There's some information you use daily (where you live). There's other information you use occasionally (your parents' address). There's information you use rarely (who played Blair's cousin Geri on The Facts of Life) (Geri Tyler). Then there's the information you don't know you need, information you might need in the future. How do you go about gathering that?

This is the appeal of intelligent agents.

Originally posted on Thursday, November 17, 2005 | Link | Comments (0) | Trackback (0)

Author-cum- design-critic Bruce Sterling has written another great book on design, Shaping Things. Ignore its surprisingly sophomore-typography-project design, with its ugly, Adaptive Path green and wacky type choices throughout; it's really a must-read for interaction and industrial designers alike.

As he did in Tomorrow Now and his keynote speech at SIGGRAPH 2004, Sterling makes a timeline of objects, starting with artifacts ("things made by hand, used by hand, and powered by muscle") and going through machines, products, gizmos, and finally spimes, the objects of the future. It's a great model. Spimes are about sustainability, creating objects so full of information that they are able to be tracked and monitored throughout their entire lifecycle, allowing us to see the impact of each item on our natural environment. Indeed, sustainability is a major theme of this book, much like John Thackara's In The Bubble.

A great nugget from the book is the idea of metahistory.

"Every culture has a metahistory. This is not the same as their actual history, an account of places and events. A metahistory is a cultural thesis on the subject of time itself. Metahistory is about what's gone by, what comes next, and what all that is supposed to mean to sensible people...a cultures metahistory helps it to determine whether new things are appropriate..."

Sterling also goes through probably the best breakdown of Raymond Loewy's MAYA principle that I've ever read. Part of this breakdown is a discussion of what it means to be "designery." To whit:

"Being designery is not an affectation. Being designery is how one manipulates MAYA in public. Being designery is what one does, as a practical measure, in order to overcome the reactionary clinging to the installed base of malformed objects that maul and affront the customer. What cannot be overcome with reason can be subverted with glamor."

Sterling goes into a discussion that should be of great interest to interaction designers about RFID tags ("arphids") and how, combined with monitors and wireless, they could form an internet of things, communicating with each other and the internet like a swarm. Really fascinating stuff.

Near the end of the book, Sterling reaches a little far into the future for my taste, but then I'm a designer, not a prophet.

For a book with a good amount of theory, it's very readable. Recommended.

Originally posted on Tuesday, November 15, 2005 | Link | Comments (2) | Trackback (0)

I have a new essay on Adaptive Path's website: The Web 2.0 Experience Continuum. It's a little out there, but kind of interesting.

Originally posted on Monday, November 14, 2005 | Link | Comments (0) | Trackback (0)

Back in 2000 when Boxes and Arrows began, it was a great source of knowledge for user experience practitioners. Since then, my feeling is that the type of material that's typically found there has disseminated onto blogs, mainstream publications, and longer books on these topics. Blogs especially have likely been the ruin of many an online magazine, as writers skip the middleman/editor (sometimes to their (and our) detriment) and publish their work themselves.

This is certainly true for me. My last B&A article "Writing Smart Annotations" was in March 2003...a few months after starting my blog. Since then, most of my thoughts on design have gone into my blog, as have those of many of the early contributors to Boxes and Arrows. My list of design blog feeds is well over 50 now, and it's more than enough reading on the subject matter on any given day.

Why then another online UX magazine, UXmatters? With Boxes and Arrows seemingly struggling to find content and its way in the new blog-centric world, do we really need another online magazine that covers the same thing? The audience and potential authors hasn't grown significantly enough to warrant two in my opinion.

Online magazines like Slate and Salon exist and thrive (sort of) because they are able to offer premium content, written by paid writers and edited by paid editors, that no one else has. That no one else has. In this era of blogging, editors of the UX online magazines need to find the premium content that no one else has. They need to do the things that bloggers usually can't do. Things like longer interviews with luminaries, and long, ground-breaking essays. This is, needless to say, very hard. I wish them both luck.

I'm going to reiterate Saffer's Law again: it's easier to create a content aggregator than it is to create content. I'm adding a corollary: It's easier to replicate the form of something than its content as well.

Philips's researchers have created HomeLab, a 24-hour testing facility designed like a house so the scientists can observe people interacting with prototypes in a "natural home environment." From the website:

Philips HomeLab looks and feels like a regular home with modern furniture in every room, Van Gogh prints on the walls, and even a fully stocked kitchen. While no one lives at Philips HomeLab, temporary “residents” can stay at the facility for anywhere from 24 hours to two weeks, depending on the type of research being conducted. During their residence, individuals or families will go about life as usual, while interacting with the new technologies Philips has installed in the facility...Philips researchers will carefully watch how their tenants are living with these technologies 24 hours a day through tiny cameras, microphones and two-way mirrors that are hidden unobtrusively throughout HomeLab.

This is utterly ridiculous. Locking people in a house for "24 hours to two weeks" Big Brother-style while you observe them is simply not good research. You are not going to get natural results in an artificial environment, no matter how realistic.

The first core principle of ethnography is: You go to them. They don't come to you and live in a lab for two weeks.

Now granted, they aren't trying to do ethnography per se. They want to see what happens after the "newness" factor wears off with a new technology. This is an admirable goal, but wouldn't the money for this lab have been better spent finding research subjects and installing the prototypes and observation equipment in the subjects' own homes? Only there are subjects going to feel more natural--because they are in their familiar environment, not some lab made to look like a house.

The trend these days, even in usability testing, is to do it in homes, using the subjects' computers. You get a sense of context then, and the subjects are more relaxed because they are on their own machines. This HomeLab is just adding a veneer of authenticity on top of artificiality. Don't these people watch Survivor?

