You can't change the world unless everywhere is deployable.

[faddat]

You can't change the world unless you can deploy everywhere.
You can't change the world unless you can't be stopped from deploying.
You can't change the world unless you can deploy.
You can't change the world unless others like you can deploy.
You can't change the world unless you can eat while building code to deploy.
You can't change the world unless people get what you're doing, because then they can choose to help.

I think things are being built so that we can't change the world. How's about you?

Hey,

I'm working on something that riffs on your project. I'd really like to chat with you about your goals and learn if you're interested in teaming up. What are we doing? Decentralized cloud infrastructure, presently based on Docker, soon to be based on Xen unikernels or another type of unikernel. We want to make it easy for people to leverage open source to build world-scale applications without a middleware layer like maidsafe / ethereum. We're pursuing these goals across hardware and software, looking to build a final platform with a cost around $20.00 per server per year, including hardware. The very tip of the iceberg can be found at klouds.org.

jake@klouds.org

[blakelapierre]

Hi Jake,

I think your title is correct. It is one reason why I have only implemented deployment to DO. I'm not quite sure anyone has given a sufficient description of what we should be changing the world into.

You can't change the world unless you can't be stopped from deploying.

Probably.

You can't change the world unless you can deploy.

Highly likely.

You can't change the world unless others like you can deploy.

Agreed.

You can't change the world unless you can eat while building code to deploy.

I can do much more than that.

You can't change the world unless people get what you're doing, because then they can choose to help.

Agreed.

I think things are being built so that we can't change the world.

Define "so".

How's about you?

I think that we (note: I have not defined this word) are likely the only people that can change the world. If you have another theory, I'd love to hear it.

Hey,

Hello

I'm working on something that riffs on your project. I'd really like to chat with you about your goals and learn if you're interested in teaming up.

I'd love to chat with you. Through text on GitHub works for me. Video or audio chat, or even meeting in person would be fine too. It looks like you're in Vietnam, so an in person meeting at this point might be difficult. I am interested in teaming up, so for now, I'll just continue our conversation here. :)

What are we doing? Decentralized cloud infrastructure, presently based on Docker, soon to be based on Xen unikernels or another type of unikernel.

Unikernels do seem to be the way to go. Right now it is difficult to balance: what are the applications, what is the infrastructure (machines, network, operating systems), and how do you run the current applications on the current infrastructure. This is a gross simplification, but I think it gets better through time (I hope!).

We want to make it easy for people to leverage open source to build world-scale applications without a middleware layer like maidsafe / ethereum.

At some level, everything might be described as some type of middleware. I think that the current task is to eliminate as much middleware as possible. However, there are many "legacy" (read: useful) applications running that depend on all manner of middleware. The challenge is to find the essential middleware and then expose it to everyone. At least, that is how I am presently approaching things.

We're pursuing these goals across hardware and software, looking to build a final platform with a cost around $20.00 per server per year, including hardware.

My currently working theory is to not focus too much on absolute costs. What we need is the automation to launch and operate global overlay networks that can deliver code and content to (potentially) every device on the planet. If you look at cloud-launcher.io you can see that all I attempt to do is estimate and account for the actual cost of running the current configuration. The actual costs depend on the configuration and the provider that is running the software. Currently, a user of cloud-launcher is billed directly by the provider. I do think some amount of monetization needs to occur and I have some thoughts on how to sustainably [maybe] achieve it. Essentially, I think you only need to add an additional line item that the user can modify that allows the user to contribute whatever amount of value to the project as they desire. Real users reciprocate real value.

The very tip of the iceberg can be found at klouds.org.

I checked it out. Everything seemed to work well. Very quick and responsive. Unfortunately, I couldn't connect to my tf2 server due to my tf2 client having a newer version of tf2. (If you're ever up for a game, let me know.)

What kind of team did/do you have in mind?