I am curious if anyone has leveraged the new features and design of the DB version to do anything interesting.
There are numerous facilities that would afford users the opportunity to get up to some pretty wild shenanigans, such as tags and properties being able to have properties. I am still fairly new to Logseq and have not really delved into advanced queries much, so I haven’t really tapped the potential of the new version. However, even just using it casually, the potential and power of the new design is very clear to me and piques my imagination.
One thing I am dreaming up (but have not really implemented yet) is a way to define inverse relationships among properties to facilitate bi-directional typed links. For example, if a “concept” node has the property “supports” that contains a link to a “claim” node, then it would be implied that from the “claim” node’s point-of-view (when it is the subject) then it is “supported by” the “concept” node. In other PKM systems, namely Obsidian, I would achieve this by querying the backlinks to a given note and seeing if any of them have the current note as a value for the “supports” property. Then I would have to indicate in the query what the inverse property to display in the results should be. It works, but it is not very robust. Being able to assign properties to properties makes me hope that I can reduce the work a query has to do by assigning metadata to a property like “supports” that indicates what other properties are its inverse. This also has the benefit of being able to achieve a one to many relationship when defining inverse properties of links. If I get around to implementing this kind of thing I will share it here.
What have you all been dreaming up?