Meeting minutes
date: 2021-11-12
Plan going forward
After discussion with the rest of the working group, added some new content to the document.
Plan for the next few meetings is to focus on producing a first draft of the algorithm, seek out feedback from impacted groups.
Also going to work on contacting Adobe to push for building the page list on generation of an EPUB in InDesign.
Impacted groups include publishers, tools vendors, retailers, and reading systems.
The Algorithm
First step is to produce a rough draft of the main algorithm for feedback.
Initial decisions, will go with a 1024 code points = "page" or locator position.
Run through each content file, generate a locator every 1024 codepoints, collect those locators into a generated list of locations, proceed until end of documents.
Open questions:
Do we start from 0 on every content document (new document, new page), or count continuously.
Do we present options to switch between those page generating options?
This algo is as much a deterrent as a benefit, so we may want to be opinionated about this to further encourage content creators to use page-list
We will not be considering CSS page breaks/ CSS object model
Concerns about encrypted content, would a RS need to unencrypt the entire document to process this location list
An XHTML parser may be required, especially if we specify that counting must be done on text within the <body>
Suggestion we built the draft algorithm with these questions included, so we can share our thought process and concerns with feedback providers
Question of what do we do about publishers that may produce retailer-specific content (would the paging be affected, we want these to be as consistent as possible, even between platforms)