We search for resources such as titles, images, and assets by various means that may or may not recognize that a page that needs them has moved around the federation. Can we formalize this process and support that consistently with resource resolution in the client's core?
In this issue, filed in 2018, we note that the Frame and Asset plugins differ in how hard they try to find the same html file. github ![]()
The suggestion then was to "consult page context when resolving asset references". This would place additional burden on one plugin to match the resolution process of another. Can we do better?
We wrote in 2016 that history based link resolution was the secret-sauce that make our approach to decentralized cooperation unique. See Collaborative Link
In 2023 we added Page.context to our Java reference implementation and then revised it when we found it inconsistent with our vaunted practice. github
, github ![]()
This visual map shows how wiki pages move within and between sites and how their Halo changes as they do.
Exploring async fetch in the Graphviz algorithmic drawing. I've launched as many as 450 simultaneous fetches and gotten good results. Amazing.