liquid thoughts




The basic premise of this exercise


The basic premise of this exercise is to work to find a path to continually improve the power of the written word as a medium for transferring human intent from one person to another.


We need to develop a gamut of possibilities and directions to invent in. What follows is from my own deeply personal perspective and I very much invite comments.




The aim of this work has to be to improve comprehension for the reader. This reader can be the same person as the author of course, but the aim must be to produce a readable document, rather than focusing on a quick way to write or a neat way to display.




If we take it as a basic premise that interaction is at the heart of what information is, and that we have much higher bandwidth visually than simply black marks on a white background utilises, then interactive visualisations make a lot of sense.




Please note that this is is about creating an interactive document, not a collaborative writing system, a dialogue system nor a task management system.


The Textual Atom


I also believe that the atomic unit of textual discourse should be a human thought – not a word or a sentence or a paragraph necessarily, but whatever length a human thought is.




My friend Dino has developed a philosophy called Polyscopy and together with Sam and Stan, we are working on ways to best make the thoughts and perspectives of Polyscopy available to various readers, so that forms the ‘case study’ contents for this work.


You can see Dino’s map of Polyscopy at and a blog post here: Polyscopy Platform Design.




The map Dino has created is the starting point for our common exploration. However, while I see it as crucial that we not only improve visual communication for the foveal area of the human eye, the sharp centre, which is only about 1.5mm wide, while taking in information in a linear way (as with traditional reading, which is in one direction), we do need to take into account ways to improve how the wider visual field can yield useful information.


As opposed to linear text, mapped text is useful to show relationships between three or more items which do not all correspond linearly.


Chocolate -> Cake -> Plate -> Tummy is an example of a linear logic which maps into a regular sentence.


Mapped text has inherent relation-opportunities of up down, right left, as well as size indicating in out, which works great for charts with a time dimension, which we naturally read from left to right, same as we read text. I feel these dimensions should be used when useful, but not where linear reading is more efficient. A key feature of a richly interactive document system will likely include the ability to move between different views, including traditional (with denseness of text) and 2/3D space (with clarity of connections).


Some information needs more than a short sentence to be made clear so mapped environments will need the ability to expand and contract to take advantage of this.


A Word is Worth a Thousand Pictures


A note: Working towards a richer visualisation does not mean dropping the last 5,000 years worth of work on grammatical symbols – words. The work I am talking about here will take the word as a first class citizen and work to expand its capability and effectiveness.




Icons are great for overviews in both traditional text and in maps, when clearly understandable but can be intrusive in traditional text where they take more attention than the rest of the document. This calls for the importance of being able to change views dynamically.


All Information Is Structure


There are at least two categories of information – information about something which exits physically in the world (or which is planned to exist in the world), such as geographic information, which will have it’s own ‘natural’ matrix structure of how information relates.


Then there are ideas about the world – human expression, such as attitudes philosophy, which does not have a natural, un-changing matrix – the taxonomy, the structures of the relationships are in natural flux.


In both cases there is no natural boundary to the information – the size of the structure needed to understand any given aspect – simply enough of the structure for any given purpose.


As such it is crucial to consider how information is connected and how best to give the reader access to (interact with these connections (to follow, view) with the aim of augmenting the reader ability internalise the structures, to understand them.


An N-Dimensional ‘Space’


Taking the starting point that all information is structure, we need to try to step away from our worldly 3D + Time visualisation. Ted Nelsons ZigZag is an inspiration for going further in many dimensional spaces.

Investigations into a Polyscopic way of seeing