Roadmap

From Biohack

Jump to: navigation, search

http://heybryan.org/

Contents

[edit]

Introduction and Reorganization


This roadmap is distinctively type-1 transhumanist; however, type-2 projects (such as the World Memory Project or Humanity Sequencing Project and Abundance4All) do play a role, however indirect. Type-1 transhumanism refers to personal futurism, the transformation of the self with the assistance of technologies, while type-2 transhumanism is more like "techno-humanism". That having been said, this technology roadmap is somewhat like a gateway to the future, listing current and possible technologies to get there.


[edit]

The Two Themes of the Roadmap

Clicking around through the wiki, one finds that many of the topics are conceptually connected in the sense of the two themes: first, there's transhumanist self-modification, for the enhancement of the brain, of the body, of the genome, etc. And then there's self-replication. What changes would you want to make given a brain implant? Randomly deleting a receptor site is a bad idea. You have two options:

Both solutions require lots of hardware. The amount of materials required for a neurofarm or a supercomputer capable of doing enough simulations is, in essence, infeasible unless we implement self-replicating machines to gather resources and build the hardware necessary.

[edit]

Self-replication

Note: the roadmap towards self-replication here focuses heavily on the transhumanist approach to "justifying" it. There is no need for justification: it is an interesting project on its own. However, it does have a context in the world.


There has been the interesting suggestion of doing in vitro synthetic circuits to show an understanding of in vivo evolved biological systems; it may be possible to leverage evolution to come up with biology, which is essentially software, and then encode this into our abstract knowledge bases. Perhaps yet another illustration of the value of self-replication.

[edit]

The necessity of self-replication

Note that there are many reasons that self-replication is an interesting topic and a useful feature. Perhaps the simplest reason to make self-replicating machines is because it allows as much as possible to occur. However, more connected to the human story of life, it can be used for personal improvement (for the more philosophically inclined, this was a suggested solution to the sloth response to Leibnizian optimism, there's a link somewhere in the bookmarks).

[edit]

Evidence that self-replication is possible

Cells have been doing it for the last four billion years.

[edit]

von Neumann probe projects

[edit]

Self-modification

For example, the brainstate augmentation setup project. It's there to help programmers maintain brainstate during programming sessions.

[edit]

How to make changes

[edit]

What changes can we make?

[edit]

Sandboxing the changes

[edit]

Database projects

[edit]

Nano

  • Molecular nanotechnology (MNT) requires atom holography and lasers for object-specific construction. I haven't been able to come up with better ideas (and I don't like waiting).
[edit]

Carbon nanotubes

CNTs are built either using arc discharge, laser ablation, chemical vapor deposition, and some forms of flame-controlled growth.

[edit]

Nanoparticle synthesis

  • Gold nanoparticles are easy. Too bad they aren't semiconductors.
    • 2008-04-30: An interesting alternative would be graphene (locally: graphene). Then use AFM nanolithography to etch circuits into the graphene. STM/AFM machines can be made for $100 USD. And the graphene can be made from some tape and pencil shavings. Quantum tunneling for graphene transistors has been mentioned in the news ( a success ) backed up in the daily notes on this wiki somewhere (or maybe not?).
[edit]

Diamondoid mechanosynthesis

Diamond mechanosynthesis annotated bibliography
Nanofactory Collaboration Project quote:

The nanofactory is a proposed compact molecular manufacturing system, possibly small enough to sit on a desktop, that could build a diverse selection of large-scale molecularly precise diamondoid products. The principal input to a diamondoid nanofactory is simple hydrocarbon feedstock molecules such as natural gas, propane, or acetylene. Small supplemental amounts of a few other simple molecules containing trace atoms of chemical elements such as oxygen, nitrogen or silicon may also be required. The principal output of the first commercial nanofactory will be macroscale quantities of molecularly precise diamondoid products. These products may include nanocomputers, medical nanorobots, products having diverse aerospace and defense applications, devices for cheap energy production and environmental remediation, and a cornucopia of new and improved consumer products. Earlier-generation research nanofactories will produce substantially less complex products but will provide an evolutionary pathway leading from the first simple DMS workstations to more mature systems. The nanofactory is a molecular manufacturing system employing controlled molecular assembly that will make possible the creation of fundamentally novel products having the intricate complexity currently found only in biological systems, but operating with greater speed, power, reliability, and, most importantly, entirely under human control.
[edit]

