Tom Moran on Everyday Adaptive Design (long)

Here's my (long overdue, and just plain long) notes on Tom Moran's plenary from DIS2002. My comments are in grey - apologies for bad spelling, grammar, and prose(!). I've added emphasis and links where possible - Moran's PowerPoint slides are available too - read in conjunction?

The final plenary session at DIS2002, by Tom Moran, was a hugely important presentation, imho, linked to several memes around the disappearing interface, or interfaces based around behaviour rather than aesthetics; open source collaboration; flexibility and modularity; pattern-based design; the importance of social process, communication and networks etc.

Moran, of the IBM Almaden Research Center - is from an architecture background originally, but then was quickly part of developing the practice of design in computer science. Hence he draws on Christopher Alexander, Jane Jacobs and Stewart Brand throughout the talk, as well as Bernard Rudolfsky [a new name for me to track down]. Moran prefaced this talk on Adaptive Design by suggesting that perhaps we can design Design. As in, what do we want Design to be? A proactive attitude ...

A brilliantly information-rich speaker, ploughing through his Powerpoint at a rate of knots, Moran began by rattling off a brief history of changing design perspectives relating to changes in computing systems i.e.

Batch systems -> Cognitive perspective
Interactive systems -> Usability perspective
Personal -> GUI
Networked -> Socio-technical
etc.
ending with
Embedded -> Experience

Which was hugely useful, reminding us of how early notions like usability and socio-technical network analysis hit the practice of HCI (Human-Computer Interaction, often CHI) within the discipline of computer science. [it also pleased me, being a comp sci grad in the early 90s and now a designer (rather than from a graphic design background, say), in that it further helps me explain to people that design is/was the key element within the displine of computer science].

Then a serious investigation into design itself, and designers.

Design lives everywhere, in all of us. Specifically in "the users", who everyday commit little acts of design by adapting systems to their needs. Important to note that adapting is not bad (as implicitly, designers often try to prevent this, nowhere more obviously than on the web perhaps) but good. And it's about continuity rather than change. This means that perhaps we should tend to specify rather than build? This is a fundamentally social process.

What is the goal of design? For Moran, design artifacts that become suitably and intimately enmeshed in people's lives. Fundamentally, it's "a humble field" - not about producing objects of admiration. We can think about 4 facets: Usefulness; Reliability; Usability; Delight. These can be correlated with Vitruvius. Usefulness is the hardest and most important - perhaps we've short-changed it?

It's more evolutionary than revolutionary; More service than product.

Moran sees a life-cycle of development, where design is a set of distributed activities, of different kids, by different people, at different times - a continuum of Design, Build, Adapt (a logarithmic rather than linear relationship, hinting that time is the best designer!)

Moran described the limitations inherent with the practice of Professional Design (can't predict usefulness; can't truly advocate the users; an inward-looking community: not least the "pull" of over-designing)

In Generic Design, where, as Herb Simon noted, "everyone designs", designing is a type of cognitive activity, an approach (as opposed to, say, diagnosis or clearer decision making) where design problems are ill-defined and ill-structured. Designing here is specifying, stealing (or rather, reusing domain knowledge!).

It can include the design of a service, where integration is the hardest part of design. For example, Moran delineated the UX of a cellphone (from research by Palen & Salzman), across various devices, arenas, modes.

So design is a social process based around collaboration, negotiation and user participation - in fact, not a profession but a community. And that, concluded Moran, was enough interrogation of design itself, clearly impatient to get away from the naval-gazing and onto the good stuff.

Onto Everyday Design, where Strickland noted, "Everybody is a designer in everyday life. Yet we share no common vocabulary for describing everyday design practice ... design is not limited to the province of specialists who have formal training ... rather, design behaviour is a fundamental element of our species' adaption."

Strickland's work includes a "Portable Effects" exhibit, or "glimpses into human mobile nature" i.e. studying use of mobile equipment (bags, wallets etc.). Notes the way that purses (Moran's example from Strickland's work was "Esther's purse"), are organised by their 'users'; to their organisation - might appear muddled to others, but utterly workable to their owner; varying in context, it fulfils many different roles. This Everyday Design is supremely authentic. It's a "continuous process of adaptation"; it's specific yet "free-flowing"; it has a tight fit to the situation. Moran praises this "informal, pragmatic, ... offhand ingenuity".

Then Moran quickly sidestepped into examples of Adaptive Design from Vernacular Architecture, noting Rudolfsky (architecture without architects, or non-pedigreed architecture), Alexander (unselfconscious design), Jacobs (vitality of the street from its diversity and density), Brand (the low road), Venturi (theory of the ordinary and the ugly). [a low-power warning light goes off in my head here, about confusing heavy use of something with popularity i.e. because something is heavily used, doesn't necessarily mean it is enjoyed/valued e.g. the Arndale Centre shopping mall in Manchester fr'instance. Of course, this kind of survival-of-the-fittest approach to usage - in Brand's words (from memory so probably not exactly), "function reforms form, perpetually" - does mean that these examples must be useful, which is where we came in. The arch-modernist in me is screaming though - luckily i can ignore it, or at least reversion it as part of some kind of dialectic process. It's summed up in Moran's point that design is, or should be, a profoundly humble practice.]

Moran picked out quotes from Rudolfsky in particular [clearly a touchstone], based around fitting buildings to their surroundings rather than trying to conquer. Moran illustrates with images of Pakistani wind scoops and Victorian houses in San Francisco, the latter indicating a flexibility of architectural space to enable hugely varying formats, still used 100 years later. Also Stewart Brand - the architecture of "the low road", with its mobile homes, ranch houses and petrol stations - a vulgar but oft-used vernacular. Aside on Jacobs quote: "old ideas need new buildings, new ideas use old buildings" [e.g warehouse space in deindustrialised areas; this links to my MA thesis on fringe zones within post-industrial cities, where cultural industries invariably initially thrive].

