Home About EIS →

Depicting Relationships: The limits of language

The heart of the english sentence (and equivalent sentential forms in other natural languages) lies in connecting ideas together and creating meaning. Like placing two portals from the recent hit sequel by Valve, you are changing the space without necessarily adding or subtracting from it. You’re using what’s already there, but rearranging it; repurposing it. Relying on a complex process of disambiguation to carry through your novel contribution to the whole of spoken or written utterances (as you learn in English grammar classes).

Have you ever considered words to be a bit constraining? I am a self avowed white boarder; I love to take a slab of potential symbols and diagrams and put things there. Able to change, able to be added to at any point. But still fixed, still temporal. Time itself is a bound on language: words occur one after the other in sequence, and despite the eloquence and mellifluence with which they can and have been applied, they do break down. At such points symbols with even more complex meaning come into play, conjuring up the various notations used in science and math to describe not only the relationships under study, but the idea as it is processed. Wikipedia’s solution is to record all changes in a log of edits, but to maintain a “canonical”, current version of the entries visible to users. It hearkens back to a world of order which curated, authoritative knowledge could be contained in neat entries with headings; where you could browse from one topic to another, and eventually the network of relationships, the implicit model of the combined understanding of the contributors would coalesce and you’d go, “Aha.” Online forums are littered with issues of contention where few such epiphanies are found.

There’s been a number of interesting approaches to this limitation — from a technological standpoint, at the very least. Mindmaps are easy to draw on paper or a whiteboard, and include creating ad hoc relationships and heirarchies between words representing ideas. Although this surpasses the linearity issue, space and the ambiguity of words leads these to be often little more than outlines and hooks into your memory. All relationships are embedded in our interpretations of those lines. A great example of this in a more interactive medium is the visual thesaurus. A more utilitarian and extensive variation on the idea is present in org-mode ( — which I’m using in writing this and heartily recommend) and freemind.

These are technologically extended mindmapping utilites that not only organize words, but associate text, code and even meaning to nodes. But ultimately, concepts are equal participants in multiple different relationships at once, and capturing the complexity into a format easily tractable and understandable in a single screen is hard.

What have you done to address this in your own work? In the complex endeavor of scholarly communication and idea development, are long form journal articles and synchronized google docs and whiteboards really the best that we can do? Is there a way to link one’s knowledge, one’s individual representations into and from ongoing discourses? Microsoft has placed it’s chips on the table (the enhanced, expensive kind) with the Microsoft Surface and its soon to be released successor. But what about both cases — when you’re alone with your idea, when they need to escape your head and be considered in some other forms — how do you see these tools evolving?

About the author:  John is a PhD Student at the Expressive Intelligence Studio researching natural language generation and authoring tools. Read more from this author

This entry was posted in Academics. Bookmark the permalink. Both comments and trackbacks are currently closed.


  1. Jon Saklofske
    Posted May 27, 2011 at 9:34 PM | Permalink

    Great post! I’ve been working on a visualization alternative to exclusively using the written word for critical note taking, idea forming and exchange. It’s called NewRadial and is an open source project that I’ve put up on sourceforge. It was originally designed to work with William Blake’s illuminated pages and to enable the addition of single or multiple user critical commentary and exchanges that are visually mapped onto the workspace via edges and groups of nodes. The tool is far from finished, but any iconic content can be substituted for the default Blake stuff. New radial can be found here: http://sourceforge.net/projects/newradial/

  2. Posted May 28, 2011 at 2:58 PM | Permalink

    Hi Jon,

    Very nice! William Blake’s combination of words and visuals was inspiring for me as well. I love how you are able to display and link the page images themselves. Stay tuned! This blog entry was one of several leading up to a tool that I’ve been developing that I think you’d be very interested in as well.