I was thinking of a profit distribution of  1/3 tax, 1/3 donate back to other open source/non-profit, and 1/3 for capital investment.   Given the state of our country, if there's profit I don't mind the tax because the free air I breath is paid by someone else who came before me.  

I don't know if the above plan is sustainable.  Regardless, the company will be financially transparent so its books will be open for all to see.   

I like to find some people to tackle the various basic fabrication skills.    For example, learn how to use FreeCAD to design a simple widget that can be cnc machined.  Learn how to use open source CAM to machine the widget.  Learn how to use open source FEA software to analyze the widget.  Learn how to use KiCAD to diy simple pcb that can be combined with the simple machine widget.    Maybe document what we've learned so other can follow.    These are some basic open source manufacturing skills that I think everyone can use.

Hopefully, as a group, the learning process will be faster and easier.  Too keep everyone interested, we can pick a simple project, ie a LED desk lamp or simple headphone amplifier, something simple that we can actually manufacture with whatever tools we current have.

You might want to check your local Hackerspaces to find people with similar interest in your area.

So, you're saying you went to a hackerspace in Los Angeles and asked if anyone wanted to work without pay and nobody seemed interested?

Are you suggesting that interns want to be paid but professionals will work for free?

It's important to clarify whether or not it will be for-profit or non-profit. There are effectively no case studies describing successful for-profit open source technology. You'd be breaking new ground.

It's a learning process.   I don't think it is going to be easy.    Most machinist are not open source users and most open source users are not machinist.  I hardly quality as either.   However, I still want to give open source manufacturing a try.    Learning to draw in FreeCAD/LibreCAD is not easy and for a while I seriously thought about buying a seat of Solidwork/Inventor and a better commercial CAM package.   However, I would much rather work and promote the open source movement than to be a cog in some proprietary product world.  

Right now I have a choice.  I have a space with a few US made cnc machines and I have the flexibility to share my resources with other like minded people who wants to learn about open source manufacturing.    The various US made equipment are standing by.    However, my pocket is not deep enough to sustain this type of cash burning hobby, much less to hire a crew of engineers to figure out open source engineering.

I hope I can find others interested in open source manufacturing and form some sort of a domestic manufacturing venture.  If not, I'll still go at it at my own tortuously slow pace. . . .

Yeah, do you have anything more specific than "I want to make stuff?" Unlike software, hardware needs to consider cashflow as integral to the project. Even prototyping something physically costs money, unlike software which can usually cost only pizza and sleep. So, where the money comes from needs to be part of the very first steps as the concept develops. Is it going to be a hobby? Is it going to be contract work? Is it going to be run on donations?

My suggestions is to find an existing company/group that is doing open source work and manufacture hardware for them on a contractual basis. Everyone who committs to open source as a philosophy tends to want to help exand the number of open source resources, so they might use you even if you're a bit slower and more expensive.

On the other hand, maybe you don't want to be a manufacturer so much as an educator/innovator. You could develop and troubleshoot open source manufacturing processes and tools, then if they work you could sell them. Maybe ask around to see what small-scale manufacturing stuff people would pay for. Like a PCB assembly line for all the guys who like to make prototype electronics. Or a modular machine shop so people can start with a lathe, then turn it in to a mill, without buying an entirely new machine.

I dunno, man. I have to speculate because you haven't said what it is you want to do.

What are you offering/suggesting then that OSE isn't already doing/planning on doing?

What are you bringing to the table in terms of skills, resources, or finances?

The cost is the 3 Haas CNC, the infrastructure to get them up and running, ie electrical subpanel, wiring, air compressor/dryer, pneumatic lines and etc, plus rent and utility to house everything.    The plan is fairly traditional.  Build small, build simple, and iterate to slightly bigger and better.  Rinse and repeat.      Same thing with marketing, find niche market and start small. . . and actually I do want to build a LED lamp. :)

Lol, wow you guys are a tough crowd.    Right now, I only worry about learning open source tools for open source manufacturing and documenting the process  regardless of success of failure, so others who want to follow can have a slight head start for the next iteration of open source manufacturing.  No point reinventing the wheel every single time and it will be nice to have a tutorial to follow.   For example, I am a Debian user and when I look at cealinux the first thing I wonder about is how can I run it under Debian instead of the Ubuntu live cd and where are the tutorial to learn the cool and seemly powerful open source tools.   It'll be nice if such tutorials exist right now.

Yeah I know, it's a long process and an overkill but that's learning the basics and hopefully we'll all have fun along the way. 

