[Admin note: this is split from a thread here, and the first post here is in response to this post in the original thread, but branches to a more general subject. This change is response to a user request to give the wider topic space of its own.]
I don’t think I deserve a privileged opinion on this, without having something illustrative from my own work to show. I am planning a large project with Tinderbox, however, and have been doing some relatively deep thinking about Tbx. Also, I am paying someone smarter than me to explore some of the hidden possibilities of Vision Pro, Unity, and some visualisation work from the military based on perception rather than cognition. We could have a new enabler for the spatial nature of spatial hypertext. I’ll present next month at the AAAI Spring Symposium on this.
The big constraint for us I believe is the document-centric paradigm we live in. For all the magic that the Mac has brought to our lives, most of it has been because it enhances this paradigm of discrete, static, linear presentation. The web reinforced that. The current value proposition of Vision Pro is that it preserves this windows-as-large-narrative -file-portals concept. Maybe we can subvert that.
At least in the context of this discussion, I see a conflict between the creation of what I’ll call structured medium form narratives (outlines) and structured concepts/insights across the same material. Ideally, these structures would work together, allowing one to go back and forth among a structured linear ‘narrative’ and a structured (possibly spatial) ‘situation’.
Whilst I can appreciate folks wanting more elegance in the creation, navigation, and annotation of the narrative flow, we’re up against an over-mature paradigm. My guess is that the original poster is looking for something that doesn’t matter much.
I am also guessing instead we could see new opportunities on the ‘spatial’ hypertext side.
Amen. I’m involved in a Sloan grant project that folk from the Future of Text caucus are working on and it is evident—at least to those with a hypertext background—that spatial hypertext has been waiting for this.
The above project is looking at work in XR, specifically academic reading†. A big take-away is the societal error of using PDFs way beyond their design intent. Counter-intuitively, a PDF has no (clean) copy of the (body) text. To get copy out from a PDF you essentially have to OCR print. Indeed there are whole product areas (v. expensive tools!) devoted to this and some niche academic work improving OCR of what the human eyes consumes without effort. Turns out text+semantic structure (HTML, XML) is a way better vehicle for putting text useably into an XR space.
Vision Pro is impressive (sadly my experience is 2nd hand) and is clearly a tech demonstrator/explorer for those with deep pockets (albeit in a consumer wrapping).
What stands out a mile, and perhaps obvious after the fact, is how immature the tools for this space are. More interestingly, how poor our writing tools are for writing text for remediation rather than print.
†. …which is decidedly not simply reading a linear text from start to finish. So, that should translate to any researcher, even outside formal academe. A nice hypertextual note is the ACM are letting us use their whole library for the 34 years of the Hypertext Conference as a test corpus. Quite right that Hypertext should lead where others then follow (or it may be that the Conference is so small that no one in a corner office really cares )
I am glad the Vision Pro exists. It most definitely is exploratory. My first thought was when I saw the demos was, nice screens, why am I limited by a screen? But, I guess you have to start somewhere.
Not unlike, why is this word processor imitating a piece of paper?
Yes! It’s not wrong, but limiting going forward. I struggle to write in word now - at least more than a quick letter to the local town hall—as it just has all the wrong affordances. If i want to use paper, I use a fountain pen. These days I’ll screenwrite to HTML (via Tinderbox RTF, Markdown, etc.). Then I’ll go to requested format (DOCX, PDF, etc.). I do academic writing in LaTeX not as because I like PDF but it’s a requirement and because it handles ancient affordances like footnotes and in-doc links much better than the other alternative (MS Word).
The world is slowly moving ahead. Since 2022, ACM (where i tend to submit work) now makes useful HTML from LaTeX input: example (disclaimer, I’m one of the authors there—but forget the content, look at the doc structure). Even better would be to have plaintext of the article too, minus boilerplate†, as it’s not clear ‘AI’ users of styled/OCR’d text clean their inputs but simply assume the AI will know which is what (if they even bother to consider this at all). Oof: garbage in, human-looking text out. No wonder we’re confused.
PDF’s designers are not at fault here. It is our fault using a format designed to give consistent printing, for tasks for which it was never intentionally designed. Safely remediating text requires more careful construction, in how the text/layout is encoded, than how it looks if printed onto a sheet of paper.
We haven’t yet built tomorrow’s writing tools, at least those written with a notion that the display/render context is unknown.
†. Though this does pose problems for where things like footnotes are captured, link info where anchor text != URL, and tables, formulae etc.
It’s intriguing to hear about your upcoming project with Tinderbox and the deep thinking you’ve been doing around it. Exploring the hidden possibilities of Vision Pro, Unity, and military visualization techniques sounds like an ambitious endeavor with exciting potential for spatial hypertext.
Your insights on the document-centric paradigm are thought-provoking. While the Mac and the web have undoubtedly enhanced our lives through discrete, static, linear presentation, it’s refreshing to consider the prospect of subverting this paradigm with innovative approaches.
The conflict you’ve identified between structured medium form narratives and structured concepts/insights presents an interesting challenge. Finding a way for these structures to complement each other seamlessly could indeed open up new avenues for narrative exploration and understanding.
Your perspective on the original poster’s desires for elegance in narrative flow is intriguing, especially in the context of consumer insights tools. It raises questions about the evolving nature of storytelling and the balance between tradition and innovation in the realm of consumer behavior analysis.
I’m eager to hear more about your presentation at the AAAI Spring Symposium and how your ideas continue to develop. Thank you for sharing your insights and sparking further contemplation on the future of hypertext.
)