CubeSpawn : is a 4 year old project to build an Affordable, Distributed, Open source, FMS. 4 months ago, the desire to embed a cryptocurrency into the operations of the machines led back to Bitcoin and the Alts.
By giving each machine its own budget, allowing it to charge for the work it does for the rest of the network of machines and letting it buy its own materials and tooling, the idea of adding the "Distributed Manufacturing" title begin to take on a dimension of plausibility once you add Cryptocurrency to the mix.
If the Meta-control system were turned into a DAC (Digital Autonomous Corporation) the system then has the potential to incentivize engineers, machinists, makers, hackers, coders and customers to use the system, contribute to its growth and the growth in the library of Digital patterns that it uses to create parts and assemblies.
I am announcing this in Ethereum first to see what level of interest is shown and weather skilled designers and developers will take an interest helping to form a global/local decentralized factory model.
My hope is that once you understand the contribution decentralized manufacturing can make to democratizing access to physical goods, You'll realize it is to material wealth what bitcoin is to money.
Thank You for your feedback
James Jones, CubeSpawn Founder
Comments
i didn t got your concept as in your store i ve only seen one kit for a 3d printer.what else are you proposing exaclty?
do you any 3d printer for metals or else?
thanks
Needless to say, they're very interested. You might look into hooking up with them.
... and I am, too, I mean, but they're farther along than I so you'd probably have more to gain from talking to them than me. I'm certainly willing to work on this, too, though.
Here is a link to some photo's of the system https://picasaweb.google.com/103828779781480193226
what i ve seen looks like to reprap type 3d printing machine
so can you elaborate and tell more about this as if the concept is interesting i would like to apply it to underdevelopped countries
thanks
I also have looked at building a laser cutter for flat work, a laser tubing notcher, a plasma table, a vacuum former, Motor winder and a long list of additional machines, including the pallet loading and transporting system.
The real advantage to the system is a modular, standards based "form factor" that all the machines conform to, it scales up from a relatively small 300 mm cube up to a 2.4 meter work cell by doubling (as in: 300mm, 600mm, 1.2M 2.4M) the cube frames manage power and data connections.
Mechanical modules performing different functions can be slid into the frames to make any machine carry out any function that has been built to that point in time. The modules order can be dynamically re-arranged to optimize workflow through a system of cubes
All the basic, general purpose machines are open source, the software is also open source.
The control system runs on almost any general purpose single board computer (like R-pi, Beaglebone, also industrial boards)
the control software is a modified version of ROS (Robot Operating System) http://www.ros.org/
The design is fairly broad in scope and reasonably mature for an alpha ;-)
rebuilding enough technological capability to repair/rebuild the ship /new ships/a new fleet etc. and therefore keep exploration alive. A happy hopeful dream. things seem to be moving along eh?
BTW the lunar vacuum would be great for 3-d metal printing no? The code for the blackbox of the technology of a civilization and the code to rebuild at another exosolar venue?
Go. 400km/sec
this is a great projecgt CubeSpawn. Press on. I am sure there is funding, base, and i agree it could be self-organizing funding.
What are the best chip technologies for Cnc and 3D printing. If cpu then bring on the arms race for the coin for the eventual repurpose for the builds.
LunarTX
We're getting there!
Ratings: Agents trusted by the network and by other agents - works like Reddit up-vote redux, meaning a rating system from users, contributors and customers of the system AS WELL AS a rating from the machines.
The machines would rate based on thing like: shorter cycle times compared to historical data, longer tool life, lower reject rates, better efficiency, and other Quality control or operations ratings.
Incentives, are paid out based on ratings at a very low period by period basis (month, week, day) and bonuses are payed on longer time intervals based on cumulative ratings and usage, all these payments represent a percentage of the fees collected by the network (similar to mining but with fabrication volume/ system utilization as the determinant)
Incentives would apply to:
Commission a new system, scaled by where its needed (small percentage bonuses for locations devoid of service at commission time)
Submitting a new machine design:
Very small incentive for a new design, modified by ratings from trusted agents.
high utility/acceptance --> high reward.
