Welcome, Guest. Please Login
IMPORTANT MESSAGE! This forum has now been replaced by a new forum at http://forum.eastgate.com and no further posting or member registration is allowed. The forum is still accessible via read-only access for reference purposes. If you wish to discuss content here, please use the new forum. N.B. - posting in the new forum requires a fresh registration in the new forum (sorry - member data can't be ported).
Pages: 1
Send Topic Print
Situated, Deep Narrative (Read 2840 times)
Ted Goranson
Full Member

Posts: 141
Virginia Beach VA
Situated, Deep Narrative
Jul 23rd, 2011, 11:27pm
It is probably time to announce what I am working on, though it may be some time before I can share the whole thing.

I am a computer/cognitive scientist, aspiring to understand and field a new class of logics that could enable machines to better understand narrative. Narrative in this use is larger than ‘story,’ referring to how we naturally structure and store information. When we recall that stored structure in order to discern/create meaning, it builds stories, apparently according to some principles we can discern.

We may be able to emulate, even anticipate these structures in a machine, both in how they are stored and how they generate meaning. (After all, a book is simply a machine.) I think we can go a long way in this with ‘reactive’ typed links and new results in the mathematics of logic.

Tinderbox as it happens is possibly capable of some meaningful experiments in this space, both because of its capabilities and the community attracted to Mark. I have a three phased project that Mark A (and others) is helping with.

The first phase focuses on building a workable architecture for structured authoring and publishing to the web. This has been a year in the works already, but should emerge soon. Strictly speaking, it is a simple matter: author in Tinderbox and export within the profound constraints of html.

The challenge is to set things up on the Tinderbox side to be friendly to the experiments in link-based structure. I have some ambitious starting points. Following (in the most general sense) the Tinderbox Way, I will discover local structure and general structural principles.

On the web publishing side, we plan two innovations. One is a way to handle the different types of links, primarily leveraging a few modes of stretchtext. The other is to have a demo of “semantic similarly” in note matching rather than the more common lexical similarity (via word and phrase matches).  

The second phase is to populate this with reflexive content: essays and examples that explore, describe and discover useful structure. Because the links cannot be simply static, we will have to create a partner application, accessed by shell scripts. The site will have four distinct areas: a somewhat traditional blog, essays on cinematic narrative, essays on the form of logic (with a focus on visualization), and essays on computer implementation. These should both describe and illustrate the ideas. The first two essay groups are likely to be openly collaborative.

The third phase implements a full up graphical modeling tool that we think we can implement in map view, at least so far as the UI. This should structure some large very challenging and dense narrative examples.

I guess a word on the philosophy of narrative is due here. The usual notion is that there is a reader and some text. The reader threads through the text building (or if you wish, discovering) narrative through what has been encountered before in the text and in what order, as well as what has been encountered in the surrounding life (and in what order).

Traditional hypertext expands this horizon because it theoretically allows greater possibilities with the order and parallel interacting threads. We focus as well on the agency within the text. ‘Everything causes something’ is the best way to describe this, with the causal, transformative influence somehow captured by (probably thousands of) internal links.

This vision of reading (or seeing a film) is a dance between the agents of the text and similar agents the reader brings. The paths of influence within the text are as significant as the paths the reader follows.

It will be a fun experiment, and if it fails for some reason, it will be a massively interesting failure.

Back to top
WWW TedGoranson   IP Logged
Pages: 1
Send Topic Print