Blog (page 56)

XTech, the European web technologies conference, has picked The Ubiquitous Web as its 2007 theme:

As the web reaches further into our lives, we will consider the increasing ubiquity of connectivity, what it means for real world objects to be part of the web, and the increasing blurring of the lines between virtual worlds and our own.

The call for participation is now open; topics we find especially interesting are mobile devices, RFID, user interface, web apps, and of course where this is all going.

Alongside Adam Greenfield of Everyware and Gavin Starks of Global Cool, S&W will be keynoting. See you there!

Web apps are currently undergoing a renaissance–or perhaps they’re fulfilling the promise made when the genre was created in 1999. The technology, skills and community that go to make these web apps is beginning to turn in many different directions. We’ll soon see a number of different web app species. One I find most exciting is Deploy to Desktop. What if the same skills needed to build complex web apps could be turned to making desktop applications, starting from a simple web app in a HTML renderer window, and iterating to use native widgets, drag and drop, and full OS integration? (More about this in my App After App talk.)

We’re on the way there. Three data-points for that journey:

Apollo is Adobe’s cross-platform runtime, based on Apple’s WekKit, that lets you run HTML/CSS/AJAX apps on the desktop. It works offline, includes an API for communication between Apollo apps, and will let you write database hooks to a local or remote persistent store. The Apollo wrapper will be distributed free, like Acrobat Reader or the Flash Player (personally I think this is the wrong model–apps should be standalone, but we’ll see). Some Apollo screenshots.

Next is WebKit on Rails which is exactly what I wanted to see when I gave that talk. It makes it easy (well, easier) to take your Ruby on Rails web app, wrap it in WebKit, the Mac HTML renderer, and run it as a desktop app. See the list of existing projects for applications you can already download.

Last up is Pyro, which wraps 37signal’s Campfire browser-based live chat application and turns it into a Mac app. Features include a badged application icon (the number of unread messages are shown), drag and drop upload, scripting support and more. Someday all web apps will be available this way.

This post is going to be about objects that celebrate their functions. This was an area of research for me during my time at the Royal College of Art. I’m going to follow on from Matt’s post on Disco and intrinsic activities. More show than tell here I think.

Here is my favourite piece of video right now. It is from the film 9 and a Half Weeks (via James Auger), and if you can wade your way through Rourke and Basinger power bonking their way around Manhattan you see this tape deck in his apartment. I’ve looped the video a couple of times and slowed it so you can see clearly.

I’m pretty sure it is a Nakamich RX tape deck. Using a system called UDAR (UniDirectional Auto Reverse) it mechanically flips the tape over at the end of each side. Something to do with aligning the heads. It is a fantastic piece of perfomance, and completely intrinsic to the nature and qualities of tape decks. Whatever it’s functional relevance might be, witnessing a mechanical operation so performative is excellent, the object is so discreetly joyful about what it is doing.

I also came across this video of a Red Raven records vinyl (via Alex Jarvis) on Kempa.com, along with some lovely research on vinyl video. It has two components. One is the vinyl, which has a large area of printed imagery on the larger than normal label; the second is a sixteen sided mirror which sits in the middle of your turn table and works like a zoetrope, reflecting the images on the vinyl as it turns and creating animation.

This a is beautiful response to the intrinsic qualities of vinyl and the mechanism of the record deck. More products should include this sort of wit and performative funtionality.

The RFID hacking workshop last week was both thoughtful and productive. I’m going to scoot through what we made and a few of the ideas, and leave any more detailed thoughts for other posts.

Day 1 was introductions and learning the technology. I’ve had a vague idea what RFID was before, but now Matt Karau told us all about it. So what is it? A powered RFID reader reads tags. An RFID tag is one of these:

Actually, that’s a smartcard together with an RFID tag–it’s from the inside of an Oyster card, the thing you use to pay on the London Underground. The antenna has broken off. RFID tags come in two flavours:

• Passive tags have an antenna and a small chip. The RFID reader sends a burst of power at the tag; the power runs through the antenna and powers the chip; the chip does something (maybe just loads an ID, or perhaps does a tiny calculation); the tag sends the data back to the reader and powers down. It might have a little memory too–perhaps 0.5-4Kb.
• Active tags are as passive ones, but with a battery.

