Andrew, Stephanie, Scott, Lisbeth, and Jessica Respond
You might have noticed that the blog software I’m using for this page doesn’t support comments. Andrew Stern, Stephanie Strickland, Scott Rettberg, Lisbeth Klastrup, and Jessica Pressman (Associate Director of the Electronic Literature Organization – can’t find a good link for her) have all written in recently via email with various-length responses to my post proposing a definition of electronic writing. Andrew also responded to the one about what I called “dynamic text” (and proposed some better names – see below).
To recap the initial post briefly, my proposed definition of electronic writing was: “Writing that requires computation at the time of reading.” Andrew’s response was basically positive – saying it was concise and appropriately inclusive. The others prodded me in a number of ways. Here are a few that I’ll respond to in the next paragraph: Stepanie and Scott both asked me to further consider the case of things like computer-manipulated poems that end up printed on paper, Lisbeth asked how my definition is different from Espen’s, and Jessica asked what my definition accomplished.
So, in response, here’s what I’ve been thinking the last couple days. There seem to be three interesting categories of work here:
- Espen’s “ergodic” – which doesn’t have to be electronic, but requires certain types of non-trivial reader actions and a text-producing machine.
- My “electronic” – which requires electronic computation at the time of reading, but may require no greater reader activity (or even less) than print writing.
- [the case that Stephanie and Scott discussed] – which includes writings that are non-electronic but in their production have used computational technologies non-trivially.
What do these accomplish? Well:
- Espen’s, to me, seems designed for analyzing texts – and creating types of analysis that don’t get confused by whether there’s a CPU involved.
- Mine is probably a little more prosaic. Mine is for figuring out what to teach in an electronic writing class, or what an organization like the ELO should be working to support and promote.
- Stephanie and Scott’s category draws interesting attention to a group of works that Espen’s definition and mine marginalize. The tradition of things that connect with Burroughs or Jackson Mac Low or many of the Oulipian projects.
Stephanie and I also talked about the fact that my definition might be improved if the term “requires” was replaced by the words “written for.” When I mentioned what I thought my definition might accomplish in an email to Scott, he wrote back and agreed – saying perhaps it should guide what is included in the ELO’s Directory. Jessica also pointed out that I should be clear that I mean electronic (not extraordinary human) computation. (Nick and Markku also have a useful, controversial introductions to Espen’s part of this. However, I’m disappointed that both write as though the term “hypertext” was defined by George Landow rather than Ted Nelson. Not that this is uncommon…)
On a different topic, Andrew wrote at more length about my “What is Dynamic Text?” post:
In Facade we’re attaching pieces of text/content (a sentence or part of a sentence, facial expression/gesture, arm/body gesture, a short bit of action) to particular places in performative, reactive, responsive behavior code/plans, of which there are a large variety. Each of these behaviors+text are intertwined together into collections at varying levels of granularity (“beatgoals”, “beats”, “beat clusters”), each of which are annotated with preconditions and effects — effectively a hierarchy of story content, dynamically sequenceable at all levels. These preconditions and effects end up defining a partial ordering — that is, not allowing any and every piece of content to happen at anytime, but to allow a sizeable subset of the total number of pieces to happen at any one time. In effect, the architecture of story content allows the system to assemble individual phrases, sentences and sentence collections in a coherent (narrative) way in many possible orders, coherently intermixed with the player’s own text (which is natural spoken language + gesture, such as looking at a character and saying “why did you marry this guy?”, not interactive-fiction command language like “open door”). The system’s text pieces and the behaviors they are attached to are authorially crafted in such a way as to allow for coherent assembly in many different orders. So, yes, that’s dynamic, but dynamic is too general of a word…? More specifically it’s “assembled text”, or “constructed text”, or “recombinant text”, implying that small-grain-size (human written) pieces are being assembled on the fly, intermixed with the player’s text.
Of course this is good clarification regarding Facade – but I think it’s also good clarification for all the things in that post. Not that all of them (from John’s transliteral morphing to Mark’s Card Shark and Thespis) operate in the same way – but a term like “dynamic” doesn’t capture what’s interesting about what they do have in common. It almost sounds like the important thing about them is that they’re made with Flash and jiggle. Maybe “recombinant” is the word? Any comments on the comments?
On a related note, Andrew will be speaking at MIT on December 2, from 12-2pm (Building E51-275). He’ll be presenting Facade at one of the “(Evocative) Objects” Lunches put together by the MIT Initiative on Technology and Self (which is directed by Sherry Turkle). It’s open to the public, according to the email I saw.
Finally, this manual response-management is helping me understand the need to work with some blog software that supports comments, and probably trackback (two-way links – it feels almost like Literary Machines). Maybe I should switch to Moveable Type for this blog – as that appears to be what I’ll be using for my electronic writing workshop next semester.