October 2003 Archives

Slashdot posted a story about an embedded HTML renderer being used for creating GUIs for medical devices. The story they point to is written by the marketing director of a company that's trying to sell the technology and, frankly, not all that interesting from a technology standpoint. It made me think, however, that basic Web technology has gotten stable enough and knowledge about it has gotten distributed enough, that including an HTML renderer for GUIs is almost a commodity idea, that it's a no-brainer to make UIs HTML-based, rather than creating a proprietary system. Even just a couple of years ago, the phone industry was trying to introduce WAP, which was a "simplified" version of HTML. Now, there's no need to simplify. From a device creator's perspective, an HTML renderer is nearly as much of a commodity as a clock chip or a serial port.

The effect that this has on device design may not be profound from an innovation standpoint, but from a development standpoint, it could be very big. Once HTML-based GUI rendering becomes as easy as putting a block on a block diagram, it's much easier to include it in engineered objects. It becomes a part of the production process, rather than part of R&D. I can imagine many more things growing rendered UIs because it's cheaper than creating them "in hardware." It's like when plastic backlit signs almost completely replaced neon signs. It not because the technology was any better or prettier, but because they were much cheaper to install, change and maintain.

In following up to my Smart Furniture Manifesto I decided to do a little research as to what's been done. My criterion for inclusion was to look for pieces of furniture that used environmental information in a deliberately functional way (rather than, say, Dunne and Raby's Placebo Project objects, which intentionally use information in a way that's only marginally useful).

I've come up with (only!?!) two:

• The chair version of Maribeth Back and Jonathan Cohen's Listen Reader, where kids sat in a chair (which contained the electronics and a speaker system) and flipped through a RFID-enabled story book. Depending on which page of the book they were on, the chair would make sounds that provided an appropriate soundtrack to what was on the page.
• The Trinity College Smart Couch which uses people's weight to determine who's sitting on it and make various things happen (there was a similar kind of technology being used in luxury cars back in the 80s to set mirror and seat position, I believe, but nowadays it seems to only be used for making safer air bags).

Are there any others out there?

[10/27: Found Purdue's Sensing Chair project. Here's a better description.]

The Tapwave Zodiac looks to be a nifty PDA-like device at just about the right price point to capture the market of 20-somethings who grew up with Gameboys, but have dayjobs. It makes more sense to me than the PDA-phone bricks that were around last year and, if it had a Danger-like mobile Net connection (not to be used as a phone, of course, but just to use the digital phone network) and a thumb keyboard, it would be a truly great device, an adult Cybiko. I'm sure they're working on it, but I hope they converge that functionality to it quickly.

An essay I wrote over the summer and have revised several times to include some of the thoughts I've posted on this blog is now on the Adaptive Path site. It's the most public airing of ideas I've mulled over in the last year or so and I'd appreciate others' thoughts on the subject. Feel free to post feedback about it here.

I just finished reading Virginia Postrel's �Substance of Style� which my (ever vigilant and savvy) Mom gave me as a gift. It's a great book. It's short, it's focused, it has some great ideas in it, and the ideas are presented with enough backing that it can be the start of a discussion, rather than handwaving.

One of the most valuable points it makes is that humans need decoration as much as utility, that �form follows function� denies a basic human desire to have things be esthetically pleasing, that esthetics is fundamental to how we judge the quality of experience.

This statement resonates with me on a bunch of levels, and with a number of ideas I've been exposed to lately:

• In sales, a traditional way to sell is to push features, functionality, without taking into account the underlying needs of the audience, many of which may have nothing to do with the features of the solution, but with satisfy other emotional needs. At Adaptive Path we've taken to trying to understand prospective clients' problems very thoroughly, to find out what they need, which is often not what they're asking for when they come to us. Since the product of our work is often abstract this means that we have to address the presentation of our results as seriously as the content. In addition to what we're going to do for our clients, how can we package and present it so that it has the most impact within their organization? The Minutemen have a song ("Shit from an Old Notebook" on Double Nickels on the Dime) that goes
  

  let the products sell themselves
  fuck advertising and commercial psychology
  psychological methods to sell should be destroyed
  because of their own blind involvement
  in their own conditioned closed minds
  the unit bonded together
  morals
  ideals
  awareness
  progress
  let yourself be heard
  