Atom holography

Brief summary of atom holography: In the 1920s, Einstein and Bose hypothesized on a new type of matter that they called a condensate (BEC); and soon de Broglie hypothesized the wave/particle duality of both light (photons) and matter.. In the '90s, researchers (Ketterle, etc.) developed the first machines to make a BEC in the lab, and a Nobel prize was awarded for this research. These machines use the laser-cooling technique to bring down matter to ultracold temperatures inside vacuums. The MOT (magneto-optical trap) is used to guide atoms into the center of its chamber and then maybe three (red) lasers shine into the atoms from different directions such that the atoms have no where to move, and in this way they lose energy and eventually become a condensate where they exhibit this sort of 'shared' quantum wave. Over the last decade in Japan, Shimizu has been working on "atom holography" where a matter beam (magnetically controlled BEC) is shot into slits on a plate. These microscopic slits have electrodes that modulate an electric field, causing the matter beam to change shape as it passes through each of the different slits. Shimizu et al. have successfully written words and symbols using this technique, implanting atoms on a surface that are later viewable with electroscopes. Meystre at the University of Arizona has stated that "we could copy objects." Indeed- that and much more, perhaps even bootstrapping MNT. The BEC setups are estimated to cost $300k USD, but this is with all factory-purchased parts. ( I have a zip file of many important papers, and maybe 80 megabytes of images of the setups, the schematics, etc. Also a large bibliography. )

[edit]

Bootstrapping molecular nanotechnology (MNT) from bacteria

See bootstrapping. Basically, the idea is to rewrite bacterial genomes and translate the lifeforms over to MNT such that some sort of bacteria programming language can be used to do a sort of MapReduce/Hadoop for parallel distributed p2p manufacturing operations. Not only does this theoretical language have to be thought of first, but a method of hijacking the genome to generate a synthetic creation is an interesting problem as well. See also the emails to the biobrick mailing lists concerning amorphous computing and amorphous fabrication. This also has important tie-ins with self-replication.

[edit]

Cryonics and cryogenics

  • [cryo.html Cryogenics bibliography] and [cryo.bib BibTeX].
    * Towards the ability to cryogenically store material without crack formation
    ** Temp-curve fitting tech to ensure smooth/proper descent to cold temps
    Insert here cryonics protocols (vitrification, preparation, etc.).

    Preservation of life, reincarnation, immortality
    Fahy, Eugen Leitl, Ben Best

    Alcor Life Extension Foundation
    Cryonics Institute
    American Cryonics Institution
    Cryonics Society of Canada

    CryoNet

    LifeNet project: volunteer network that goes where-ever there are firestations and police stations. The goal is to minimize the amount of time to reach anybody who dies on the continent within 30 minutes and to cryogenically store them (they are already "dead": they might never know). Calculations show that there would need to be at least 150k locations and that there is one death every 14 hours per 50 km^2 average in USA.
[edit]

Nobody should die

Burial is an archaic practice. There is no reason to treat the human body in such a primitive way. Cryogenic storage for everybody: no exceptions. Automated cryonics units can be used to help with the process of death: release design schematics and let community volunteers build these devices for their aging members of the population. There is no excuse for allowing death to occur.

[edit]

Neuro

paper archive (brain implants, MEAs, mindlinks, neurohacking)
previous notes

[edit]

Brain atlas projects and the Blue Gene project

The brain atlas projects profile gene expression across brains via mRNA probing and gene chips (microarrays). For example, the Blue Gene project uses this data plus theories of ion channel simulators to computationally model the brain. This may one day be useful for figuring out how to optimize the human brain, or perhaps eventually doing "brain" better than biology currently does. This is not direct mind uploading. This is sort of like ai, but these projects have wider implications regarding the ability to self-enhance with the selective attention to certain genes and modifying one's own cognitive faculties.

[edit]

Neurocomputation project


Or, for lack of a better name, the following.


On Tuesday 18 March 2008, Andy Ellington wrote:
> Here's a worthy challenge:  while it's relatively straightforward to
> make switches, gates, and even circuits, what should we do with
> them? What can you do with transcriptional switches that would be
> interesting computationally?

