So now that we have a starting point, all we need are five different relationships to explore and then we can begin fleshing things out in actual code and testing these ideas on the actual blockchain. A few posts ago, we talked about three different relationships. One dealing with an author's most recent work. One dealing with an author's most recent like. And one dealing with an author's great proportional support towards a piece of content.
Another interesting relationship to explore would be something like the most recently resteemed author's most recent post. In some cases, this would simply give us the most recently resteemed post, in other cases, we may see new content from an author the current author liked in some point in time. This allows a little more flexibility as opposed to simply the most recent resteem since an author may not resteem very often and thus such content may be older than 7 days old.
The last relationship I'll consider is to select the most recent piece of content of a voter of the current piece of content which the highest voting percentage committed. Since everyone commits VP this method is a fair way of selecting a voter that leans slightly towards smaller users who don't yet have the option to make partial votes. This should exposure readers to other readers like themselves who happen to both come across the same piece of content.
But what if these relationships don't turn up any results? We are going to need some way to back fill the results. We could simply use some of the other relationships if they turn up multiple results. We could use backup relationships to fall back upon in the case that some relationships fail. Or we could simply show less results.
These are all decisions that have to be made in the future, but luckily for us, they do not need to be made now. But now we have a solid foundation from which we can begin building. We have five relationships to begin testing with. We have a way to initialize the entire process. Now we simply need to begin building the project through experimentation and code.
Moving forward into the next couple weeks, we'll consider different development options and libraries available to us. Start looking at the blockchain in code and begin implementing these relationships one by one. I'll even start adding pictures and diagrams to make things more interesting. But for now we have an idea and a motivation. Now we need to make it a reality.