I don't even worry about marketing the LED lamp.  The worst case is everyone participated in the manufacturing process gets one.  So we only made a few.  No big deal, at least we learn the process for the next project.   That's just the reality of business.  Not every product will be successful, but success or failure, it should be a learning process. 

The first step is just the learning process so a lot of the details can be over look, ie build a few led lamps that only the developer will want.   

The first post is fairly long, but I stated my philosophical motivations and a brief overview of what I want to accomplish which is something potential partners will want to know.   If someone wants to empower the world through manufacturing in authoritarian countries, then this manufacturing startup is probably not the right one for that person. 

The first project I'm working on is fairly simple and I'm beginning to see the error in choosing 304ss over 303ss. 

http://tux-lab.com/tux_project_detail?id=2

I want to move beyond just a "hobby" hence the first post, but let's not go over the same cost and what to make part because at this point, those are fairly irrelevant.    We're just going around in circles. 

Documentation is already a given and was mentioned many many times.   My ruby/rails skill is pretty pathetic, but feel free to suggest a better way to document the process.

The goal is to learn open source 3d CAD, circuit design/layout, FEA, and whatever open source software it take to design and build something.   What to learn together and speed up the process?  Great, let's figure out what we have, where we're at, and what we want to build.    The fallback project is a led lamp.  

Some simplified documentation on FreeCAD is available at http://opensourceecology.org/wiki/FreeCAD.  I documented the capabilities that I figured out during the development of CAD models for the OSE Steam Engine of last year.

- Mark

@Sensor:

Warren Buffet once said that total historical investment in the automotive industry as of I believe the year 2000 netted very close to zero dollars. This was prior to the GM bankruptcy. He also stated that total historical investment in the aviation industry was negative. Furthermore, the US government issued tens of billions of dollars in loans to US solar companies. 535 million to Solyndra. 1.43 billion to First Solar. 1.6 billion to Bright Source Energy. And, the list goes on. Engineering projects are expensive and prone to corruption and failure for technological reasons. The losses are enormous. Yet, humanity benefits tremendously from the technological march. The financial benefits of patent protection and hoarding trade secrets
are not sufficient to warrant large investments into technology. Furthermore, companies should not reap the benefits of risky loans at low interest(for the risk) while leaving tax payers with the risk of default.

What is my point? All non-defense(or dual use) government funded projects should be open sourced and available for public criticism before grants are given or loans are issued. For something like solar energy, the whole world benefits from the spread of the technology. For every solar thermal or photovoltaic plant that the US or China builds, the whole world benefits from reduced CO2 emissions. Granted, manufacturing would flow to lower wage countries with good infrastructure, but manufacturing need not all flow to the same country, which would alleviate some national security issues. Countries such as Mexico and South East Asia would be well positioned to export to the US as well as China. In addition, the power plants themselves would be setup in the US, Spain, and other countries with solar thermal resources. This would allow the distribution of financial benefits. Furthermore, an open source license would encourage all countries to contribute to projects with the potential to obtain some benefit.

Another area for open source projects is medical and agricultural equipment and chemicals.

As a step forward, I believe model projects and licenses are required. I would propose the following.
1.)Create a viral open source manufacturing license similar to the GPL. The license would basically state something like the following:
"Whenever this widget is manufactured and sold, the recipient must be given all the mathematical and computational models, manufacturing drawings, and software with each widget sold. The information can be made available through a DVD or online."
2.)Develop an Open Source Engineering Project Kit that would easily allow an individual or group to setup development servers, development machines, and manufacturing machines. Some of this already exists. For example, CAELinux may be a good choice for development machines. And, CNC Linux provides a good choice for manufacturing machines. A development server that deploys tools such as trac, svn or git, and ejabberd on a stable and secure platform like OpenBSD would be ideal. Scripts could be written to install all the necessary software.
3.)Model projects would need to be started in areas such as solar thermal and medical devices. I'm considering starting a project in solar thermal.

Matt_Maier, as far as I know, there are no mature open source engineering projects outside software. Thus, any financial analysis would be conceptual in nature since no empirical data exists. The empirical data would come from analyzing patent law and public funding of R&D, but there would not be any basis with which to compare these things to open source engineering. I searched for an open source engineering license that was like the GPL license for software. However, I found no such thing. Thus, a first step in a more rigorous legal analysis would be to draft a sample GPL-like license for open source engineering/manufacturing. Once drafted, a line-item debate on specific points within the contract could begin. After the debate, a new version of the contract could be drafted and a new debate could begin.

One example of a very large and successful open source hardware project is the Arduino.

Arduino is really great, they are programming a Arduino to replace the proprietary board in my wife's dental chair cause the manufacturer went out of business and replacement parts for the controller are unavailable