That's a very interesting challenge, is this what we'd work on answering
over the summer? I have been playing with an idea that comes close to
answering your challenge, so I want to see if you like it:

That engineering knowledge -- for switches, gates, circuits in vitro
(eventually in vivo) -- needs to come full circle back to the source,
the brain, or at least neural tissue slices in a dish. This is because
the function of neurons is computational in nature*, and now they can
design new computations to make it full circle and complete the circuit
(pun intended). For starters, this would mean recognizing
neurotransmitters that bind to in vitro DNA (eventually in vivo RNA)
and then performing basic logic functions, like calculations, [this is
where it stops being 'for starters'] and spitting out information to
influence the neuronal network 'backwards' (up-propagation / bottom up)
in a way via releasing specific molecules that could trigger other
circuits in neighboring cells (data packets, like on the internet),
since the DNA would theoretically be able to store a 'map' of the local
neural cluster, but this mapping system has not yet been engineered
(keeping track of what neurons down the dendrites send per each signal,
something very basic with minimal overhead, would be a logic circuit
with using DNA as RAM for the map). I have not completely evaluated the
feasability of this approach, and understand it'd be far more than a
summer project.


A first step might be to illustrate data transfer between two cells via either rewriting the signal transduction pathway or making an in vivo RNA logic gate system with some sort of exporting function. And in general, it should not be dependent on neurons, but cells in general; although neurons have an interesting property in that they are physically connected and present a specific engineering project.

[edit]

Notes on subminds

Add notes on subminds plus mice experimentation.

[edit]

Intelligence augmentation & intelligence amplification

  • Nootropics
  • Alternative neurochemistries
  • Alternative neural tissues
  • Alternative cognitive architectures
  • Mindbots and agents for export of cognitive functionality
[edit]

Multitasking and breaking out of the action bottleneck

Multitasking and breaking out of our "action bottleneck"

The human action bottleneck is the amount of action that can be taken at any one time. While the human physiology maintains itself with parallel circuitry, metabolic pathways, and trillions of cells, human output is reduced to ten fingers, arms, legs, vocal output, etc., capping information output (not necessarily information production). There are rare individuals on the planet and throughout history who have seemingly broken out of action bottlenecks, in fact many have and are usually known as professors, who have the unique opportunity of exploring idea space with the help of their students, assistants, etc. However, this mindlink (proffessor-student) is poor and still suffers from action bottleneck especially as the system is scaled upwards and a beaurocracy develops. To remedy this situation, I propose research into direct neural interfaces with "subminds" such that information is directly transferred from the lobes of the mainbrain to other beings, either the typical brain-in-a-jar (with a robotic body), or to an untethered human body (maybe a clone, or maybe a monkey (opposable thumbs are useful)).


Submind connectivity will be similar to "mindbot" modularity. Mindbots are defined as software agents (or "bots") that interface directly with neurons and listen (sometimes speaking) for commands, analyzing the signal spikes and transforming the spikes into some computational action. The similarity between a submind and a mindbot (or agent or avatar) is useful and may provide for readily switchable brain components. For example, a simple and common mindbot already in production is the auditory prosthesis that listens to the air and whispers sounds to the neurons, or the more recent research that is expected to lead to a mute being able to talk via 41 neurons and some fancy analysis software within the next few weeks.


By porting redundant mental operations, or even physical tasks, to mindbots/agents/avatars, such as cooking, mowing lawns, calculating and doing basic algebra, screening job candidates, etc., the brain can leave the body to do other (important) tasks while maximizing the amount of action that one can take per second. Recently, Todd Drashner of Orion's Arm emailed the following, which is quite relevant to this idea:

On Thursday 25 October 2007 22:33, drashner1 wrote:
> As an example of the ideogenetic process, a modosophont might
> experience a snatch of rhythmic noise that inspires them to imagine a
> tune, which they may eventually turn into the basis for a piece of
> music, usually at some point days, weeks, or months after the initial
> experience. In contrast, a first singularity transapient experiencing
> the same bit of noise might produce several dozen pieces of complete
> music, usually in at least half a dozen styles, and a similar number
> of literary, visual, and performance art works, each supported by
> several hundred thousand words of commentary, notes, and scholarly
> writings on the subject of its own work. All within a minute of
> hearing the inspirational noise in the first place. And this same
> creative impulse can be applied to virtually every other event that
> the transapient experiences from moment to moment, all the time.

[edit]

Decoding neural signals

[edit]

Brain-in-a-jar

  • Brain transplantation (White's experiments with monkeys)
  • Neural tissue engineering
    • Researchers have been able to hook up mice brain slices to computers to play with flight simulators.
  • Neurofarms -- farming tons of petri dishes of brain tissue slices or brains-in-a-jar.
[edit]

Brain implants

Current tech does 5 MB/sec in/out of the human skull with at least 100x100 tip MEAs. Kevin Warwick has demonstrated sensory augmentation/addition and human-human neural links. Roadmap: more biocompatible materials. Tissue damage due to heat and metal is not good.


