I personally have an issue reading long, grey columns of text, such as this from ACM Hypertext. Compare it to the very first substantial printed document, the Gutenberg Bible–which even includes red highlights of key letters–and it becomes quite sterile:
The perspective is that not all academic papers are worth reading every word of so it should require no more effort than absolutely necessary to get the gist of a paper and when deciding to read it deeper, to get the layout of the paper. Additionally, when reading something outside of the user’s field, further support can be useful (see below), not just a different layout.
I have therefore been experimenting with ways to make the text less uniform and more visually distinct, in dialog with Bob Horn and Jim Strahorn.
One approach is making every other sentence bold (as Ask AI example here when reading):
Another approach is to bold key terms in each paragraph (when authoring, exporting or via AI, as below):
Re-flowing the text to space out sentences is also interesting, something I call a ‘Focus’ view in Reader:
All of these which are applied with the ‘Ask AI’ command (chatGPT) take a few seconds to process but users can choose to ‘Save Metadata’ to have it instantly appear in the future. Below is an example of Flow and Bolding which does not make the page more readable. Further studies will happen.
Reading New Types of Material
When reading material that is important to the user, but is outside of their main expertise, layouts can still matter, but supplemental material can also be useful. Take this study on reading on paper VS. digital and–to me at least–there were terms I did not understand and which Google searches did not clarify well:
I ran an Ask AI prompt asking for an ‘Academic Introduction’ with supplemental information and this was presented, giving me a ‘way in’ to further study this paper:
Here is a Focus & Bold on that same paper:
I also spend quite a bit of thought and time on the colors of text, primarily in my Author, rather than Reader, since I have more control, something I hope we can also do in XR:
The next stage is to use the spacial affordances of XR to extend such experimenting into how to make text highly readable.
Grey, not Bold
I’ve added one more article to the mix, where I have used lightening/greying for non-key text instead of bolding for key text, since I think bold is very heavy and not great to read. What determines what should be grey and what should be black is a discussion in itself, with perhaps the author bolding text and choosing on export to convert bold into regular text and non-bold into grey.
Marking Text
This is similar to how in Author the user can choose to Mark text (make it bold orange) for their own skimming when writing, and have this Marked text converted to bold text on export, or kept as bold orange, or converted to plain text or even removed entirely (if the user employed marking for their own notes).
This has allowed me to use the Journal in Author (just a regular Author document but accessible instantly through a keyboard shortcut) since I can Mark text I should action, to easily see it again when skimming the document, or when folding the document into an outline, since Marked text appears there as well. When actioned, I simply un-Mark the text.
Vint Cerf suggested that maybe the reader could be interested in what the author marked, so maybe it should be embedded in the document metadata to be seen on demand.
Hide/Reveal
This very much resonates with my thinking on hiding and revealing in Maps. A single Map (concept map etc.) cannot contain all the information relevant to the user so there needs to be mechanism to connect Maps so that a hierarchy (not linear or one way) can be established and elements can be hidden and revealed instantly at will.
In a written piece this reflects in me (text in brackets) and subordinate clauses as well as sections in paragraphs which primarily serve as context to the uninitiated in the field or binders for flow. All necessary and useful when reading straight through but not useful for navigation or for skim reading when one is familiar with the context and the work.
I have therefore also experimented with notions of collapsing brackets as a kind of stretch text like this () where clicking on the empty brackets would expand to show what was inside them. Since my current exports are generally to PDF this is not very useful but could work in HTML. However, I feel they are quite visually ‘loud’ and I’m therefore experimenting with other means to do such folding/hiding and revealing. I have written a few alternatives in my Author Journal:
Examples:
Folding bullet lists into showing just the first bullet with one or two words followed by an ellipsis:
• First point…
Folding bracketed text with double brackets on the right, to indicate click:
This would be normal text and you can see that you can click on the following to expand to see the full bracketed text (extra)) inside a sentence.
A dagger is used in Author to indicate an Endnote which is turned into a superscript number when exporting for reading, which could be another way for the author to add further information:
What follows is what we called an Endnote but should probably be called a Popupnote† at this point.