browsing a well sorted library is rarely a random walk is it? usually you go to a specific section _because_ it will contain a plethora of views and ways to e

browsing a well sorted library is rarely a random walk is it? usually you go to a specific section _because_ it will contain a plethora of views and ways to express them, in that section, and the librarian, if asked, will guide you to the core works and what to read first to get an overview, or has them singled out in an extra shelf or table. and in addition, your skimming is not page limited. discoverability is the second, subordinate case the article makes, though. persistence is the even bigger challenge: how do I know the article I've recommended to you is the article you're going to see? not just some little edition and helpful factual correction, but a subtle rewrite, twist in tone? or even worse, differently tuned articles depending on who we are, as identified by, say, our cookies, we see a 'libdem' and a 'repcon' version of the same article under the same headline? and we'll never know because your device shows one, my device shows the other one?