Tags have large antenna loops, and are generally sealed inside a disk of plastic or paper. There are complexities, of course. RFID tags can be joined to sensors, so they report environmental data, or have more complex chips capable of running a tiny OS and cryptography applications, like the Oyster cards. There are different frequencies, different ranges you can hold the reader to invoke the tag, and different standards…

…but, essentially, the basic thing we’re using is a loop of wire that somehow reports the same number each time when you ring it with an electromagnetic field generated by the reader.

(Also on day 1 we had a go at calling the tags “spychips” instead of “RFIDs” every time we referred to them. The specific privacy fear isn’t a view I subscribe to, but it’s enlightening to see your new ideas being inflected by the language you use to reach them. I think it helped us see RFID as a technology in its own right, rather than relying on a single, badly-fitting metaphor.)

Day 2

We started considering the range of RFID on the second day. These thoughts, about how to have interactions in a sphere of thin air a few inches around a hidden tag, led in part to last week’s post on RFID and forced intimacy. They also led to this:

(Thanks Timo for the photo.)

In the palm of each white glove (Jack and I both have one) is an RFID tag. Inside the white polystyrene box is an RFID reader hooked up to a microcontroller (on an Arduino board). The board is also hooked up to a vibrating motor.

When you put your hand near the block, it begins to rumble.

On my glove – I’m on the right of the photo – is a flex sensor, wired up to the controller. The more bent the sensor, the bigger the rumble of the box. So when you approach the reader with an open palm, there’s a gentle vibration. As you make a grabbing gesture, the vibration grows and the white box begins to lead about. It makes a fair racket.

Why do this?

Jack was keen on celebrating the magic of this kind of remote action. What if the vibrating motor was actually a toy car motor? You could approach a car, and push it with your glove, imbuing it with acceleration through empty air. By clenching your fist, it’d zoom faster! You’d have to chase it to keep it moving. With two gloves and two motors, you could control it too.

Are proximity and tensing the hand the correct interactions for these kind of toys? We can only think by making.

Day 3

We’d spent the second day trying to learn some of the intrinsic properties of RFID tags and readers. Some of the ones we discussed were:

• You can’t see the tags
• Tags are ways for machines to tell the difference between different lumps of plastic–heading towards what Matt Jones has called a robot-readable planet
• Knowing a tag’s ID is proof that the reader was geographically there, like the patterned card punches used in orienteering
• The functional bit is the reader, not the tag

This last point was a revelation to me. RFID interactions are not like button-pushing interactions. With buttons, the smarts are behind the buttons themselves. The buttons trip relays and activate switches. With RFIDs, the smarts are all in the thing you use to push the buttons.

Imagine a computer keyboard, but the keyboard isn’t plugged into anything. Instead, cameras in your fingertips read the letters on the keys, decipher them, and send those letters via a USB cable in your wrist to the computer. That’s more like RFIDs. With buttons, the button pusher is the fungible bit (any finger can be a poking device) and each button is special. With RFID, the button pusher needs to be clever.

Continuing this thought, imagine a keyboard which worked like this. You’d print out paper ones that were better for Quark or Photoshop or whatever. You’d draw ad hoc macro keys on the desk, in erasable pen. Taking this back to RFIDs, does their true potential only emerge when people can make and write their own copy-tags to fill their environments?

Anyway.

Since the reader is the functional bit, different tags can behave in different ways. Perhaps the reader could have a slot in the top, and that’s where you slip in a tag to state the kind of tool the reader is at the moment (a telephone, a camera, a query tool), and there are other kinds of tags that represent objects, like people, places and things. What about stacking tags, or having them interlock in different ways, or… or…

The possibilities multiplied.

On the third day, we came back to where we’d started on the first day: Trying to get a feel for the field of an RFID reader.

Above is a drawing Timo made, with the RFID reader under a desk and a pen through a hole in the middle of an RFID tag. He drew only when the reader was picking up the tag ID. It’s a beautiful image. (Thanks Timo for the photo. This was when we were seeing if a magnet would distort the field. Other drawings didn’t have that in place.)

Jack and Timo spent time in the workshop at the Architecture School making a robot pen to do the same thing.

This pen has an embedded RFID tag (Timo’s photo again). It’s wired to the RFID reader, which controls a solenoid pushing the pen up and down. The pen is only down, and drawing, when it’s within range of the reader. You slide it around, and the automation does the rest.

