In an effort to develop new forms of reading, I’ve built a minimalistic Bible app (only on iOS for now), called John 1:1. The Bible is well-suited because it: (i) has a history of driving reading innovations and revolutions (e.g., 2nd century: scroll -> codex; Gutenberg in Mainz, 1454: codex -> movable metal type; Estienne in Geneva, 1551: introduction of verse numbers); (ii) remains popular and widespread, with public domain translations in many languages (the app ships ten translations in seven languages); and (iii) is intrinsically self-linking and annotated over millennia (31,102 verses; +500,000 cross-references). Few texts have such a dense pre-existing edge list.

John 1:1 tries to extend the Bible’s arc of reading innovations to and through the mobile device by organizing verses in a swipe feed that can be navigated vertically and horizontally. Navigating via these two axes enables two distinct forms of reading: linear and lateral. To read linearly (Genesis 1:1 -> Genesis 1:2) a user swipes up, and the next verse appears below. To read laterally, means to read verses that are connected or related, but not next to each other. Swipe left from Genesis 1:2 (“formless and void, and darkness was over the surface of the deep”) and Jeremiah 4:23 appears from the right, with that same, rare language (“I looked at the earth, and it was formless and void”). That swipe (Genesis 1:2 -> Jeremiah 4:23) crosses +19,000 verses or ~ 61% of the Bible. One could read the Bible linearly for decades and never realize that Jeremiah 4:23 echoes Genesis 1:2. Lateral reading can therefore reveal a hidden structure, and meaning, that cannot be detected through linear reading. The app ships with 292,374 lateral connections. The key challenge has been how best to score “a human will see why these two things are related” at build time. If anybody knows or has a good idea, please tell me!

John 1:1 is free (no ads, no accounts, no in-app purchases, fully offline) and there are no plans to monetize it. However, lateral reading can be applied to any text with addressable units, a pre-existing edge list, connection logic, and ranking. And there are many promising applications in law, academia, or any other domain in which humans will continue to closely read complex texts and benefit from a tool that makes hidden information visible.