How Web Reader Modes Modes Work
Daniel Aleksandersen (via Ricky Mondello):
This article is part two in a series on web reading mode and reading mode parsers.
- The non-standard rendering mode
The history of reading mode, a look at the different parsers we have today and how they came to be, and a small criticism of the Apache 2.0 license.- Determining the main page content
There are many approaches to content-analysis and extraction, and most only work well with English-content. Why is reader mode so slow to activate, anyway?- Title, author, and date metadata extraction
Visual page inspections, standard metadata, or guesswork? Everyone has their own ideas about how to best determine the metadata describing an article.- Inconsistent and bad reading experience
Encourage publishers to fix their designs, and standardize reading mode now.
I went on a bit of a journey exploring how our websites are experienced across these technologies and how we can make sure that our content is interpreted the way we want, regardless of how people are consuming it.
For demo purposes let’s look at what a simple HTML page looks like in Instapaper, Pocket and Safari’s Reader mode and compare
<div>only content and semantic HTML.