The machine-aided drawings aren’t as beautiful as the totally hand-drawn ones, but ain’t that always the case.

So I’ve been thinking about hands and arms. I started by thinking of extremely small hands, on my hands. So here are some drawings from that thinking.

This drawing is of a toy that shrinks your hands down so you can play in a small world, with small figures. Your fingers are all connected up to a group of flex sensors, which converts the analogue movement to a cluster of servos. The servos collectively control fingers on the small hands by tightening or loosening. So the movements of your fingers become roughly and awkwardly analogous to those of the small hands in the toy. There is also a screen inside some goggles hooked up to a small camera in a glass ball between the two small arms. So when you look in to the goggles, you see what is in front of your arms. There are two wheels which you can twist to point the camera in different directions, like an eye. Kind of like an analogue version of virtual reality, only right in front of you and not virtual.

I would also like to have a very small hand at the end of my finger. To pick up pens and things. You control the small hand on one finger using your other fingers, with flex sensors (same as above). You lose one of your big hands to gain a little hand on the end of one of your fingers.

I came across Chad Thornton‘s work. He is at Google now, but he made a mechanical finger as part of his work at Carnegie Mellon Interaction Design programme (nice video here).

Maybe I’m carrying some latent affection for the Radio Shack Armatron here, I don’t know. These themes are common in films. This must be informed by Ripley’s Power Loader from Aliens:

The belt buckle, and rubberised keyboard make her rig seem really convincing, her trainers too, and how she locks into the unit. The cyborg fingers for typing in Ghost in the Shell are nice too.

No doubt there are more. It makes me think of Robocop‘s gun hip too although slightly off topic.

I like them, robot arms. I see them as a celebration of industrial process. I predict they will become a more widespread part of our lives. They are cheaper now (it appears that non-load bearing ones don’t require three phase power either) and since they are multi modal they can perform many tasks, in strange contexts. No doubt FDM or other fittings are/will be available, implications of that could be very large. Imagine a robot arm in your drive thr(o)u(gh), changing a tyre, and then printing out your happy meal. Our lives could become peppered with arrays of multi-buildy-arms.

Robotlab (via Roger Ibars) are a German partnership who have used industrial robot arms to perform a DJ set. Witnessing the arms is as important as their role. I find them disconcertingly accurate, mechanised confidence in something typically so analogue and expert and careful. There is also something about their inflexibility, their inability to reach inside certain arcs, too close to themselves. I like the way they occasionally find a sync with each other, and at other times drift out. I think these guys have a business model set up around this, so I’m very interested to see how that develops.

I want one.

S&W is here in Oslo with Timo Arnall’s Touch project for an RFID hacking workshop (check out that hand-drawn antenna field map). Yesterday was introductions, learning about RFID as a technology, and some preliminary explorations.

The work group met for breakfast today, and we discussed promising interactions and potential projects. One of the topics that came up was RFID tag visibility.

I know it’s obvious to state it, but RFID tags are generally hidden. To read a tag, first you have to find it with your reader. Design can make the location of the tag obvious… but wouldn’t it be interesting to embrace the invisibility constraint? Could we take advantage of the seeking behaviour that has to occur?

Consider a car showroom. Imagine no salespeople there, and no prices on the cars. When you entered the showroom, you would be given a RFID reader. There would be an RFID tag, holding the price, hidden somewhere on each car. You’d have to find the rough location of the tag to read the price.

Okay, this would be enormously annoying. But it would force you to step closer to the car, to examine the wing mirror to see if a tag was there, get up close to the paint-work on the door. What would happen?

When you get bright lights and noises in films, you feel excited whether the narrative of the film is exciting or not. Playing Project Rub on the Nintendo DS, the blowing interaction forces you to get adrenalised. Your body is tricked. Maybe getting really close to the car would be a kind of forced intimacy. You would feel better disposed to the car whether you liked it or not, simply by virtue of almost hugging it.

Okay, that’s car showrooms. What about parties for teenagers? We made up a Spin the Bottle type game. Oh, and gave it a pirate theme.

The scenario is this: Everyone who comes to your house party gets a token that looks like a gold coin with an RFID tag in it (holding a unique ID). People at the party take turn with the RFID reader. They have to wear a pirate hat, and the reader looks like a buccaneer’s sword. Let’s say you have the sword. You press a button on the handle, and it chooses a random RFID tag. This is the tag you have to find, by going round to everyone at the party and sweeping them with the sword.