A similar pattern applied to Enhancements to existing machines for accuracy, speed, efficiency and a mechanism for scaling demand for fulfilling system requests from all agents (ex: 500 people upvote a request for a feature - fulfilling in a highly rated way nets a large initial bonus and a decent residual from all installed instances.
And, of course, features to the control software, materials handling, supply chain management etc etc sub-systems would work in a similar, if not identical, fashion...
Other ideas: sponsoring a new system might be a worthwhile event to incentivize, helping to bootstrap new participants to add capabilities or to add a new location to the network.
All of this fits in to 6 general incentivized classes:
Each class would use a bonus/residual model (Bonus to build/design/deploy, residual for using it.
Adding to the network i.e. growth (new systems)
Improving the networks software, hardware or design (engineering/technical)
Running production capability for income (like a franchise)
Sharing capacity with the network i.e. distributed manufacturing
Bringing jobs to the network i.e. marketing and sales
Adding digital templates to the system for products and selling new templates
In this way the system charges the minimal fees needed to keep these six areas paid and manufacturing cost can cut out transportation expense by reducing or eliminating most of the hauling (make things locally!) plus it opens up an opportunity for creative materials recycling solutions to lower material costs.
Lastly, if small scale local power generation systems are added to the digital templates library - energy costs can be adapted to the system operators environment for the cheapest way to generate/supplement power to the array, lowering costs and increasing access to a complete solution to produce democratized access to material wealth.
I have another bullet point . What about integrating a market for engineering services. Much like a fusion of stackoverflow and crowd-working. One posts the challenge, and experienced engineers flesh out a manufacturing plan whereas the design(er) that is chosen by the client is rewarded in ETH and added reputation. I guess this somehow extends the point of "Adding digital templates to the system for products and selling new templates" and would allocate engineering capacity to where it is needed, plus draw in less technically declined people.
Closed loop food, water, power, recycling, industrial production - we have several novel systems at various stages of completion:
Cactus-->Biogas(methane)-->fuel cell-->electricity,
A wood gasifier powered internal combustion generator,
Solar PV,
Solar thermal, (similar to http://www.solarfire.co, but more automated) and a team of participants - this is a no-blue sky, pragmatic group - they have their own money in it, so we are only looking at what we can actually do.
With the goal of evolving it toward better solutions for the foreseeable future... ;-)
Getting Ethereum driving the self funding portion is very important to making this a self spreading technology.
I like the vision
Thanks @Ursium?
Well I guess it depends on the regulatory and disaster recovery protocols in place.
It would be great if we could ...create...engender....instantiate...or conjure (up) an Ethereum based resource recovery cybernetic management software modular solution that could be easily deployed across similar technologies.
(http://rt.com/shows/keiser-report/episode-569-max-keiser-191/)
If all classes of financial transactions decompose into 30 or less "atomic" transaction types, then any management system will similarly break down into a finite class of specific relationships, tests, and feedback loops - so proposing general purpose configurable DAC templates is the way to go, since the structural details of such a solution will likely be similar irrespective of what your managing.
Combining the contributions of many developers with different, but overlapping solutions should speed up the process of getting a working skeleton available...
So - any waste and make power.
So do we "ethereumize" this process?
Does a DAC drone - tell a delivery drone when the garbage is full and then the delivery drone - takes a resource recovery center close by - then the RRS (resource recovery system) chops and shreds the waste to digestible units - then processes via the gasifier chamber - makes some power - gets some ether/BC?
Ur thoughts?
This is probably blasphemy on my part, but the Ethereum code (in CubeSpawn at least) does very little actual direct system control, simply because there are better, purpose built, mature tools to control robots and automation:
(ROS, ROS-i) http://www.ros.org/ http://rosindustrial.org/
I have talked to Shaun Edwards (he is the founder of ROS-i) extensively about CubeSpawn...