Check out the page on brain implants. One of the brain implant designs involves viral gene therapy on a lab on a (microelectrode array) chip. The idea here is that we can program new sequences to modify our brain in real time, but this is rather dangerous. This is why we need massive neurofarms or computational biology simulations and sandboxing so that we can test out modifications before potentially harming ourselves.

[edit]

Neuroengineering competitions

In the spirit of competition, the Neuroengineering Competition will consist of either mice or human runs with brain implant technology (perhaps even viral gene therapy on a lab on a chip as an evolutionary experiment with some sort of selector for intelligence). See the Inner Space Foundation for more information and interested individuals.

[edit]

Extra

In a few years I will be making an announcement to the internet:

I will personally install brain implants to anybody who shows up at my door. Absolutely no exceptions.

-- in the mean time I need to come up with enough cash to work with MOSIS on the implants (mice don't cost much).
At-home neurosurgery known as "trepanation" (do not forget metal plate to screw into skull).


Miguel A. L. Nicolelis MD, Ph.D re: monkeys moving robotic arms and exoskeletons (Idoya)

[edit]

Mind Uploading

- http://minduploading.org/
- Mind Uploading Research Group (experimental worm-mind uploading)
- Joe Strout and his page
- Anders' page on MU

* Where is the old grass-roots open source group that was researching worm mind-uploading?

Experimental mind uploading:
- Whole brain simulation (brain data gathered via a variety of methods: Keith F. Lynch nanotech urine idea, slice-and-scan, gen-eng'd colored proteins in neuro for imaging)
- Brain replacement (neuron-by-neuron).
Important: Non-nanotechnological mind uploading tactics.
-- Slice & scan method (see the connectome project)
-- Soft, incremental uploads or brain replacement (lobe-by-lobe)
--- Relatively recent "artificial hippocampus" (on a chip)
--- Cochlear implants
Uploading does not necessarily have to preserve total identity. Amazing technological progress would be partial uploads.

People are cryogenically frozen in the hopes that they will one day be thawed when a cure for some ailment is discovered; similiarly, it should be possible to maintain parts of people on life support indefinitely given AdG's anti-aging research. This will allow for neural tissue cultures or brain regions to be on life support (perhaps indefinitely) linked to silicon interfaces for living an otherwise typical life. How long can we keep neural tissue cultures alive?

[edit]

Astro


- Microlaunchers, open source
- Getting off the rock (organizations)
- Space vehicle designs
- Stable, reusable rockets
- Review of satellite engineering
- Laser communication
- LEO/XEO colony designs
- Asteroid mining bots (cite recent NASA news re: landing on an asteroid)
- Hydrogen harvesting procedures and planned operations (distant, yes)
- von Neumann probes (see my implementation notes)
- Megascale engineering plans (not immediately necessary)
- Space tether/evalator review (here?) (lots of fanboys)

Mallove, E. F., and Forward, R. L. Bibliography of Interstellar Travel and Communication. I. J. of Brit. Interplanetary Soc., 27, 921-943 (1974); 11. J.B.I.S. 28, 191 - 219 (1975); 111. J.B.I.S. 28, 405 - 434, 1975).

Interstellar communication bibliography - I need a copy
Earth-to-orbit Transportation Bibliography
Joshua Fox recommends Centauri Dreams: Imagining and Planning Interstellar Exploration


There is a flourishing community of space pioneers in what is known as "NewSpace" (in contrast to dinospace) consisting of Bigelow, Masten Space, XCOR, Unreasonable Rocket, and lots of other teams that are busily working to get all sorts of probes, rockets, satellites, UAVs and other craft out into orbit and beyond. Local (citizen) rocket clubs. However, most of this tech and all of the collective experiences seem to be "behind closed-doors" in comparison to the transhumanist community.

[edit]

Solar Power Satellite

[edit]