Now let’s say you’re a person with a coin. If you’re not too keen on the person wearing the pirate hat, you’d put the coin in your pocket, or under your collar, so it could be found quickly. But maybe if you fancy the person with the hat, you’d conceal the coin a little, to make it harder to find. Gosh. I think it could get a little bit sexy.

I like this game because it celebrates the invisibility of the RFID tags, the fact they have a short detection range, and that the range can be shortened with material. It’s a treasure hunt game, but it doesn’t matter whether you have the chosen coin or not–it can be flirty in any case. It supports social interactions (ahem) rather than displacing them.

We had other ludicrous game ideas: Croquet where the hoops had tags and the balls had readers, where the balls would speak what you had to do next. “You have taken 4 hits. Now get through hoop 2,” your ball would say. You wouldn’t need the rule book. We also considered playing penny football with RFID tags, the readers snapping onto the edges of the tables and recording the number of goals. They could show the score on LCD screens, and play cheering sounds whenever a goal was scored.

Both of these games feel as silly to me as Slapz, the electronic game that replaces the children’s game Red Hands (or “slaps”).

The forced intimacy treasure game feels just as silly, but much more fun.

When I wrote about the Coke Happiness Factory ad, I mentioned the activities that happen around a product just because it’s something that’s bought and sold. Examples included selecting, purchasing and showing off.

Often the intrinsic activities that surround a thing are ignored. This is a shame. They’re opportunities for design to communicate brand and celebrate the constraints.

That’s why I’m totally in love with Disco, the new Mac CD/DVD burning app.

It can take ages to burn a disk. Your intrinsic activity is waiting. What does Disco do? It puts a fluid dynamic smoke simulation on top of the window (follow that link to see a movie). And get this, you can interact with it, blowing the smoke with your cursor.

You can watch, you can play. Inspired.

First talk!

I visited the London office of Agency.com a week last Friday and reprised Engaging Technology. The gags went down okay – something I’m always nervous about, especially with an end-of-the-working-week audience – and I made a few small changes, mainly to focus the Acts Not Facts slide on interactive agency work. I said:

When I buy a holiday, Expedia makes it feel like I’m engaging with tedious bureaucracy. My ringing phone embarrasses me when it rings on a train or in the cinema. But when I purchase something, that’s not just a cold fact… it’s the first time a product and I engage, and if the purchasing experience is lousy then the brand is damaged. […] And when I’ve talked to advertising planners and folks working in brand communication in the last few months, they’ve all told me that the days you want a product in your life because that product is “cool” or “reassuring” or has a particular lifestyle, whatever the fact… those days are gone. It’s all about the acts instead. What’s it like to purchase? To show off to friends? To sell? To clean? To run in? What are the stories? How do I engage with it, and how does it engage with me?

I keep meaning to post more about how to design for intrinsic activities. Remind me.

Second talk!

This Saturday, I spoke at an away-day for the folks behind BBC Radio 4’s The World Tonight (my slides aren’t online). It was rather disconcerting to see the faces of voices with which I’m extremely familiar–I catch the programme almost every weeknight. I was speaking about the leading edge of media consumption, especially new media, covering old and new media channels (giving some demos–I think hands-on is important in understanding threats and opportunities), social media, citizen journalism, and a little trend speculation on where people will be spending their time in coming years:

Virtual worlds such as Second Life and the creative renaissance evident in Make magazine and more (the example I gave was Instructibles, where the empowerment of creation from the internet has come together with the loosely-coupled collaborative ethos)… both of these feel as energetic and full of potential as blogging did in 1999.

Thanks, too, to Dan Hill and Suw Charman, who were both generous with their help when I was putting together the talk.

I wanted to mention, here, how I concluded the presentation. I’d begun by stating that while technology changes, people mostly don’t. At the end I highlighted one social change that, from my perspective, looks as if it might be taking place:

Far from being an antisocial medium, the internet is enormously social. We in this room are media literate, from being surrounded by tv, cinema, radio, magazines, adverts… People growing up online are surrounded by people, and are socially literate. They’re fully in-tune with small-p politics. Interpersonal politics. Social politics. Sometimes it shows as a lack of respect, because they know how people work and aren’t willing to think that some people are special, just because they have authority. They understand that people are just people. […] There is a small-p political literacy that comes from continually socialising. It manifests both as people forming communities online, and as a lack of respect for big media that means media needs to personalise, and meet people in their communities as peers.

