The Semantic Web, Imagined. Designed?

Further reflections from DIS2002 concerning the futures of interaction design

There’s a buzz about this notion of the Semantic Web at the moment, to say the least. Paul Ford’s excellent speculative fiction about Google is as good a place to start as any (though read his further comments on his essay too). (Ed. This piece originally written and published in August 2002.)

[A note here: Paul’s approach reminds me of the exercises Tony Dunne & Fiona Raby get their students to do at the Royal College of Art. Obviously, scenarios (with personae) have been standard design process for a while now, but this more imaginative approach allows a little more freedom. Dunne & Raby, presenting at DIS2002, ran through a few of their ‘Value Fictions’ — akin to science fiction, but extrapolating social/cultural values rather than extrapolating technology — in which students imagine possible futures for contemporary technology, in order to flesh out the cultural changes associated with technological change — nice, because it inherently sees “design as a social process”, a notion later expanded by IBM’s Tom Moran at DIS2002.]

Back to the Semantic Web: See also Matt Webb’s basic roundup, and further initial thoughts. And Matt Jones had no sooner asked the question of what this all means for web design, than Dan Gilmour had begun to pick it apart. Liveblogging action from a day at IBM’s Almaden Research Center, where Tom Moran works (there’s some interesting asides in there on the craftsman/engineer debate from earlier, too), Gilmour’s notes are a little random but, echoing many concerns at DIS, do address where interface design is going.

Back to Matt: as he notes, the sheer brazen utility of apps like NetNewsWire effectively questions the value of website design per se — as in, building a site for information/interaction — whereas NewNewsWire immediately illustrates the value of freeing information from a bespoke, unique housing. Why build this site the way I have (Ed. this was originally published on cityofsound.com), when you can access these pieces quicker, easier, and on your own terms? Indeed, help yourself to the cityofsound RSS feed, and further avoid the design of this site.

Where does all this leave the designer?

I’m still reflecting on Tom Moran’s DIS2002 plenary, on adaptive design (his slides are now up at the DIS site). More in a later post, but his suggested strategies include of over-building infrastructure (the interior forms), and underdesigning features (the exterior); building an architecture of layers to integrate change (after Stewart Brand’s How Buildings Learn, in turn borrowed from Frank Duffy) and enabling users (adapters, in his terminology) to “manage the at-hand”.

It’s clear that products like Movable Type exemplify aspects of Adaptive Design. Ben & Mena Trott have defined a brilliantly flexible system, interpreted in thousands of utterly different ‘exteriors’. Similarly, as Clay Shirky noted in a recent talk for us at the BBC, Livejournal’s infinitely flexible approach to organisation is Adaptive Design at the information architecture level. There’s a related shift towards focusing on behaviour (Dunne & Raby) and open systems (Moran).

As the-designers-formerly-known-as-interaction-designers, I suspect we’ll become akin to engineers of possible behaviours, in the interaction sphere sure, but also, as Matt Jones hints, in terms of networks and organisation, increasingly leaving many aspects of aesthetics to the numerous apps and services adapter-users employ i.e. they’ll arrange it how they like. Else, it’s a question of employing a few stylists.

The best interface design may be invisible, but the practice is unlikely to disappear. Ubiquitous computing, and the Semantic Web, will probably mean a proliferation of interfaces across a proliferation of platforms, protocols and products (recent memes include the return of the command line (cf. Jones, Webb, Neil Stephenson but not Peterme) and gestural interfaces (linking Jakob’s latest and Spielberg). Leading to a possible divergence between:

• an increasing engagement with software engineering (the return of old-school HCI/CHI)
• information architecture (perhaps building systems which users deploy and manage; though with further schism between Interaction Architecture and Information Architecture, aka doing and finding)
• purveying style, interaction and communication across numerous different interface spaces.

The technical gulf between, say, gestural and command-line is not insignificant either. So do we see this increasing division of labour within what most of us designers do? An industrial revolution?!

In large teams perhaps. Specialisation should mean better products. But part of me hopes that the designer is still seen as a connector or conduit between differing worlds, and as such, needs to be able to speak many languages (across IA, UXD, code, ethnography, ‘editorial’, cultural change etc.); a player at all stages of the process. A richer view of design ought to lead to richer experiences for users/adapters (given humility in the designer); a richer view of design notes that experience design may increasingly become less about the visuals — no bad thing, I think. It tends to distract away from the areas which we’ll need to know about next. To be frank, the great imaginative leaps aren’t likely to require our graphic design skill-set — all the interesting work is around technical/social systems.

Anyway, enough navel-gazing. One of the many wise things Moran said at DIS2002 is that design is essentially a humble trade. Just as the best interface is invisible, the best designer probably is too. Designers are the glue in the design process — interfaces will be the glue for most of our social/cultural/business activities. Whilst not trying to sound all Soviet Propagandist about it, we must become more like these next generation interfaces! You are what you interface. You interface what you are.

This piece originally published at cityofsound.com on 5th August 2002.