von Neumann probe

  • 2008-03-09 - Lunar ark (moon-base) to contain DNA and encyclopedic information
  • The basic idea of a von Neumann probe is to have a space-probe that is able to navigate the galaxy and use self-replication (see RepRap and bio). The probe would contain hundreds of thousands of digital genomes (sequenced DNA), DNA synthesizers and sequencers, bacteria, embryos, stem cells, copies of the Internet Archive and a significant portion of the WWW in general, plus the immediate means and tools to copy all of the information and create a material embodiment, kind of like running an unzip utility on top of the thousands of exabytes predicted to be inexistence today. This would probably include many people, societies, even entire civilizations if we can collect enough data and begin to 'debug' civilization. The system might end up using an ion drive and a hydrogen collector, with on-board nucleosynthesis to create the biomolecules necessary for life, plus ways to attach to asteroids and begin replicating and copying the data and biomaterials. Self-replication is, in general, an interesting problem. See the Center for Bits and Atoms, RepRap, some MNT websites, etc. In debates and discussions in wta-talk and extropy-chat, there has been many a suggestion that mastery of self-replication will lead to post-scarcity economics or acceleration in some way-shape-or-form. The three methods mentioned on this page are (1) atom holography (generally: precision control of matter/energy), (2) molecular nanotechnology (MNT), and (3) bacterial bootstrapping (with the open source synbio/biohacking community, biobricks, the synbio mailing list, Endy/Weiss/Tom Knight/etc.), though RepRap might get to it first with the 3D fabricators (or even fenn's hextatic). An often unexplored and unmentioned method is semiconductor manufacturing but a good first step would be a si-fab that can make a fab on a chip. The idea of the von Neumann universal constructor is that of some machine that is able to construct all possible structures within some given environment, including itself. Molecular nanotechnology (MNT) is an important alternative to investigate along-side atom holography technologies.
[edit]

LOX Infrastructure

[edit]

DIY LOX

  • Air intake valve, molecular seive to take away unwanted components of air, oxygen storage, oxygen cooling via compressor,
    • (2008-03-09 15:58:39) hjohnson: you then take, say, half of it, and let it expand, and use that really cold newly expanded gas to cool the remaining compressed gas, expand half of that and repeat until you get liquid out.
  • Fractional distillation
  • Don't have to buy pressurized oxygen.
  • Gas compressor
    • Piston design: have a giant metal tube with a plug on the other end, insert the compressed oxygen, then ram a giant piston down the neck with lots of horsepower, then let it go sit in a pressurized chamber somewhere else for a while.
    • Study the gas laws again (not the ideal gas laws).
  • See notes on How to design, build and test small liquid-fuel rocket engines (the Leroy book + Sutton).
[edit]

Orbital LOX fabs

  • Make the fuel in LEO.
  • Refeul en route.
[edit]

SugarShot2Space Project

  • Sugarshot - the goal is to use sugar propellant with potassium chlorate.
[edit]

Space Habitats

  • Space Habitats in Orion's Arm
  • Open air space habitats - "A home in space need not be the enclosed volume usually described in most movies, books and articles. If a strong enough material is used, a rotating cylinder can be so large that it holds an entire atmosphere against its inner surface." (Forrest Bishop)
  • Ideally, we should be able to house people or at least plants in space habitats.
[edit]

Rotating paired Module Habitats

[edit]

Stanford Torus

[edit]

Bernal Sphere

[edit]

O'Neill Cylinders

[edit]

Bishop Rings

[edit]

McKendree Habitat

[edit]

Plant panels

Surface area used with either algae or plants or other photosynthetic organisms with automated deployment of water and other nutrients, as well as automated harvesting of hydrocarbons. This would essentially be many thousands of square kilometers of grass.

[edit]

Open source spacetech


Open source aerotech and spacetech has not come along in full force because there has been a historical tendency for the major governments of the world to keep information behind closed doors so that foreigners don't figure out how to make missiles and so on. Thus why most of the computational fluid dynamics software is very expensive and only for U.S. citizens, some even only for DARPA contractors with specific clearance, even though ultimately this software can be copied from first principles and further experimentation ("what one person can think, so can another" -- Damien Sullivan).

[edit]

NewSpace

[edit]

Competitions

[edit]

Lowering the barrier to entry

  • Launching barrier
  • Machining - see CNCzone
  • Team recruitment forum over at OSAERO to lower the barrier to entry
  • Cost of fuel for launch (millions of dollars at the moment)
    • See DIY LOX
    • Sugar propellants (candy propellants)
[edit]

RepRap in Space

Transistors have been solved with semiconductor nanocrystals or the Artemis Society and their method of manufacturing silicon on the moon (which is rather complicated). Demonstration of RepRap technology must proceed on earth first.

[edit]

Asteroid mining

Go to the nearest asteroid field, track and tag at least 10,000 asteroids, and then identify the rocks that have the most valuable resources and start mining. Fire packages of mined material back at a central processing node, and give each 'pellet' a solar panel and some ability to propel itself and navigate, but not much, at the other end send a signal that would cause the pellet-guider (fins, I call them) to detach and let the pellets continue on for further processing.