Highly speculative, of course! I also touched, briefly, on social capital.

It’s been a while since I felt like I was banging the drum for the internet. But being online is a large component of most of my friendships, and a lot of those friends I wouldn’t have if it weren’t for the net. Often people who have purely utilitarian use of the internet don’t see that.

I find it enormously heartening that there’s a general intelligence, online, about how people work in groups, gleaned from folks living on mailing lists, making stuff together, and chatting. I wanted to get that across.

Jacob Neilsen’s latest Alertbox Participation Inequality: Encouraging More Users to Contribute highlights the lurkers/participants ratio in social software:

In most online communities, 90% of users are lurkers who never contribute, 9% of users contribute a little, and 1% of users account for almost all the action

Neilsen points to less equitable distributions, and gives suggestions on how to increase the number of participants. Previously the Guardian have discussed this as the 1% rule.

It throws up a ton of questions. Is the growth of a social software system led by the number of participants, or the number of lurkers you can gather who somehow turn into participants? What are the distributions for different sites, and with what are these correlated? Are people who participate in one community more like to participate in another? Does having a consistent identity across services promote or damp general participation? With which distributions are the people lurking and participating most satisfied? How does this compare to television, newspapers, the postal service, the music industry and so on?

I don’t like the word “inequality” because it’s so loaded – the whole phrase seems to promote deliberate and vocal participation as a good thing – but I’d love to know more about distributions of different modes of engagement with communities, completely online and otherwise. Any suggestions for reading material?

I’m currently reading Dan Saffer’s Designing for Interaction. Well written, well structured; a good introduction to interaction design that’s centred on the web as far as I’ve read, and looks as though it’ll take a broader approach in the second half of the book. This passage, on icons, grabbed me:

A confusing image can obscure much more than it can illuminate. For example, the disk icon has come to mean “save,” even though, increasingly, many young people have never seen a floppy disk (except perhaps this icon!). Then again, it is difficult to imagine what image could replace it.

Well that’s a challenge if ever there was one.

People don’t use paper files like they use to, and besides, computers aren’t office focused but for the home now. And at home, it’s all about the media.

Could play, pause and the rest replace save and open?

We might have to twist the metaphors a little, but consider the regular set of icons (and I’m sure I’m only thinking of this because I’ve been reading about early play and record icons; more).

How about this: On file open dialogs, the play icon would replace the open button. On toolbars, the play icon would be used to trigger the dialog. To close a document, saving automatically, the pause icon would be shown.

The metaphor here is that a document is a continually evolving piece of media. There’s no concept of “save,” what you see on the screen is what’s on the disk. You can never lose work. Play and pause simply mean start and suspend editing.

Ah, but what if you wanted a safe and stable version of your document to always come back to? That’s what the record icon would be for. It’d be analogous to “save as,” kinda, but a better description would be that record means “tag this as stable” in version control. You’d still have your continually evolving document, only the recorded one would be tagged as a version to rely on–an inflection point. This would tie nicely into Apple’s Time Machine, which will let you leaf through previous versions of your files–maybe rewind and fast-forward could be used to step between stable versions.

When I had a dual cassette player, years ago, the second tape deck had a record button for deck to deck dubbing. It only worked if the first was playing. The record button, in our new system, could also be attached as a label any place the currently playing document could be channelled. So record would also show up next to the printer, to dub the open document to paper, and it’d be on the email application, and next to IM buddy names too… it’s a bit of a stretch, this one, but I’d like to try it.

Naturally stop would send your document to the trash.

This change in metaphor, from document as a discrete thing to something which is continuously changing, would affect much of the desktop GUI. Rather than an application which has the command “send to printer,” it has to be a representation of your printer which has the record button on it.

And what does “now playing” mean, given our computers multitask? Perhaps it’s a good thing for us to think of them playing documents over one-another, in a cacophony. It means we’d start thinking about which we focus on, and how to quieten the others–would we have a “stifle” button, to suppress alerts from playing documents that aren’t at the pinnacle of our focus?

Documents as music. What else could replace the disk picture on the open button?