I’m developing a project where users can design their own productivity workflows. Each workflow is built from a graph of routines, where each routine can link to an interface, smart contract, external API, etc. For example, a user wanting to create a routine to mint an NFT should be able to paste the smart contract, select an interface, specify the standards that the contract’s inputs should follow, and that will generate a webpage to let anyone use their minting contract.
Ideally, developers who design the interfaces won’t have to worry about hosting. They can either send money themselves to a wallet that handles the costs, or let users pay for it using microtransactions.
I did some research, and it seems the only way to accomplish this now is with Akash. To host an interface on Akash, you must create a wallet, put in some initial funds, create a deploy configuration file, then trigger an automated bidding process. As long as the wallet has enough funds, then the interface will continue to run. I figure that the setup process can mostly be automated with a script, so it all seems perfect.
EXCEPT that Akash is on Cosmos, so the Ada paid by users would have to go through the Cardano<->Ethereum bridge, and then the Ethereum<->Cosmos bridge. Obviously not ideal, and very expensive.
NuNet sounds like it could be the solution, but I was not able to find any information on how the deployment process works. Specifically, I would like to know:
- Is it feasible to host interfaces using NuNet? What about the backend services, such as Cardano nodes?
- Are the costs expected to be greater/less than Akash? What about AWS?
- What would the general process for getting a service running look like?
- How long would it take to run a service? Could interfaces that are needed within a few minutes be spun up on demand, and be ready when the user needs them?
I understand that NuNet is in private alpha, but my project is also in the very early stages. If what I’m asking for won’t be available for several months, I’m okay with waiting.
I can’t answer your questions because as you say, it’s too early, but I will suggest you propose your project for DEEP Funding with SingularityNET. If your proposal is voted in, you’ll get funding and also help integrating your service into SNET and eventually be able to run on NuNet.
Thanks for the info. I will submit a proposal😁
One thing I would say in relation to your timeline of several months is check the NuNet Roadmap timeline. e.g. Public alpha is a 12 month time frame that hasn’t stated yet. Until there is compensation, reputation and penalties for NTX miners the service would likely not be stable for something hosting a financial transaction front end. Not saying its not going to happen because I think it will i just think the timescales might not fit with your project as it stands.
Thanks for pointing that out. You’re completely right, my timeline is a bit optimistic. I guess you could say I’m running on Elon time😅. But my project can still work without NuNet in the short-term - it’s just a matter of how decentralized it will be. My original plan for Akash is to manually send funds to the wallets until I can figure out how to best solve the Cardano<->Cosmos bridge situation, so already there is centralization involved. I don’t expect there to be many different types of interfaces at first, so worse case I could settle for using Kubernetes or AWS Elastic Beanstalk, and do all the setup/maintenance myself.
The main focus for this year is not on full decentralization, but rather making a useful rapid prototyping and workflow-building tool for Project Catalyst proposers. NuNet is not required for this, but it is required for my long-term vision of decentralizing the means of production. I believe this is required for adopting new governments, since centralized production is a huge attack vector that incumbents can use to keep their power.
@MacForaday @samski If it interests you, I just submitted a proposal for this which explains my game plan in more detail.
Awesome! Best of luck on the voting