[edit]

Asteroid tagging

[edit]

Asteroid following

[edit]

Asteroid landing

Hitting one dead-on would be a really good start.

[edit]

Asteroid mining machinery phase

In this phase, deploy the machinery on to the surface, begin drilling process and start mapping out the internal infrastructure of the asteroid. Explosives might be required if the asteroid gets too dense.

[edit]

Pellet system

The fins need to be reusable and somewhat programmable as for their destination.

[edit]

Pellet processing system

Maybe a giant net can be used to pick up incoming pellets, which are conveniently tagged kind of like an RSS feed. Except it's a physical, constant feed.

[edit]

Cloning

Present here an abstract look at what is possible with technology- cite Kuwabara, the Japanese researcher who built an "artificial womb" or the more recent researchers who have done IVF-on-a-chip. Find the original "Dolly the sheep" PDF.

No longer requires embryonic stem (ES) cells (see recent Yamanaka research on iPS cells as well as Shoukhrat Mitalipov on cloned monkeys from skin cells)


  • Cloning humans from scratch.
    • To what extent do we mean to say "scratch" ? See from scratch for the different levels and attitudes of from-scratchness. You could clone a human from an egg, clone a human from a skin cell, or does this mean cloning a human from a single strand of DNA known as ssDNA? Or does this mean using a DNA synthesizer to take the digital bits and bytes and create a new synthetic oligo strand, amplify/replicate via PCRs, and then go through some other steps to create the human via an artificial womb? (While unrelated to cloning, synbio presents another "cloning from scratch" project -- namely, the idea of rewriting the human genome and making something that ends up looking and acting exactly the same, but with different genomes.)


Cloning also provides the unique opportunity for neuropsych research into the brain. With artificial wombs and Skinner boxes ("baby-in-a-box") it is possible to completely specify all stimulation and inputs (like nutrients) through a model organism's life. Will similiar brain architectures emerge? What, then, will we learn through slice-by-slice brain upload scanning? Will we discover how precisely different stimulants alter neural tissues (comparative neuroanatomy)?


Cloning increases redundancy, and increasing redundancy means reducing the possibility of total death: you are free to go out and do risky items of business, knowing that should you happen to die, you have automated systems back home that will recreate your body (though definitely not necessarily memories, resurrecting dead memories is a difficult art).

[edit]

Alt-bods

- News: fuel-powered artificial muscles and semi-artificial blood vessels.
- Redesigning the human biody, biomedical engineering, artificial organs (cochlear implants included),
- Genetic engineering, antiaging, gerontology, Aubrey de Grey
[Aubrey.zip Aubrey de Grey paper archive] (zip, 25 megabytes)

AdG's research deserves a more thorough review. Basically, research is needed to figure out how to clean up metabolic waste and all of the extra gunk leftover in aged cells. Lysosomes need to be replenished. Mitochondria mutations need to be eliminated. AdG suggests stem cell replacement therapy every decade, a massive animal experiment to engineer immortal animals (for testing purposes and to stay ahead of us), and longevity-mice experiments within the next ten years. Insert here a more thorough summary of his papers (from the zip file).

Alt-body projects can be of two main types: keep the human body or not. In the first case we see the development of "open prostheses" groups and in the second case there is the development of robots in labs all over the world. Robotic development is mainly geared towards AI research and not body replacements. However, with brain interfacing and brain-in-a-jar, robotic bodies can turn into ideal replacements.


[edit]

Biogerontology (anti-aging)


Aging is a very complex issue, given that life is complex as well. Aging is the natural process whereby the body is changing its own state in a detrimental manner, such as by the accumulation of junk molecules (undegradables), or the death of stem cells required to keep the body alive, etc. Drugs to counter aging are going to be difficult to discover, thus one proposal has been to have tissue farms and zoos to keep track of tens of thousands of specimens all tested with the latest and greatest treatment of medicines and different technologies to keep the body and its components healthy. This is much like combinatorial chemical library testing for massive drug searches. Can we use the immune system to fight senescence and to 'homeostatize' the aging process?

[edit]

Body refreshing

In this method, the body is replaced cell-by-cell usualyl with molecular nanotechnology.

[edit]

Stem cell refreshing

Aubrey de Grey suggests going in every 10 years to get a stem cell transplant (bone marrow transplant).

