How easy will it be to build apps via cloud services like AWS and then integrate with Holo/Holochain post beta launch, compared to starting building on Holochain/Holo now? Will there be an API for existing apps like Facebook and Twitter to integrate with Holo and be hosted by it? What about online marketplaces and other apps involving economic activity?
Hum good question/conversation…
My honest answer is I don’t know, what my intuition tells me is that when it comes to data structures, migrating an existing app designed for web2 to be integrated into hololand is going to be a big big effort…
Regarding app design and data structures: many holochain applications require tricky patterns regarding signatures, hashes and so on, and if the web2 app design doesn’t include those patterns I think it would be very difficult to migrate. Example of a tricky pattern: we are developing a ticket zome to be able to join a network, and all the interactions behind that: get the product hash and send a direct message to the seller, the seller bridges and then signs the hash of the ticket, you store that signature alongside the ticket and provide it when joining the new network.
Regarding holo architecture/design: I think that the architecture that holo will support will be publishing a js library so that the users communicate directly with holo and call the holochain zome api directly from their frontend (see chaperone). I think this is mainly to guarantee security on behalf of the user, since the private key is generated on the fly in the browser, you won’t have it in your AWS service. If you want to go another route, maybe it will be more difficult without that support.
I think that main integration between apps is going to be done the holochain way: bridging, cloning dnas and joining networks. And also integration with other systems by “soft bridging” on the frontend. We’ll see though.
Happy to continue the discussion!