund the city.
There are lots of people who go live on a boat for a year, including families.
W: No arguments about SFS. You know my Venice strategy.
Cruise ship industry is big and efficient. Underwater robotic welders. Hard to compete with. Big, safe, stable. Whereas small boats are not safe and stable right now. Not meant to anchor outside a harbor. Easier to beat. (P: But ship-building is a big industry too.)
Prizes
We should compare other engineering strategies to: “Just define exactly what we hope to achieve and offer a $500K prize for achieving it.” Vince is serious about this as a strategy.
Patri: I see value in having an organization which grows community, does additional fundraising etc, but I definitely like this as a strawman for comparison purposes. And if you just replace $500K with $250K, it’s even a reasonable strategy from my perspective.
Success breeds success. Look at the X Prize – they awarded one, and it made it easy to raise funds for more.
W: Did you ever hear of the Cheap Access to Space prize? Well, nobody ever collected on it. Just offering prize does not make people suddenly decide they are going to go for it.
A prize incents the goal directly. And you don’t pay unless you get what you want! With contractors, you may pay and not get what you want.
W: It depends upon how you write the contract. I think we need to be smarter about writing our contracts.
Suggests an Ephemerisle contest for real seasteads – head out to the big waves, see how they handle. Needs to have a launch that can tow it (and rescue people if structure sinks).
Misc
Wiki should be on “Key Areas”.
Money will go further if we partner than if we try to do things ourselves. Example: Vince wants to build a full-size Water Walker, rent it out as an Anguilla vacation home. TSI could help fund it, and make money off the rent. Getting to use someone else’s capital and labor along with our own.
You can’t orient yourself to avoid a rogue wave. They can just pop up right underneath your structure.
Having access to design tools is certainly highly desirable and a seemingly likely effort multiplier. Unfortunately having access to the software simulation tools is only one part of the puzzle. Apparently actually using them correctly involves a great deal of art in addition to the science.
For example designing a building or bridge to be structurally stable and safe is challenging under static loads. Adding wind loading makes it harder. Adding wave loading makes it significantly harder still. And the way waves behave internally and interact with a structure is a non-trivial field. It may not be rocket science, but there’s apparenly a lot of specialized knowledge involved. It’s not like a car design where things are so well-understood and tools are so well-developed that someone who’s relatively ignorant can plug in the desired parameters and a suspension design pops out of the software without much effort. (That said, the results of underinformed use of such sofware isn’t always great either. GIGO still applies.)
That all said, improving the design and simulation tools for marine structures seems like a great way to leverage the progress in software development, the network effect, etc. Software development is something the world has gotten better at, though perhaps really complex, expensive engineering problems aren’t always the first problems solved using open source software, etc. There’s so much value to add and cost involved that it’s probably something major indusries develop for themselves first, at great expense. But I’m definitely in favor of gaining better access to and knowledge of the design tools.
To back up Wayne’s comments, motion of the stead in waves is only one factor. Structural strength is another major factor, and perhaps the most important one. A structure that’s uncomfortable in waves is bad. A structure that’s not strong enough to survive in them is much worse. The relationship between waves and structural strength is probably very complex.