Gum-ball Coin-op

[brian-lc]

Concept

Make compute/storage/any simpleIO service as simple as a gumball machine. Coin goes in, gumball comes out.

The Problem

Developers of simple web based services are forced to create large/difficult structures to charge for them. This project would serve by creating basic primitive to charge for usage. The concept is similar to ethereum gas but is a way to apply that same concept to any composable service.

How does it work?

1. Develop a service you would like to monitize
2. Add the coin-op shim/side-car process can track usage by account
3. Register service and define/document it's API and rate structure
4. Create OpenLaw contract for SLA, liability, security of service using boilerplate
5. Publish service and advertise it's features to other service developers (through a central app store like "API store")
6. Other developers/users can pull in API definitions into per-language/framework wrappers for easy RPC-like calls in their projects/services.
7. Other developers/users provide eth address to wallet, as they use published service, wallet balance is drawn down based on usage.
8. Wallet empty? No more service provided.

Existing Work

Unclear. This is essentially a new way of metering and charging for services ad-hoc. There must be others doing this or something similar. In theory this is better because it lowers the bar for monitizing a useful service in a way that doesn't require forming a business entity.

Fit and Fitness

The idea is that this expands the potential users of ethereum. It goes after the long-tail side of webservices. The ideal service candidates are ones that are novel and useful but are not right for building a business around. With this system a developer can launch a service and charge for it right away. No need to provided services for free because setting up business entities to accept payments require a lot of work. The centralized registration service provides a channel for marketing and promoting the service similar to an app store.

Assumptions

• OpenLaw can provide a contract that provides limits to liabilities similar to LLCs
• Service providers can effectively meeter their services.
• Customers are okay with using services run/created by individuals and not businesses.
• The registration system is centralized and can provide some way of vetting API's/developers to prevent malicious behavior.
• Side-cars/shim assume some kind of docker/k8 host so the services need to run on some cloud provider.
• Data processing can be encrypted end-to-end using customers public eth key.

[brian-lc]

Look into use of microchannel payments for settling up usage https://solidity.readthedocs.io/en/develop/solidity-by-example.html#micropayment-channel

[brian-lc]

Some kind of "app store" or index might be needed. Looking for more examples of something like this http://apis.io/ Following the sponsor links leads to some 404's. Not sure if/what is still active.

[barlock]

Twitter Survey: https://twitter.com/jwolpert/status/1114998520203939842

[brian-lc]

Someone building microservices following these principles would produce services that are good candidates for 'gumballizing' them... https://medium.com/@jbgisbert/microservices-overview-30e505316a8

[brian-lc]

Turns out this idea is so good that other people have been working on it for quite a while 😸 https://docs.iota.org/docs/getting-started/0.1/introduction/what-is-iota

[barlock]

👏