Lots of complex issues raised in that discussion. I think your instincts franzap are on the right track, assuming I followed the discussion correctly.
I firmly believe that WoT won’t work unless it’s based on explicit trust attestations. Trust proxies like follows, zaps, likes, replies, and the like are a crutch which have limited utility and MUST BE ABANDONED. We will not make progress until we accept this thing that we all know deep down in our hearts to be true.
Explicit trust attestations won’t work unless they’re context-based.
Context based trust attestations won’t work unless we solve the UX problem which stems from the fact that the number of contexts has no limit.
The UX problem won’t be solved unless trust in any given context applies automatically to all subcategories.
For trust in one category to apply automatically to all subcategories, we must represent context as a hierarchy, with parent-child relationships.
The hierarchy of context MUST ultimately be curated by the WoT, not by devs.
For a site like yours franzap, a store with products, the hierarchies of context would include things like: Alice trusts Bob to provide reviews of zap.store products in the category of electronics, with smartphones being a subcategory of electronics. As the app developer, at first you would create the hierarchy of categories yourself. Eventually, the hierarchy of categories would be curated by the Grapevine. The tapestry protocol shows how to do all of that, in detail. I’ve already used it to curate lists, and curating a graph is going to be conceptually not much harder than a list.
But we need to take baby steps, which is why at first, you as the dev should manage the hierarchy of categories yourself. There are other ways to break it down into baby steps from a dev perspective. I’d love to brainstorm with you and hammer out details on how you could do that with zap.store if you’d like. 😃