Originally posted on Friday, November 11, 2005 | Link | Comments (1) | Trackback (1)

While not the worst conference I've ever been to, DUX 2005 was a disappointment. I think three things worked seriously against this conference: the location, the format, and the chosen content. Which is to say, almost everything.

Social networking is as (if not more) important than the program itself, and every effort should be made to foster those connections. And I'm not just talking about formal receptions. Fort Mason is a great location, but not for this sort of event, especially considering there was only one track at the conference. If you didn't like the topic, there was no place close by (in the same physical space) to hang out in, and nothing really outside of the conference within walking distance. Everything in San Francisco was either a long walk or a car/taxi ride away. Not ideal for interactions around the conference.

One track programs are an interesting idea, but don't work with a large group with no common theme. If you don't like the session topic, you are stuck either doing nothing or sitting through something bored. A little variety is great and makes the conference seemed crammed full of good stuff. It also makes for more varied conversations ("What was that session about? Oh interesting...").

I heard many people complaining about the five minutes of time each speaker got. Me, I frequently found that to be a mercy. It's only when someone is a good presenter with interesting content that you want to give them more time. And I'm sad to say that I found that happening with only a handful of the chosen presenters, like Maren Costa of Amazon, Jane Murison of the BBC, and Jan Chipchase of Nokia. Granted, I missed a few sessions, so unfortunately I didn't see everyone.

For a conference about design, there was very little talk about design. The five-minute format lends itself to much more of a show-and-tell format than to in-depth case studies. And that's too bad, because DUX is supposed to be the practitioners' conference (although there were certainly a handful of pure academic projects that were better suited for CHI or something).

Now, lest you think I'm just a hater, I have great sympathy for those responsible for DUX. It can't be easy trapped between all those organizations, not to mention that the "peer-reviewed papers" format puts you at the mercy of both the submitters and reviewers for your content. Perhaps for a non-academic conference, you could dismiss with those for choosing sessions? A better solution is one where the conference has a curated theme and hand-selected presenters, like (allegedly: I've never been invited to either) TED or Design Engaged.

We need to use some of our tools on ourselves and our products (like this and other conferences) occasionally. Aside from the spotty wireless, the DUX conference format could have been from 20 years ago. We should figure out what attendees want (their goals) and then figure out the best ways of achieving those--the best form for this sort of conference. My guess is it won't be seven people on a stage talking about seven different projects, switching every seven minutes.

Originally posted on Monday, November 7, 2005 | Link | Comments (0) | Trackback (0)

In Erving Goffman's legendary book, The Presentation of Self in Everyday Life, he identifies three social norms for conversations ("facial engagements"): opening, maintaining, and leave-taking aka ending the conversation. Instant messaging is great for openings ("Accept this message?" and the little ping) and maintaining (the open window, the noises), but is terrible about leave-taking.

There is no good, standard way without seeming rude to end an IM session. In person or on the phone, we usually give some indication ("Bye!") we're ending the conversation, but with IM, that's not typically the case, at least not with my conversations. Thus, by the end of the day, my screen (unless I am using Adium with its tabbed IM sessions) is littered with conversations that have simply trailed off long before. Just closing the window seems rude, especially if the person on the other end sees you have disengaged with them.

I'm not sure if a solution needs to be designed into IM clients, or whether after a few more years of IMing becoming mainstream, social patterns will work themselves out.

I'm attending DUX 2005 for the next three days and will be one of the hosts for the Adaptive Path Studio Tour. Hope to meet old and new colleagues there!

Yes, I know this entry breaks blog time...

Originally posted on Thursday, November 3, 2005 | Link | Comments (0) | Trackback (0)

One funny thing about blog entries is that they are designed to be timely one-offs, and yet every time I've posted something with a relative time reference ("Yesterday, I..." or "Last week there was a discussion about..."), I've regretted it. Most of the people who visit my blog (and probably yours too if you have one) don't follow religiously every day and likely stumble onto your site via a Google search.

Search destroys time. Sure, "date posted" could be part of the search algorithm, but that could be good or bad depending on the situation. People searching for "mood boards" (my #1 hit for some reason) don't care about what happened three days before I posted that entry.

The temporary, the timely, ends up being permanent. Be careful what you say and how you say it.

Originally posted on Wednesday, November 2, 2005 | Link | Comments (0) | Trackback (0)

Perhaps I am naive about this, but I feel it's the job of Congress, specifically the Senate Judiciary committee, to examine presidential nominees to the bench for competence and extremism, not ideology. If an nominee isn't unqualified (Harriet Miers) or a demagogue (Robert Bork), the nomination should pass. After all, there is no guarantee how a justice will vote once on the Court; some of the most liberal justices have been nominated by conservative presidents, and even Sandra Day O'Connor has moved to the center over the years.

From what I've read about them, John Roberts and Samuel Alito seem like reasonable, conservative choices, as well as honorable men with solid records. I do not agree with their judicial philosophies, per se, but I agree even less with the Democrats' knee-jerk reaction to block both nominees. We should be grateful these nominees have been as moderate as they are.

I understand the concern about rolling back Roe v. Wade and I share it. But there are many, many other complicated issues going to face the Court in the next, oh, 25 years, and having qualified, reasonable jurists there able to handle complexity is terribly important. For now, all we can hope for are qualified candidates who don't adhere to strict ideology and who are able to take nuanced views. People unlike Clarence Thomas, in other words.

Our day will come again.

Update: As I semi-predicted yesterday, Alito is finding support among liberals who are able to look past the man who nominated him.

Originally posted on Tuesday, November 1, 2005 | Link | Comments (1) | Trackback (0)

« October 2005 | Main | December 2005 »