Tuesday, August 17, 2004
I've been arguing for years now and have tried to put it in to every editorial style guide I can that the internet is a medium like radio and television and thus should not be capitalized in documents. Finally, Wired agrees.
Thursday, August 12, 2004
Designing for Gizmos and Spimes
Writer Bruce Sterling gave the keynote speech at SIGGRAPH this past week and Luke Wroblewski has written an interesting summary of it.
In the speech, he expands on some ideas about objects he presented in a book called Tomorrow Now that I read about a year ago. To summarize, Sterling categorizes devices into machines, products, gizmos, and (now) something called spimes. Machines are 19th century, and products are the 20th century children of machines. Gizmos are mainly what we're designing now. Quoting:
"For a gizmo, the function is the decoration. A gizmo...has more functions than the user will ever be able to master, deploy, or exploit. It's designed to have baroque or even ridiculous amounts of functionality...A gizmo...doesn't want you to accomplish any task in particular. It wants a relationship; it wants to be an intimate experience...It wants you engaged, it wants you pushing those buttons, it wants you faithful to the brand name and dependent on the service."
Even more outrageous (for designers anyway), Sterling goes on to claim
"End users don't want to solve problems. A solved problem is actively dangerous for them. Any end user with a permanent solution has lost a job...This also explains why end users don't settle for cheap, simple, fully usable software. After all, if software is simple and useable, then anyone can use it. End users...can't afford to be just anybody, because this is a swift ticket to poverty."
This of course goes against almost everything that has been written in the field of design in the last, oh, twenty years as we've used design as a method for solving user problems. But Sterling, erm, might also be right. Sure, some people want simple ways of tackling problems, but they also might not. In looking at users' tasks and goals while working, it's easy to forget they might not want you to solve their problems for them: they instead want the tools to solve it themselves. They might not want something made simpler; they just want to do something better. There's a difference.
At SIGGRAPH, Sterling upped the ante and introduced a new object into the mix, objects of the future he's calling spimes. Quoting Wroblewski paraphrasing Sterling,
"Spimes are objects that have "swallowed" our past by combining social networks, RFID tags, GPS systems, self Google-ing, peer-to-peer networking, and more. Spimes can reveal most anything about themselves. They are precisely located in space and time, have a history and identity, and make their nature transparent to us. Spimes are "user groups first, and objects second." But most importantly, spimes allow us to make good on sustainability through a traceable lifecycle. Because spimes have identities and complete histories, they create accountability: we know where they end up and we know the impact they have on our world."
The impact of spimes on the world and the world of design is pretty mind-boggling. Imagine designing an object that has such total transparency that you can follow through its lifecycle as users buy it, use it, and even discard it. Then imagine the privacy issues that will have be navigated. We're seeing only the smallest taste of this with RFID tags right now. Now imagine designing this same object with all of its innards exposed for adaptation and hacking and you begin to see the world we're heading into.
There's a class at CMU called The History of Objects. Maybe one should be taught called The Future of Objects.
Sunday, August 8, 2004
Every year, it seems, we get a new organization in the realm of design. A couple of years ago it was AIGA's Experience Design. Then, two years ago, AiFIA. Last year it was IxDG. This year, it's UXnet. And while I'm a member of some of these fine organizations, I've come around to my friend Tom Alison's position (well-expressed on the defunct uxDesign blog, second from the bottom), that we don't need more organizations. We need more egomaniacs.
"I don't think any ho-hum professional organization is going to make even a ripple in advancing the awareness of information design, or influencing anything that has anything to do with the information technology industry at all...If you look at anybody who's made an impact in the industry, take a close look at what they're like: eccentric, self-aggrandizing individuals who attract attention to their cause because they are in love with themselves and love being heard. And they also sometimes happen to be geniuses."
He goes on to give examples: Richard Saul Wurman, Steve Jobs, Bill Gates. And, of course, our pal Jakob Nielsen, who has done more for usability than the UPA could ever hope to achieve by constantly pushing it in people's faces. Or take Edward Tufte (please): he's done more for information design than, well, probably anyone ever. He also seemingly has an ego that is so large, it is difficult to chart.
Each of these folks has pushed their agenda to the point where it has become the industry's agenda, and all of us are better off for it. Some of these folks (Tufte, Nielsen, Don Norman) are considered the main thinkers in Design these days, although none of them are designers. We need more design egomaniacs (RIP Paul Rand) to promote themselves and, by extension, design, not more organizations.
So get to work, you design slackers! Publish books. Flood the web with blogs. Find your "thing." Find work. Do your own projects. Be obnoxious and eccentric. Wear a costume. Stroke your own ego. We'll all thank you.
Friday, August 6, 2004
DIS 2004: Day Three
To say that the final day of the Designing Interactive Systems conference was a DISappointment would be an understatement. In fact, to say that about the entire conference would be an understatement.