[edit]

Finding novel ways of degrading the 'undegradables'

[edit]

Synthetic Biology

  • Synbio review
  • Venter's minimal cell project
  • Cellular synthesis
    • Construct artificial cellular membrane that just barely works and hope that injected cellular mechanisms can slowly repair our poor substitute. Evidence that cells have membrane-repair tech: if meiosis allows for cellular replication, and cells are plus or minus the same volume, where's the surface area of the membrane coming from? So we do not need to mimic a mature cellular membrane.
[edit]

Synbio - neuro

Artificial neurotransmitter receptors ("deceptors" by Freudian-slip) to link with GTP molecules in underlying membrane surface re: cellular signal transduction pathways. "Mind expanding" in a different way.

[edit]

Genetic engineering

[edit]

See the Open Biohacking project

The Open Biohacking Project or biokit project was released late January 2008 and received upwards of 20,000 hits and over 120 GB of transferred data in its first week of existence.


Groups have been appearing around the globe. Check out the Boston DIY bio group.

[edit]

Genetic circuits

[edit]

See also

[edit]

Energy

[edit]

Meat on a stick

The meat on a stick project is to develop artificial meat using a supply of stem cells to grow myocytes (muscle cells) in a tank of filtered nutrients and energy. The inputs would ideally be only grass and water, the output being waste byproducts and meat. There are some questions as to how this would work with the immune system being absent and may require gastrointestinal bacteria of the host species to be transported into an artificial stomach of sorts (or maybe not). Contamination may or may not be a problem. What are the protocols for culturing myocytic stem cells? What sort of motor do we need on the filter? How much sterility would be required? And how would we copy the stomach if that is indeed needed?


Re: the immune system problems. This can be solved by freezing meat and having backup energy reserves. Should the immune system happen to fail, and the currently growing meat supply begins to die, use a portion of the energy reserve to track down the pathogen, isolate it, characterize it, and determine a solution whether an antibiotic or (in the worst case scenario) burning the current food supply and starting fresh from frozen stem cells and eating from the cache until everything is back online.


Controlling our food supply will allow us to separate ourselves from the ecosystem more fully -- to the point where we can harvest grass in giant space habitats (2D arrays pointed at the sun) and feed it to the boxes to generate meat most optimally (no reason to allow brain function, complex circulatory systems, etc.). Ideally, we could just use energetically efficient nucleosynthesis plus biochemistry to build food from the fundamental elements (nitrogen, oxygen, hydrogen, carbon, etc.), but we do not yet have this technology.


See also the stomach-bacteria project, which may lead to the elimination of the "inbetweens" of food from the starlight to the forms that the hu can process. This is not necessary for the MOAS project.

[edit]

Materials

[edit]

Nucleosynthesis

Importance of being able to transmute hydrogen into elements required for our biochemical existence. Note the absurdly large amount of free hydrogen in the universe. Nucleosynthesis is the natural process that occurs when stars explode and go supernovae upon the cosmos. They then convert massive volumes of matter into heavier atoms with maximized number of protons and neutrons. This is how we are able to have carbon, nitrogen, sulfur, aluminum, magnesium, lithium, etc., instead of just hydrogen. (This all connects intricately with condensed matter physics, nuclear fusion, cosmology, astrophysics, etc.) After collecting a massive amount of hydrogen, we could depost the hydrogen into a giant star and make the star go supernovae through some other method, but then we are slowly killing the galaxy and that would not be acceptable. Instead, perhaps we can control the process of nucleosynthesis via putting it on a chip or at least in a lab in a giant chamber at first.


Why might we want to be able to control the process of nucleosynthesis? The universe has massive amounts of hydrogen for the taking, it makes up a relatively large percentage of the cosmos including the molecular nebulas throughout the galaxy and beyond. For this reason, it is a useful source of matter, and with meat on a stick technologies (where food can be grown from the biochemical substrates without growing entire cows and pigs) we could have a replenishable supply of food -- it is maximally replenisible to the extent that the universe would allow it to be replenished ... once the hydrogen supply is gone and used up, hopelessly radiated away faster than we can recover it because of either nuclear reactions or stellar explosions for hundreds of billions of years into the future, then it's all gone and we would have maximized our use of it. Of course, we have to ask whether we really need to automate this process. It is important to note that such a scenario means that we become the grey goo even though we may seem moderately "intelligent". Having automated cloning and automated food production from automated hydrogen harvesting could allow humanity to use up all of the resources of the galaxy and beyond, and is this necessary? Is this needed? This is not a question of ethics, but rather a question of programming of the human genome and the human metaprogram. Are we to encode restrictions in human potential, so that we do not consume the resources of the universe hopelessly beyond reuse? Or will we let ourselves become a cancer of the cosmos? Signal-to-noise. Thermodynamics. Heat. Dissipation. Energy gradients.