  And all respect to D.Boon, but he was wrong. Admirably idealistic, but wrong. Which brings me to the next idea,
• The surface matters because we superficialy judge things even though we're not "supposed" to. Lakoff and Johnson talk about how Westerners use metaphor of �essenses� to understand morality. The metaphor states that there's an essence to people which defines what they really are. The surface is seen to be separate from the essence and the essence is immutable once set. Someone who did a bad thing is �bad� forever, because they have shown that they have a bad essence. Someone who did good things, a hero, is difficult to imagine being bad, because they have a good essence. But woe to those heroes that show themselves as fallible, that their surfaces are not identical to their essence: Michael Jackson's behavior went from being charmingly eccentric to irredeemably suspect without really changing all that much, but our perception of his essence changed. According to Lakhoff and Johnson, that's because we feel that he betrayed his evil essence by pretending to be good (grabbing his crotch in videos was merely cute, rather than threatening, then one day it was weird). Britney Spears used this dichotomy brilliantly by playing someone with a pure essence who was just �being bad� in jest. However, the minute she actually does something racy, such as admitting to having had sex, doubt appears, the jest comes off as sinister, and record sales drop. Postrel talks about these same issues:
  

  Critics of ornament have aimed some of their sharpest attacks at bodily decoration�at all the ways in which individuals create �false� selves and at the temptation to judge people by their appearance.

  We are creatures of first impressions, surface judgments and sensual experience. In a world of teeming with information this becomes especially important, since we have less time to evaluate beyond the surface. Pretending that we can see the essence of things is one way to help us filter through the deluge in the short term, but it denies how we actually work. Moreover, since we relate to objects anthropomorphically, I believe that all of these ideas also apply to our relationship to objects. As Postrel puts it: I like that. I'm like that.
• Henry Petroski reaches the same conclusion in another very good book, �The Evolution of Useful Things,� where his studies of common functional objects (forks, tin cans, paperclips, etc.) reveal that, in his words,
  

  form follows fashion

  Even in the design of paperclips.

As I understand it, the Panopticon was originally a prison plan, designed under the idea that omniscient vigilance would prevent sin. People would get out of the habit of sinning and thus become rehabilitated if they thought they were being watched all the time, or at least surveillance would minimize the damage they could do.

As a metaphor for The Modern World, this has been discussed ad nauseum by Foucault and his followers. I don't want to go there. I want to wrap the thing back around and think of actual panoptic prisons and rehabilitation. Panopticism, and the anxiety created by it, is a pretty bad thing, but there may be worse things, especially as far as actual prisons are concerned. Modern prisons, as I understand it from reading the paper, suffer from the problem of being “colleges” for prisoners, spreading knowledge and—more importantly—creating social networks of cons and ex-cons. These are the original smart mobs. This is terrible for rehabilitation since, as theory has it, social networks reinforce themselves. I'm sure there are whole schools of criminologists studying methods for reducing the social network effects without denying basic rights to individual prisoners. My thought is: why not actually use modern panoptic technology to make house arrest more of an option? If every permitted object gets tagged with an RFID tag, nonpermitted objects and people aren't permitted, software is there to monitor movements, and [include every other technology that's has been shown to be privacy-invading], wouldn't it be possible for people to then either continue to live in their homes or live in communities that are not prisons, communicating and dealing with people who are not other prisoners? Yes, their home would be a prison, but with a key element missing: the social network.

[And note that I think that prison reform doesn't start with technological solutions, it starts with an acceptance of reality by lawmakers... an acceptance that I believe is currently severely lacking. But that's a different rant.]

My book was reviewed by Christine Wiegand for the SAP Design Guild. She liked it and said it was long. I'm glad about the former and can't argue with the latter. ;-) She liked the practical aspects of it and correctly identified that there should be a shorter description of some of the techniques. I agree and I've thought that there's a place in the market for "The Guerilla Guide to User Experience Research."

