Gandhi on Ubicomp

Remember Gandhi's steps of a revolution? "First they ignore you, then they laugh at you, then they fight you, then you win." For as long as I've known the term, ubiquitous computing has been largely ignored, written off as a scifi pipedream from the people who promised you AI and cars that would run on water. That's beginning to change, as hardware such as the Arduino and programmable mobile phone handsets enabling artists, researchers, and makers like Eric Paulos, Elizabeth Goodman, and Julian Bleecker to join the digital and physical worlds in new and interesting ways. Now that it's harder to ignore ubicomp we're seeing laughter.

Nicholas Nova pointed to an article in Backbone Magazine where "Austin Williams, technical editor of the Architects’ Journal and director of Future Cities, a forum that critically explores city issues" wasn't so impressed with Paulos's work. "Williams, who calls such technology-driven projects indulgent, points to more urgent urban problems awaiting solutions, such as the loss of social connections between city dwellers." While not technically laughter, I put this belittling in the same bucket on Gandhi's progression.

I think Williams is wrong because he fails to allow for the rate that technology matures. Those practical ubicompers like Paulos, Goodman, and Bleecker have only had affordable easy-to-program embedded hardware and open mobile handsets for a few years. They're explorers taking first steps on a new world. Explorers sometimes find gold, sometimes find deserts, sometimes get eaten. It's the nature of the game. The explorers I named would be the first to tell you they're not buliding products, things for wide deployment that are meant to be consumer-ready, shelf-demonstrable, and poised for their 30s spot after Leno's monologue.

But in the legions of developers, hackers, hobbyists, alphageeks, and tinkerers who look at their work there will be some who see a product to build—the homesteaders who build cottages, shops, factories. From digital photo frames to the Chumby and the Dash (disclosure: O'Reilly Alpha Tech Ventures is an investor in the Chumby), we're seeing the first products emerge. Some of them will be like Roanoke Colony, some will be like New Amsterdam. The next wave of products is going to be really interesting, and in there I'm expecting to see things that tackle "the loss of social connections between city dwellers" and other matters that Williams was concerned about. That should go some way to silencing the laughter.

Gandhi's progression invites the question: if next "they fight you", who will be fighting ubicomp devices? Not just those who fear wifi, but entrenched business interests. Any ideas? Leave them as comments below.

(Updated with Julian's new URL)

Teaching Kids Programming

For the last two years I've taught a computer club at the local primary school. I'd get six or eight kids aged 8-10 for two hours at a time, once a week for three or four weeks. They varied in previous experience from "have computer at home and play games on it all the time" to "Mum doesn't believe in computers". Last year I wrote up my experiences from 2006, and what follows is my summary of the 2007 club.

I repeated some of the same exercises as the first year (building stop-motion animations) but the big change this year was that I introduced Scratch instead of Lego Mindstorms. Executive summary: it was a great success.

The Lego Mindstorm kit hadn't been a straightforward win for me. On the positive side, the graphical programming environment worked well. Kids are not good at typing. They were able to grasp loops and conditions and so on, without ever having to struggle with parentheses or indents.

Against that, though, we discovered into several significant problems. The hardware isn't of high enough quality—tracking a line across a sheet of paper sounds like a great project but the light sensor wasn't able to reliably do the job. Kids need strong positive feedback when they get it right, otherwise they rapidly lose interest. The compile-install process was a downer for them (added complexity with no benefit). The range of projects possible with the hardware in the box wasn't exciting enough to maintain their attention over the long-term. They like the idea of a robot, they're not so excited by the reality of it.

I had much more success with Scratch. The kids got quick successes from moving drawings and bouncing them off the sides. They learned the same concepts as were in NXT, but got to do more things they could relate to. They made sprites have conversations with one another (using the kids' recorded voices), built games, and were constantly calling each other over to say "look at what I did!". A girl, whose parents firmly don't want a computer at home, built an animated summary of the first chapter of her favourite book. Boys wanted to make guns that shot bullets.

I think there's a lesson here: doing something in hardware isn't automatically cool, particularly for kids. It's harder to make things happen, so we veteran geeks get a thrill from it. We think that because it's physical, real, and a Robot, kids will automatically be excited. But for kids who are learning, and who don't appreciate the significance of the challenge, it's just hard and unrewarding.

This might change with age, but I think that even 15 year olds should have had some exposure to programming in software before they start on the hardware. The only way around that would be for the teacher has a huge toolbox of peripherals to bring in and make some more impressive demos than LEDs blinking and Legos rolling back and forward across the floor.

I've posted a longer report, including the non-programming bits of computer education that I did with the kids, on my personal blog. (updated to fix URL)

Radar Roundup: Ubiquitous computing (ubicomp)

