It'd be nice if a majority of websites were exposed like this: if you wanted to save an article or blog-post for example, you could just fork it into a reposito

It'd be nice if a majority of websites were exposed like this: if you wanted to save an article or blog-post for example, you could just fork it into a repository. What's saved is just the original text, or anything that has a flat representation. Then you could edit and re-publish with changes; I think this is more in align with how knowledge works, sort of an implicit graph with splits and merges. The basic idea is to separate the content from how it's rendered (html, css, js). How content is rendered can be a repo-by-repo decision. Benefits of this is better portability of data and easier way to organize things on the internet: you wouldn't have to web-scrape with some type of dom parsing heuristic library since the content is readily accessible. It's also feasible in implementation: plenty of people are publishing via Github pages and Jekyll; I just haven't seen the ability to fork posts yet. Bloomberg Beta definitely could've done this for their website.