In the interim, please note that the book is written so that pieces of it can be read without needing to read the whole. It's designed to be used like a cookbook, rather than read like a novel.

Random idea of the week: a disco floor made of scanners. I was walking down the street on Monday night and saw a scanner sitting on the street. Technology is tossed to the curb all the time in San Francisco, and lately I've been seeing more scanners (for a while it was managed switches) showing up the trash. This led me to think "Hmm, what could you do with a bunch of scanners?" and the idea of the scanner disco floor emerged.

Picture a floor with constantly moving back-and-forth lights, making long streaky scans of people's feet and ankles. The images could then be used in a video wall. Imagine if someone was breakdancing over this floor, or making snow angels. I think it would look great. There's already a whole camera-less scanner photography movement (as beautifully exemplified by Kevin Lyons' Floraphilia site). This could be the equivalent of scanner video.

I looked into it, and and the Linux scanner drivers can support up to 100 simultaneous scanners, which is a 10x10 grid and should be enough for a decent sized dance floor, or a corner of one. Of course people couldn't actually dance on the scanners themselves, but if you placed them directly under some hard plexi, maybe the depth of field would still be good enough to get decent results.

I'm writing this in the jury waiting room in the San Francisco Criminal Courthouse, waiting to find out whether I'm going to be picked for jury duty. The place is incredibly full of a remarkably broad group of people, like an airport waiting lounge during a snowstorm. The 1960s clock is broken, always pointing at 8:45, and there's a sign underneath it (printed in rainbow colors) that says "Decorative Clock Only." That's one solution.

So this is the perfect time to address to all of the responses to my smart furniture manifesto. First of all, thank you to everyone who responded. I like the dialogue. Second, I restate your points below, partially to condense the length and partially as an exercise to understand what it is that you're saying so that I can respond to it appropriately. My apologies if I misrepresent your statements.

• Peter asks for a definition of smart furniture. The manifesto is my definition of smart furniture, but if here's one more like a dictionary definition: Smart furniture is furniture that uses the information technological advances of the last 30 years to make furniture more functional and elegant.
• Andrew asks what the connection is between furniture and information other than the literal one (computers have to sit on something). The connection is that furniture, as it's currently designed, has a poor information processing model. If you think of furniture as tools, then current furniture designs are using very little of the information available to them. Some pieces use the weight of the person (the user) to adjust the geometry of the chair, which is clever, but the mechanics are baroque (though not as baroque as the Aeron) and there's so much more information available about the user and the environment that isn't being used.
• Cassidy asks why smart furniture is better than dumb furniture. Paraphrasing: Dumb furniture is simple to make, simple to use, lasts a long time without upgrades and won't tell on you if you do something it doesn't like on it. All true. And I don't think that dumb furniture is bad. In fact, I love dumb furniture and I don't think it's going away. Whenever you sit on a rock you're essentially making that rock into dumb furniture, and I don't think people are going to abandon sitting on rocks or hanging their jackets on trees any time soon. However, I also think that furniture—the tools of our lives—will be
• Dave says several things, which although meant in jest actually seem to support my thesis

  Dave says	I hear
  Furniture is not dumb, people are dumb	People aren't going to get any smarter, but furniture can and should compensate for our failings when it can. Really this applies to all technology, and I think furniture design can do better.
  People weren't designed to sit in most chairs	Current furniture is often designed for looks rather than comfort—smart furniture would allow more flexibility to achieve both
  You will pry my Aeron from my cold, dead ass	Yes, Aerons are undoubtedly comfortable and an achievement, but they're like an HR Geiger monster mechanically and I think, unnecessarily.
  The Murphy bed is a great space saver and closer to smart than most	Absolutely. It was an elegant technological solution to a problem (small apartments created by rapid urbanization in the early 20th century). It cleverly leveraged off of the metalworking expertise that was created in the late 19th century. We should have more of those today, leveraging off of the information technology expertise created in the late 20th.

