A fascinating story about Flores in London after his escape from Chile post-Allende. he was a formidable personage in his times and places. He has been in Chile this decade as I understand, after disengaging his seminal firm Business Design Associates at the end of the last decade. While not the inventor of most of the work he’s known for, Flores was the innovator, in the sense of creating products (Coordinator) and systems (independent educational programs) that advanced and embodied these ideas. As second order cybernetics suggests, we need to re-enter our practices with our own theories, intervening rather than observing to understand.

A journal like Interactions is primarily practice-focused, so my pique about the miss on Flores was overtstated. It doesn’t matter that much in scholarly terms, because you’re spot on about the precedent. But I’m more interested in legacy, as perhaps you are. The Coordinator just has a huge legacy for its audacious early entry into constructing an interactive world of its own conversational theory. Although there was plenty of heat discussing it in the ACM world – mainly after it died – We also should have learned more from its innovation than we did.

Our current email ecology is a weak transaction oriented system that we have adapted to, rather than adapting to conversation. (And gmail is not that innovation we were waiting for either).

Pask was a deep influencer and should be cited more than we find in published articles. I’m reviewing second-order cyber papers for another piece and I do notice where he’s missing now, where I would have not before this discussion. But then, Jantsch, Ozbekhan, and Dervin don’t get cited for their early influence work either. Its helpful to notice, their ideas stand the test of time.

But Pask was doing conversational stuff earlier. His 1961 book (An Approach to Cybernetics) is full of reference to conversation (I have published on this, but only in German), and his earlier work is interactive in a much richer sense than almost anything since: a manner that is truly conversational (conversation being essentially interactive).

Pask wrote about the use of conversation in a similar manner to some of the use in this paper, in “The Architectural Relevance of Cybernetics” (1969).

Pask and I discussed design over the years from 1967 until his death, and played with several ideas. I have published on the centrality of the conversation with the self, as being at the heart of design. But I didn’t publish this till later.

There are always precedents, And there are always precedents that occurred before the precedents I (or you) cite. It’s nice to have Winograd and Flores brought into this rather under-referenced piece above, but they really aren’t the origin and it’s a shame that their inclusion is proposed in the manner in which it is.

Wonderful: a conversation about conversation! Thanks for posting.

You have us thinking about new possibilities: - We’d like to solicit an article for “Interactions” on the history of conversation models. Would you be willing to write one? What models would you (and other readers) include? - We’d also like to solicit an article looking at “The Coordinator” 20+ years on. We’re hoping Terry Winograd might be willing to write that.

Now to the substance of your post: We accept your point that our article should reference Winograd and Flores. That’s an oversight, at least partially corrected below. Your further suggestion, however, that our article “re-invents and reformulates” their work misses the point of both their work and ours.

Winograd and Flores work is important, because it offered a new way of looking at “computers and cognition”, challenging the flawed orthodoxy of artificial intelligence. Within that context, the authors also developed a model of the possible sequences of interaction between participants in what they termed “conversation for action,” essentially a highly constrained (and somewhat mechanistic) view of coordination, itself one portion of the larger space of conversation. The authors needed to take a narrow view of conversation and coordination in order to model it in software, “The Coordinator”, another important achievement in their work but one limited by the software systems of the time.

We acknowledge that The Coordinator was an important advance, but we advocate an approach to “designing for conversation” which differs in at least three fundamental ways: 1) A robust model of conversation must include participants’ goals. 2) It must also acknowledge the central role of processes that build understanding and agreement. 3) And it must, at every level, allow for the possibility of true interaction (e.g., negotiation of goals and actions) between participants.

Of course, that’s the essence the models articulated in Pask’s “Conversation Theory.” Here, we should further acknowledge the shared roots of Winograd and Flores’s work and that of Pask. Both are grounded in the second generation of cybernetics. Both reject notions of “objectivity;” both embrace the subjective role of observers and frame observers as participants; both owe a debt to the work of Heinz von Foerster.

For readers interested in more information, please see Winograd, T., and Flores, F., Understanding Computers and Cognition: A New Foundation for Design, Addison-Wesley, Menlo Park, 1986.

• Hugh Dubberly (for Paul Pangaro and Hugh Dubberly)

While I applaud interaction designers investigating communicative theory, robust mental models ought to embody the debates of the scholarly literature. So relying solely on a single notion of “what is a conversation,” limits what might be a very interesting approach to building interactivity.

Good work in pushing the boundaries of what interaction design ought to aspire to. I would encourage further theoretical debates to be embedded within actual designs.

However, interactions IS a professional journal. How could an article on designing for conversations not at least cite and reference Stanford’s Winograd, Berkeley’s Flores, and their work on the Coordinator from 1986? They not only wrote a book and peer-reviewed stuff, they built one of the first commercial email systems that embodied these principles (The Coordinator). There were debates for years in SIG CHI around that. And interactions is a SIG-CHI publication, yes?

And as scholars, their treatments cite and rely upon their sources for the conversation theory they embodied, Berkeley’s John Searle and his intellectual forebearer Austin.

I’m adding this comment because I believe it is critically important in the rapid-fire social media age to retrieve authenticity in our discourse that honors the legacies that led us here. Take an older friend to lunch this week and find out about their intellectual influences. And share them with others on the social media so we can enhance our collective wisdom.