[edit]

AI

  • Henry Markram, but he is not focusing on ai. Instead, he's doing computational neuroscience to simulate the whole brain, starting with neurons. Notes on his work can be found [[[Henry Markram computational neurosci imitationproject|here]] and here.


Artificial intelligence is typically used to meant to refer to the computer science approach to problem solving by trying to model the brain. Historically, this has meant linguistic modeling, theorem provers, and other attempts at very generalized intelligence. However, recently there has been an attempt to more closely model intelligence by looking at the biological substrate in computational neuroscience and so on. Whenever asking questions about ai, the real question becomes to what extent do you want to use the word 'artificial' and what goals do you have? Do you want to write an intelligence from scratch? Wouldn't you have to use some generalized principles? Do you want to use a problem space and apply various methods? Or do you mean to say that you want to set up an evolutionary experiment to eventually generate intelligence among the members of the population? Supposedly, you can go set up organ farms and neurofarms and use tissue engineering (particularly neuroengineering and go out to build farms of alt-brains that are unlike previous brains, and yet to what extent would this count as artificial? What if you used an evolutionary experiment to make a new type of brain by evolving one from scratch, just like the Urey experiments to evolve amino acids, except this time for an entire brain (condensed into a period more managable by a single person, instead of billion years by the natural coarse of history and so on?).


The historical method is not "wet" and does not address the wetness. The brain is, itself, open wetware, although not so open (not yet) -- while the genome is accessible to all of us, the design principles and ideas within it are not exposed to the general public or accessible to agents willing to eat away at the energy gradients hidden within the brain. Again, the traditional methods are more static and crude, and do not allow for the natural biases and errors that appear within the brain, not because the brain is doing any exact Turing computation, but because the brain is a general tissue that is responding to certain noises and certain signals based on historical context and different molecular gradients, protein expression and the various established thermodynamical structuralist relationships (in the sense of Prigagonine and Salthe). Maybe a more important question we should be asking is how we can encourage the development of intelligence without stifling intelligence. How do you design a selector for an evolutionary experiment aimed towards ai?


  • Failed methods in AI
  • Comp sci vision research
  • Combinatorial explosion / search problem
  • Failure to define "intelligence" (maybe "intelligence" is not so useful of a concept)
  • Create AI paper (p)reprint archive (ex: http://arxiv.org/)
  • Cortical simulator
  • Neuronal modeling (ANNs)
  • Our inability to answer the question "what is intelligence" suggests that completely giving up on 'intelligence' might be the right path. Whatever it is that we are trying to do with our computers, it is something other.
  • Insert Goertzel's additions here.
[edit]

Perspective shift: we *are* artificial intelligence

That humans are self-replicators is a good head start, what further that we are already intelligent (by definition). Although accessing the source code to our intelligence is not possible, due to the lost selection experiments throughout evolutionary history, we do have a strong foundation for self-modifying intelligence, intelligence amplification, alt-cog, and so on. It is, in general, not wise to rely on any future in which artificial intelligence is a reality since it is a black swan. It would be interesting if ai does develop, yes, and even useful, but its creation might not be predictable.

[edit]

Information aggregation

A human create is limited by information production/consumption rates in our "internet information ecology". Just as we have the Internet Archive and many search engines, need to help fix the single-point-of-failure nature of the WWW, as well as provide for a method of content distribution and backups, esp. as we move closer and closer to mind uploads. Natural solutions have been progressing, coming about on their own, but that doesn't stop top-down visions for guidance to pop up over the place.

[edit]

Social

WTA and the extropians
- Roadmap objectives: maintain high signal-to-noise ratio of engineering/science; educate transhumanist-newbies that are willing to become researchers (high priority). Context-sniffing, context training, ability to work on internal coherency rather than defense of stale ideas and plans.



- Highlight important researchers, what they are researching, how to fund them.


silicon
International Technology Roadmap for Semiconductors


Include a "doomsday ASAP help" file-- i.e., who to contact in case of experimental emergency involving transhumanist tech.
a.k.a. "Help! I'm the guy that's caused grey goo! Now what?"

From #SL4:
- Suggestion of a central coordination agency for this roadmap
- Long term projects, feasability studies,
- Measurable goals, establish philanthropic funding for the pursuit of transhumanist technology

Truly parallel computer architectures (not the "multicore" stuff)

Retrieved from "http://heybryan.org/mediawiki/index.php/Roadmap"
Personal tools