• The Street as Platform (Dan Hill): amazing essay by Dan Hill (yet another genius formerly at the BBC) about the invisible cloud of data in a city street. "We can’t see how the street is immersed in a twitching, pulsing cloud of data. [...] This is a new kind of data, collective and individual, aggregated and discrete, open and closed, constantly logging impossibly detailed patterns of behaviour. The behaviour of the street."
• Service Design Notes: Tools, not Services (Chris Heathcote): frustrated by the limited functionality in his Nike+ because the service is intentionally feebly aimed at feeble "typical" consumers, Chris dashed off a quick rant about the trap of designing services. Users want tools, not services. And by building tools, you can build a service people want. The last two paragraphs are gold: "Tools will be bent and misused - which means you sell even more. And you don't have design in the usefulness - just find the useful functionality and package it up in an open-enough way to show possibilities."
• A Very Long Conversation with Dopplr's Matt Jones (SecondVerse): a long interview, mostly about design stuff, with Matt Jones. The bit that resonated with me was "Mother Box is not in the Box", which I translate as "you buy products that are front-ends to services", which is a short hop away from "these days a device needs a network to be useful". If you think of ubicomp as "it's about sensors, outputs, and computation", you can't forget the network that connects them all--and what life is like when that network disappears.
• Review of Everyware by Adam Greenfield (heyblog): a very detailed review of "Everyware" by Adam Greenfield. Matt Jones recommended Everyware to me as the first stop in a quick catchup of ubiquitous computing. "Everyware strikes a good balance between the impenetrable proceedings of the UBICOMP conferences and design writing. Adam expects the reader to get references to “Ctrl-Zing something away, “elevator pitches”, and “user experience” and something about how people behave with mobile phones".
• Being Human (Microsoft Research): subtitled "Human-Computer Interaction in the Year 2020". Love the "Transformations in Interaction" section: "The End of Interface Stability; The Growth of Techno-Dependence; The Growth of Hyper-Connectivity; The End of the Ephemeral; The Growth of Creative Engagement".
• William Gibson - The Rolling Stone 40th Anniversary Interview: "One of the things our grandchildren will find quaintest about us is that we distinguish the digital from the real, the virtual from the real. In the future, that will become literally impossible. The distinction between cyberspace and that which isn't cyberspace is going to be unimaginable. When I wrote Neuromancer in 1984, cyberspace already existed for some people, but they didn't spend all their time there. So cyberspace was there, and we were here. Now cyberspace is here for a lot of us, and there has become any state of relative nonconnectivity. There is where they don't have Wi-Fi. In a world of superubiquitous computing, you're not gonna know when you're on or when you're off. You're always going to be on, in some sort of blended-reality state. You only think about it when something goes wrong and it goes off. And then it's a drag." I linked to it from the first Radar Roundup but I know you skipped it, so I had to quote it all here. You made me do it.

Radar Roundup: Web

• Active URLs (Ned Batchelder): OmniTI have done something clever with their URLS—turned them into active verbs: their testimonials are at URLs like /helps/ning, their jobs page is at /is/hiring, etc.
• What's This Fascination with Ad Networks (John Battelle): I had breakfast at ETech with jbat (who runs the Web 2.0 Summit) and got a braindump of his thinking around web-based advertising. His end-game is fantastic: brand marketing as a value-adding interactive experience on the net rather than a proliferation of small text splash. He's beginning to blog his thoughts in the leadup to his conference on Conversational Marketing (as he calls this interactive brand marketing).
• Debategraph: anyone who has tried to have a conversation online with a lot of people will welcome any attempt to bring order to the chaos. Debategraph is a wiki debate visualization tool with RSS feeds, open modification, and more.
• In Japan, URLs Are Totally Out (Cabel Sasser): the founder of Panic Software talks about a trip to Japan where he realized nobody gives URLs any more, instead they show the search terms in a searchbox that will give their site. I saw this last year in New Zealand, where an airshow was advertised with huge black and white signs that just read "GOOGLE AIR SHOW". After briefly being confused ("man, is there ANYTHING that Google isn't doing?") I figured it out. Only problems I can see are that you're at the mercy of PageRank or you are committed to outspending whoever else wants to buy that keyword.
• Mail Trends: utility built for GMail but extended to any IMAP-capable server that lets you graph trends and activity in your email. First step to seeing a mail program that gives us insight into our email. Next step is to have the mail reader use that insight to manage our inbox. Are you listening, Mozilla Messaging? (see also 21 ways to visualize and explore your email inbox which has a high number of spam-related visualizations but is still worth checking out)

Disks have become tapes

Via Michal Migurski's delicious feed, I found Disks Have Become Tapes, a fascinating set of observations about why MapReduce is so successful. In short (as observed by Doug Cutting at OSCON last year), MapReduce hits the sweet spot by operating at the transfer rate of disks (growing at 20%/year) rather than seek rate (growing at 5%/year) as relational databases do. It also mentions the growing field of column databases, which we're watching at Radar (e.g., Michael Stonebraker's appearance at MoneyTech) as driven by capacity growing faster than transfer and seek. Absolutely fascinating stuff.

Subscription Music: Which Is It?

The question is: given Yahoo!'s decision that DRM subscription music doesn't work, why does Apple think they can make it work? The lifetime payment idea sounds a reasonable one: the iPod has a slightly higher cost but it comes with infinite downloads. Most important about that model is that it hides the payment to music companies; remember, by misinterpreting p2p ("the market wants convenient digital music" was interpreted as "some dorks are stealing our stuff!") and then by acting like bullyboys as they futilely chase the people they wish were customers, the labels have created a generation that expects music to be free.

Then again, it may well be that mobile devices operate under different economic rules than desktop ones: in New Zealand, Vodafone sells more music singles than any other outlet, including the iTunes Music Store (that fact's in the comments of the article, not the body). I run a very dumb phone because I enjoy being away from email, and I'd buy a bridge before I bought a music single on my phone, so I'm probably not the best person to assess the viability of the phone. I realize this is probably like asking, "why do kids these days dress so funny?" but I have a question for readers: can you help me understand why people pay money for music on a phone when they already have a laptop full of music?

Simplicity

I got a chuckle out of this comic on app simplicity and usability. So true, so painfully painfully true.

Recent Posts

• Radar Roundup: Brains on March 17, 2008
• Radar Roundup: UI on March 11, 2008
• I Thought You Guys Were Supposed To Be Utopian: The EFF at Etech on March 4, 2008
• Radar Roundup: Data Mining and Visualization on March 2, 2008
• Radar Roundup: Collective Intelligence on February 26, 2008
• Radar Roundup: Bio on February 25, 2008
• Radar Roundup on February 23, 2008