So to interactive systems, and Adaptive Design in that arena. Moran lists various customization techniques (scripting languages and macros, rules, features, parameters, skins and rearrangement, and so on). He notes though that this is essentially a "limiting view" of adaption.

The essence here is the difference between thinking about User vs. Adapter. The User assumes the system is ready for a purpose, and merely puts that into action; hence the usability is the designer's problem. The Adapter makes the system suitable for a purpose, thus usefulness is the adapter's problem; they make the system their own. [What an insight! Usability is a self-defining problem for the design, due to the initial, fundamental approach to design, not the effectiveness of a particular design. If the design was less prescriptive, and more adaptive, usability issues could diminish significantly]

Examples of Systems for Adaptive Design include wikis and blogs; spreadsheets; email (which gets used for various tasks as well as sending/receiving); messaging (unforseen use of SMS); cellphones (when used for rendesvous); the freeform space of the desktop, adorned with post-its.

Moran related various studies to do with Adaptive Design (incidentally, it was refreshing and enlightening how often Moran referred to recent research as the basis of this thoughts) such as:

• Mobile Work (planning on what to carry, redundancy dealing with uncertainty, tools for differing contexts, lightweightness and flexibility, where facets include "micromobility", "instant-on", and connectivity to local resouces)
• Email (people "progressively appropriate [email] as a habitat in which they spend most of their workday"; notetaking, negotiation, to-do lists, filing, contact management. But important to note that filters [the designer's attempt to enable organisation?] are only slightly used.)

Moran compares Professional and Adaptive Design - a useful separation of mindsets:

formal vs. informal;
anticipated vs. situated;
ill-defined vs. concrete;
reflect vs. act;
specify vs. build;
program vs. arrange;
adventurous vs. conservative;
make it right vs. make do.

In terms of actually Designing for Adaptive Design [which we'd all been waiting for!]:

• Moran returns to Stewart Brand, and his notion of an Architecture of Layers (I've mentioned this before - it allows slippage between layers, fast layers exploring change, slow constraining fast and providing continuity, integrating change - as seen in the flexible Victorian house layout perhaps, where concrete/steel rsj's, foundations, plumbing etc. provide solid structure for years, with many walls proving temporary and transient, and airy spaces provide opportunity for speedy change, daily re-arrangement of 'stuff').
• Then we have Platforms, Not Solutions, or rather, overbuilding the infrastructure and underbuilding the features. Provide basic reliable features, but generally under-design (like his idea of deferring quick decisions, but providing opportunities). This also enables quick experimentation [MT is a good example here - a basic set of features, implemented in numerous ways by users, leading to much experimentation and further refinement of original offering.]
• Space To Evolve - leave some spaces rough - the equivalent of low-definition spaces (like a garage, loft, or porch [link to Weinberger's points about imperfection?] - also non-minimal and not over-optimised i.e. "roomy"!
• Managing the At-Hand - allow people to arrange their stuff in their spaces. Moran doesn't seem over-keen to prescribe at points like this - when he asks, rhetorically, what looks and feels adaptable, he will only commit to suggesting that it should convey "opportunity and potential" and have an "aesthetic of ongoing process." [Probably wise to hold back, and wiser still to make the audience do the work - again, specificity and close fit are relevant here!]
• Modularity allows recombining and repurposing, via cellular spaces (hierarchic) which are joinable and splitable, preferably open-source in attitude, to accomodate heterogeneity (as opposed to closed systems which are expensive and limiting).
• There are also important points around Process - if people have local control, you have to provide documentation, service and support, make adaptations sharable [again, the open source approach] and documented. [this means the designer has to be a continual communicator about systems of design and engage in multiple dialogues - rather than overdesign a system and move on, the usual hit and run we engage in.]

As to what systemic trends are supporting these ideas, Moran pointed to open standards, web architecture, "portalization" and "freeform technologies". In terms of interaction design, which Moran noted was still fundamental, we're talking lightweight, flexible, looser less crammed, and interchangeable/interconnectable. Behaviourally, we need to think about the time to adopt, about maintainging vs. changing habits, about amenity and function vs. style, reflection vs. on-the-fly action, about engendering the desire/ability to experiment, and where the threshold is in terms of local control - when do we inhibit adaptability? [obviously context varies this last greatly]

Moran concluded with an excellent list of research material (check the last pages of the ppt), directly relevant to the examples of Adaptive Design he'd mentioned. And then the summary:

"Adaptive design runs rampant! It is vital, creative and messy"

And we - the design community - have a choice. We can dismiss it as vulgar, try to clean it up; embrace it; or design to support it and improve it. Naturally Moran's talk was heading in the direction of the latter two, but he was smart enough (again) to pose it as a question; to make us think about actually taking action on the choice, rather than sinking back comfortably into the auditorium's seats.

The talk had been so rapid-fire, information-dense and, well, brilliant, that the delegates seemed slightly shell-shocked (in a good way!) - thus the Q&A was something of an anti-climax. Moran noted a few more research angles (Alan McClean's work - "Buttons" (citations here) and reiterated a few key points (adaptabilty as a social process, "sharing layers", "specification as a dialogue" [link to XP?].

Update: Many of the ideas triggered by Moran's talk ended up in my Adaptive Design presentation to AIGA Experience Design London. Similarly, my Adaptive Design category here collects further thinking around these themes.