  Lane takes dayment's facetious comment about beanbags and points out that beanbags pay more attention to and are more responsive to the needs of the body, unlike Aerons which are difficult to adjusted, and don't adapt to the shifting needs of your back, arms and neck. He then points to chapter 2 of Galen Cranz's book "The Chair" as required reading. I haven't read it yet, so I can't comment on the book, but I agree with the beanbag comment. Beanbags are also a product of the technology of their time, since most of them were actually not filled with beans, but with styrofoam pellets, and utilized the adaptive property of those pellets to create a novel and more easily managed update of the traditional pile of floor pillows.
  Molly agrees with the fact that cars are furniture and succinctly defines them as furniture incorporated into little capsuled environments. I agree completely. She then asks if smart furniture is better than dumb furniture and what the relationship is between smart objects and smart furniture (i.e. do you need smart furniture if you have smart objects)? I think smart furniture is better than dumb furniture. If it can do what its primary purpose is more elegantly and efficiently because it's smart, it's better. I also think that the issues with making furniture smart and having smart furniture as part of an animist environment are two separate issues, connected by the underlying technology but little else. She's also creeped out by furniture that can reveal details about us that we're accustomed to keeping private. I agree. I think that there are a lot of potentially negative issues with the animist environment I've outlined.
  Erikpoints to "Chasm City" by Alastar Reynolds, where furniture is "aware" of the environment. He remembers they may have moved out of the way (no stubbed toes) or in your way (a place to sit). I think to some extent that's exactly right, and Erik's comment that it's pretty cool shows that the ideas are attractive. Why shouldn't your surroundings adapt themselves to your life? I, too, would like to have my dining room set arrange itself for the 6:30 dinner party listed in my calendar. I think that dinner parties are a rare enough event that it's likely that that wouldn't be a big consumer sell, but a footstool that scoots itself under my feet when I lifted them may be popular with the La-Z-Boy market, which already welcomes "nontraditional" furniture technology.
  Michael points to what he calls my earlier manifesto. Thanks for pointing out the similar style, but that's less of a manifesto than a succinct summary of my book. ;-)
  Adam first asks whether I'm jonesing for is some sort of utility fog. No, not in the extropian nanotech sense. The basic utility fog premise assumes a generic technological platform that can create tools on the fly from basic quasi-intelligent nanobots (or something like that). I'm actually going the other way: I think that tools need to become more specialized and enable specific activities.
  Adam then makes several points:

  Adam's points	My responses
  Desks and chairs are part of a system, and good ones are designed to be reconfigured by the user to suit their needs, rather than some program of the designers. That's one of the reasons why integrated seating/working surfaces fail.	I completely agree that all of these things need to be part of a system, and nowadays designing technology (which furniture has always been, even when it was made of wood, wool and iron) in isolation no longer makes sense. Allowing for adaptability is a tricker question, though. Sure there should be flexibility in the design, but one of the reasons that traditional furniture had ambiguous function was that it had to. Then, again, it would be weird to have six Aerons at the dinner table or a leather club chair next to a work surface, so it's not like "traditional" furniture design is all that flexible, either. Besides, maybe smart furniture would present possibilities for more flexibility?
  Adam likes his Murphy bed and doesn't see why it should be quaint.	I'm not saying it should be quaint, just that it is. It's of a class of furniture, like wash basins and hat racks, that are no longer nearly as popular because of how our society functions. Modern McMansions are build under different social considerations than when your house was built, so they don't have drawing rooms or root cellars, but they do have media rooms. The Murphy bed is a reminder of that time, so it comes off as quaint.
  Dumb furniture is easier to maintain and doesn't depend on the electrical grid to work.	I agree that dumb furniture is easier to maintain, but I'm not sure how much that plays into people's considerations anymore, so I'm not sure how much it matters to my point that furniture needs to become smarter. With the exception of people who seek out older furniture (which is a sizable chunk of the population, but probably not the majority), its function in society is as dictated by fashion as anything else. That makes it as frequently disposed-of as everything else. I too like furniture that lasts (in fact most of my furniture is pre-Modern antique), but my values are clearly not those of the majority of Ikea visitors. And so many things aren't going to work when the grid goes out that I'm not too worried that maybe my smart table won't work as advertised. Granted, key devices should still more-or-less function even without juice, but that's part of good design.