Monday, July 18, 2011

The Challenge of Not Dying

I have made a list of classes of things that can kill a person.  In our technological future it would be nice if these things were kept to a minimum.
  1. Temperature Extremes
  2. Pressure Extremes
  3. Acceleration Extremes
  4. Radiation Exposure
  5. Impacts - Penetrating, Crushing, Shearing
  6. Electrocution
  7. Chemical Exposure - Acids / Alkalis, Solvents
  8. Respiratory Failure - Hypoxia / Drowning, Poisons
  9. Starvation - Improper Food, Vitamins, Micro Nutrients, Poisons
  10. Pathogens - Bacteria, Viruses

Replicator Technology should allow us to create objects needed to guard against these dangers and should not accidentally introduce any of these things into our environment.


On April 14, 1912, a clear, moonless night in the north Atlantic ocean with unusually calm seas, impact with an iceberg fractured hull plates and rivets made brittle by the frigid water leading to the loss of the RMS Titanic and the deaths of 1,517 people.


On May 24, 1993, Joe Bill Dryden, one of the most knowledgeable and experienced F-16 test pilots in the world took a brand new, properly performing, aircraft on a clear day over level terrain and performed a Split-S maneuver into the ground. He ejected safely, but his parachute descended through the resulting fireball and he was killed by the impact.


On January 28, 1986, hot gasses leaking past a redundant pair of frozen o-rings on the right solid rocket booster weakened the supporting structure and ruptured the external tank of the space shuttle Challenger seventy-three seconds after liftoff. The $1.2 billion spacecraft and its payload were destroyed and all seven astronauts were killed.


On February 23, 2008, at Andersen Air Force base in Guam, water intrusion into three of twenty four skin-flush air-data sensors caused faulty air speed and attitude information to be sent to the flight control computers of the B2 stealth bomber Spirit of Kansas, leading to a crash immediately after takeoff. The two pilots ejected safely, but the $1.4 billion aircraft was destroyed.


On March 27, 1977, at Los Rodeos Airport on Tenerife in the Canary Islands, the KLM Boeing 747 Rijn piloted by Captain Jacob Veldhuyzen van Zanten, head of KLM’s flight training department, began takeoff without proper clearance and collided with the Pan Am Boeing 747 Clipper Victor taxiing in the opposite direction on the runway. Both aircraft were destroyed and 583 people were killed.

I have used these boxes to describe disasters caused by reality reaching out and grabbing the unwary. The world is not a safe place. Mortality or extinction may be closer than even the most paranoid realize. It is also true that when people are very good at what they do, and are very careful, it is possible to achieve virtually unimaginable things.


I once visited the McDonald farmhouse at the Trinity Test Site near Alamogordo, New Mexico. This is an uninsulated wooden, pier-and-beam house typical of the early twentieth century. I could look down through the cracks between the floor boards and see the dirt below.

In this house, in 1945, the absolute top scientists of their day, given the essentially unlimited resources of the Manhattan Project, gathered to hand-assemble the world’s first atomic bomb.

The Challenge of Pattern Design

Patterns must describe the desired object in a form that can be implemented by the Replicator. 

For a start, let us think of a Replicator Pattern as a kind of recipe.  For reference I have included a particularly good recipe.  Any cook knows that you (or your mother) must have read The Joy of Cooking before you are actually allowed in the kitchen.  Thus, you have the background knowledge to be able to measure flour (don’t pack it down), and what it means to “cream the butter”.

I have modified the presentation from a normal cookbook format to include the required equipment.  If you try making the cookies yourself, see how many little things I have left out.  Experiment.  Try baking cookies on a clear, cold day and again on a hot, rainy one.  Do all the cookies come out nicely brown?  What do you “get to know” about your oven.  Do you really leave them in the oven for 10 minutes, zero seconds?  Or do you peek and take them out when they look and smell nice. 

How do you get the cookies off the cookie sheet?  When?  Are we going to need a cookie jar?


Chocolate Chip Cookie Recipe

Equipment:
    blender
    grater
    mixing bowl
    measuring spoons
    measuring cup
    spatula
    cookie sheet
    oven
Ingredients:
    2 cups    butter
    4 cups    flour
    2 tsp.    baking soda
    2 cups    granulated sugar
    2 cups    brown sugar
    5 cups    blended oatmeal (measure oatmeal and blend in blender to a fine powder)
    24 oz.    chocolate chips
    1 tsp.    salt
    1 8 oz.    Hershey bar (grated)
    4    eggs
    2 tsp.    baking powder
    3 cups    chopped nuts (your choice)
    2 tsp.    vanilla

Cream the butter and both sugars. Add eggs and vanilla; mix together with flour, oatmeal, salt, baking powder, and soda. Add chocolate chips, Hershey bar and nuts. Roll into balls and place two inches apart on a cookie sheet.

Bake for 10 minutes at 375 degrees.

Makes 112 cookies.

Now we have a batch of cookies.  And a terrible mess.  How do we know what to do with the leftover ingredients and the dirty dishes?  Not even The Joy of Cooking is any help here.  We have common knowledge and some vague instructions from the dishwasher’s Operator Guide.

I believe that a Replicator Pattern must restore the system to exactly the initial state.  And it should have provisions for cleaning up and recycling the inevitable accidents - like the egg that got dropped on the floor.
  • How many patterns would be required to recreate the Replicator itself?
  • How many patterns would be required to support a sustainable human population?

  • How many patterns would be required to support a modern technological society?

  • How many patterns would be required to support a colony in space or on Mars?
I expect the Replicator to be able to internally derive processing steps and self-optimize operations without explicit direction.

Patterns must include tolerances and test parameters to be used during production to detect flaws.

Patterns must describe deconstruction steps to be used during recycling.  This must work with incomplete or damaged objects. 

Deconstruction patterns are also used to handle production waste by-products such as materials and used tools.

Software tools must be available to allow ordinary human beings to create and edit patterns safely and successfully.

Patterns must be standardized in such a way that they can be transmitted from one Replicator to another.

Reducing Space Station Down Mass

Consider the resupply situation for the International Space Station.  Every few months or so for the past ten years supplies have been sent to the space station aboard the space shuttle or Russian Progress cargo ships at a cost of approximately $100 per pound. 

Used equipment, packing materials, and the perfectly good, working cargo ship are then discarded to burn up in the atmosphere.  All because there is no safe way to store or recycle the materials, some of which could become extremely hazardous in the confined volume of the station.  Also, the lack of maintenance and limited lifetime of certain components (especially seals, pyrotechnics and the fuel systems) mean that the cargo ships would be unsafe. 

Careful rethinking and redesign with the goal of eliminating all possible “down-mass” operations would make inhabiting space much more cost effective.  Designing equipment to be disassembled into reusable material or components inside the station and designing the vehicles themselves for permanent attachment to the growing station would provide much needed additional space as well as shielding and safety from orbital impacts. 

If cleverly done, this could provide the basic research into techniques for building really large structures in orbit.  Questions about orbital boost and decay, dynamic stability, structural strength, modes of oscillation, etc., could all be studied.  But not if the design requires that we continue to regularly discard tons of mass that we already paid to put into orbit.