So in my view the Ethereum portion of the system acts as a web appliance to collect performance and status information from the CubeSpawn array, then broker the connection to other geo-physically close systems for all the magic schedule coordination, capability advertising, capacity sharing, payments and so forth...
The Ethereum "Module" is very compact, and lightweight, harnesses the CubeSpawn arrays capabilities and Ethereumizes the Web facing interface.
I can't imagine that the EEW system (whether adapted to a CubeSpawn form factor, or otherwise) would be much different - most of what the Ethereum Module "knows" about the system state would be summary information as tasks where completed so building the interface between #some system# and the ethereum module (which I envision on a beagle bone, R-Pi or industrial embedded Single Board Computer or System on a Chip) amounts to formatting a summary report from a system controller or providing each system with an API to talk to the other for discovery...
on the Web facing side, collecting the cryptopayments for services rendered, and paying the Cryptobonuses for "events" (Task Completed, Status ok) and residuals then becomes a simpler, purely performance report related function
I think the 6 category payment scheme mentioned earlier would work fairly consistently across most materials processing without too much variation...
Also, sorting and separation processes "upstream" will have pulled out the "elemental" trash (glass, metals, assemblies) and any intelligent system will trend toward goods designed for recycling
Your feedback appreciated
I've had some working experience with Supply Chain Management. How will the
chains of supply created interact with each other. Operations are in my view are
definitely a big constraint, business networks are formed as a result of human
communication. Decision making, in my understanding should remain with users.
As you said ideas and implementation have a huge gap of hard work in
between. I understand you must still create an interface for effective decisive control.
I am currently working on my own designs for an Enterprise Software with a few
teammates. Would love to gain some wisdom from your experience in the field.
@SatCa - full supply chain management might be overly ambitious to start with, but earlier designs had a OSS ERP focused toward SCM in the diagram...:-)
what I see here starts out considerably simpler, its a reputation system with higher rated agents carrying more influence. All the "intelligence" in the system is human intelligence, moderated through a management system for narrow, purpose specific signals to the software.
I had a discussion today with Daniel Larimer (Bitshares CEO) about a DAC for manufacturing and he expressed well reasoned doubts about the ability of the current state of the art in DACs to manage the complexity, he made several good points, but since I do not see this system as needing to evaluate its decisions quite as thoroughly, I'm more optimistic. Here is why:
You could implement this as a "coupon rewards" program - as in, buy a system and we, (meaning a foundation (group of individuals, cooperating)) will give you a coupon discounting your second system, if you do a $1000 worth of business on the system, we will give you ANOTHER coupon worth 5% of that gross thoughput - But, for this to happen we'll charge you a membership fee that is a small percentage of your system's Gross income (1,2,3%) ...
See!?
This is not an intelligent cybernetic marvel, its just a partially automated incentive program - making it a DAC helps automate the process, to participate you have to provably have a system and meet other requirements (as yet undefined) starting out in this simple way still accomplishes the desired goal (incentivise users, engineering, programmers etc) WITHOUT setting the threshold so high its un-doable, or pushing the implementation out so far its not meaningful.
K.I.S.S still applies, even in this abstract realm.
As time passes progressively more complex and sophisticated decision making can be baked into the architecture, and simplistic models can be phased out.
my 2000 cents! ;-)
James
Obviously, this is the very "back end" - but the fact that it ultimately makes electricity form "waste" is the interesting part.
But - how to incentivize folks to "use" it - its simply the same old "cost/benefit analysis" or some green credits, etc.
But - if they could receive some ether for their trash or better for their electricity made from their trash - until they brought it in-house
Of - even if "in-house", possibly do a solar-city type of thing that Elon Musk's brother-in-law or cousin is doing (in-house but paid for by SC with shared benefits).
So - instead of paying for my rubbish removal - the future model is that the resource recovery systems pays "us" for our waste or resources. This may be way further out "there". But just a thought.