The biggest missed (or is that DISsed?) opportunity was the Beyond Human-Centered Design panel. What could have been a really brilliant, edgy panel, became instead a pointless discussion of innovation. The panel never got to the keyword "beyond" or even, unless my memory fails, seriously discussed other philosophies of designing aside from UCD. I feel like there should be a Do Over on this panel.
The final day also featured a discussion on the future of DIS: whether or not it should be combined with DUX (Designing User Experience) or take some other form. What DIS should be, basically. Here's my thoughts on the matter:
To start with: Physician, heal thyself. Let's find out what the users (the conference attendees) want out of the product (the conference). From what I heard, these seem to be a place to:
The problem is that the majority of the conference is designed around the first goal: for academics to present papers. Which would be fine if it was done differently: smaller, workshop-like sessions, to discuss the paper, not just get a lecture on it. How it is now is deadly dull, like boring undergraduate lectures. Having written a good paper doesn't necessarily make you a good presenter.
I'd recommend having more than two tracks, with several smaller rooms instead of two big rooms. Smaller sessions would mean more chance for discussion and personal involvement. Having the papers before the conference would be a good thing too, so that there is discussion about the papers instead of just rehashing them.
A major problem with DIS is attracting less design researchers and academics and more working designers. And not just as attendees, as presenters. They should look at the IA Summit as a model of how to do this well. By attracting more designers, DIS would have more case studies and applied work, instead of the theoretical, which it was over-burdened with this time. It would also have more work to see and interact with. Reportedly, DUX did more of this. This means that there needs to be an outreach to the design communities and to corporations. Many people I've spoken to have never even heard of DIS.
Aside from meeting people, what I really wanted out of DIS was to hear about the bleeding-edge stuff, either projects or theories. The inspiring stuff. And I did get some, just not enough. DIS left me wanting more, much more.
Thursday, August 5, 2004
DIS 2004: Day Two
The second day of the Designing Interactive Systems conference was, to my mind, the best day of the conference, featuring as it did the excellent Designing for Hackability panel, a tour of local design studios and research labs, and an evening barbecue in Somerville. I also got to spend some quality time with not only my CMU comrades, but also a bunch of new friends and acquaintances including Matt Jones, Dan Hill, Chris Heathcote, Peter Merholtz, Eric Wilcox, Liz Goodman, Jofish Kaye, Michele Chang, Katja Battarbee, and probably a host of others I'm forgetting. Meeting people and hearing about their work was definitely a conference highlight and certainly more interesting than most of the papers and presentations I heard (more on that later).
The Designing for Hackability panel was the best "official" presentation I heard. The definition of "hacking" floating around was pretty loose: everything from appropriating things for not-designed-for purposes to breaking open devices to rewire them. Thus, a lot of stuff was considered hacking and it became an unofficial buzzword for the rest of the conference. (Leaning against the wall became "Hacking the Wall" for instance.)
The interesting conflict on the panel for me was whether or not we should bother trying to design for hackability. Panel moderator Anne Galloway pretty much said there was no use in doing that, since people will find a way to hack things even when the manufacturers specifically design something not to be hacked. But I found a more interesting answer in Dan Hill's comments about creating "loose layers" on top of a more permanent structure. This allows people to peel back and hack the loose layers while keeping the base solid. Dan also had another interesting comment that I wish had been followed-up on in the Beyond Human-Centered Design panel: namely, that HCD/UCD doesn't scale well. When you have millions of users, UCD methods are inadequate for designing for them.
There's a lot of great ideas from this panel I am still unpacking. If only the rest of the conference had been as interesting.
For the studio tour, I visited the beautiful office of Orange, saw some really interesting work on threaded conversations and group information management at IBM Research, and saw the start-up spirit at work again at Ambient Devices. I really enjoy seeing other designer's work spaces.
The day ended with beers and BBQ at an MIT student's house in Somerville, eating ice cream sandwiches in the dark, talking about design.
Monday, August 2, 2004
DIS 2004: Day One
I'm in Boston for the Designing Interactive Systems conference, meeting people, playing with interactive displays, and looking at posters (including the interesting Friction as an Interaction Design Metaphor).
My favorite paper presented so far has been Making Tea: Iterative Design Through Analogy. The paper basically came about because a design team was trying to understand how chemists use lab books. Since the designers had very little domain expertise, they had to devise a way to ask questions and create a common language between them and their subjects.
So, when in doubt, make tea.
They basically used an analogy (or, really, to my mind, a metaphor): CHEMISTRY AS TEA. They basically used the metaphor to explore the domain of chemistry, figuring out how chemistry was or was not like making tea. It gave them a "safe, known, observable, repeatable, and interrogatable" process to compare a chemistry experiment to. They had the chemists make tea and record how they did it in lab books, asking questions about how the lab book was being used. Then they later made tea using scientific equipment instead of traditional kitchen items. It was a way, alongside other design methods, of bridging the domain knowledge gap.
this site (xml)