Last: January 2004
Next: March 2004
February 25, 2004
rdf as barcodes, and geowarchalking
This is a summary of the talk I gave at conconuk, with a few more references, examples and links thrown in.
In my talk at Etcon (Powerpoint, 1.2Mb), method number 28 to find your location was geowarchalking. This originally was a simple idea of people marking postcodes and lat/longs around the urban environment in chalk, similar to the wi-fi markings - making the invisible visible.
These methods are still hard to get in and out of an electronic location system. You don't really want to be typing in numbers to 8 decimal places, or triple-tapping postcodes. Machine understandable is not machine readable.
So how can we make this easy and automatic - RFID tags will soon be available in spray cans, anyone can sticker the city.... with barcodes.
An aside: Japanese mobile phones. Many modern mobiles read 2d barcodes, in a standard format, using the on-phone camera. These barcodes include a phone number, a URL, or an email address. This has been expanded to include a whole business card. People print them on their paper business cards, adverts often have a barcode link.
But - we want to store other data. We want latitude and longditude. We want places. We want lots of things - we want to barcode tables, chairs, everything! How can we create all these specs?
Most of them exist already. There are RDF schemas for position (geo), places (locative) and even homepages (FOAF). If you want to describe something else, there's an entire dictionary in wordnet and a poor description of how things work with cyc, and, to be honest, RDF schemas aren't that hard.
So, the idea: RDF as barcodes. I'm proposing to use the same standard for barcode encoding as used in Japan: QR Code. It's scalable (up to about 4kb), it's in the public domain, handles odd characters (such as Japanese scripts) and it copes pretty well with being readable when the barcode is damaged.
Here's a geo tag (the example from the geo spec) encoded as a QR code:
(the black line is because I'm using a demo version of a barcode creator - I think)
Now, it might be worth being a bit pragmatic, and not including the RDF blurb at the beginning (i.e. just encoding the actual geo packet). This might actuallly be valid RDF these days - I know the validator has this as an option.
This reduces the size of the barcode considerably:
The other thing to add is a human-readble indication of what you're encoding - e.g. geo, locative, foaf. This means you can work out what kind of thing you're scanning or looking for.
And just to prove how crazy this can get, here's the description part of my FOAF file:
And if you include all my email aliases, it gets to a point where I don't think cameraphones could cope:
So, let's annotate the planet!
What we need next -
a standard proposal (this post may even be it)
free, open reader and writer software - both for PCS and phones
crazy geo pirate graffiti gangs
Some questions that arose at conconuk -
Why not use RFID?
Too expensive, and not generally available - all you need with barcodes is a printer
Why not just use a pointer or a URI?
This isn't just for connected devices such as cameraphones. Your digital camera might use a geo barcode to geoposition your photo. Your GPS might learn about places and things via barcodes. The system's a lot more hackable and open-ended this way.
The other thing I haven't really given is services that might use such a system. This is just an enabler, a piece of glue that relates computer and Internet data to the physical world. I'm really interested in everyone's thoughts, problems, pitfalls, ideas.
(I'll leave the reverse idea - barcodes displayed on mobile phone screens and read by an environment of devices for another time)
UPDATE:
A few people (hi joshua) mentioned by email and at conconuk some other points -
Why not use n3 triples?
I thought more software and hardware would understand some form of XML rather than n3 (I have never seen an n3 reading app outside RDF testbeds).
Why not zip/gzip the data first?
A pretty good point, and something I'm going to investigate. I'm worried that this is introducing more complexity, and things to go wrong. I want to see if doing this makes the barcodes less fault-tolerant.
Some links:
DoCoMo barcode spec
an english explanation
QR Code official site
Japanese software to create QR Code barcodes: QR Factory QR Manager
general barcode creation software (demo)
good overview of all 2d barcode techniques
Some examples of the smaller geo packet in other barcode formats -
data matrix
maxicode
(registered at the Post Office as a "dangerous British ammunition export")
link | Comments (10) | TrackBack (1)
February 22, 2004
"excuse me, did you guys just lose your little world?"
the waitress said. It was late Thursday night, and everyone was hysterical and overemotional. She didn't realise she was a surrealist mindbomb. Yes, yes we had just lost our little world. Not the rubber globe thing she was holding (never trust a company that gives away promotional juggling balls), but Etcon has finished - we'd survived - but now we had to get back to reality.
A hard cold British reality would have been too much for me, so instead I spent a few days in San Diego, and am now sitting in LA worrying about the upcoming British reality.
(illustrations accompanying this story available here)
I couldn't really get a grasp on San Diego. Downtown was dead and deadly, especially the infamous Gaslamp quarter. The food was good, the price was right, but it didn't amount to anything. I went to Old Town, "birthplace of California", and found the state park was just a bad garish reconstruction full of tacky tourist shops. Then I went to Fashion Valley, a shopping mall that screams of alternate states of reality.
Horton Plaza was the same, a new level of hell constructed for a city populous. Seaport Village is a vacuous tourist reality that prevents fun unless it serves Mammon.
A few things changed my opinion of San Diego: the zoo, Balboa Park, and neighbourhoods such as Little Italy and Hillcrest.
The zoo was great. It didn't seem as cramped as London Zoo, the animals seemed to be ok, and it's a nice walking environment. Balboa Park is full of stolen architecture styles, and some great museums. I am particularly enamoured by the model railway museum, and the outside pipe organ - both are labours of love, maintained by the public. There are very few ways people can actively help museums in the UK, so all the volunteers and docents I saw at the museums fills me with joy. And you haven't really lived until you've heard Copland's Fanfare for the Common Man played on a very loud slightly out-of-tune open air pipe organ.
Little Italy and Hillcrest suffer from grid system sameness - a highway to cross every few steps - but at least the shops, the restaurants, the people had character. Areas are more than neighbourhood names.
How do you make an American laugh? Tell them you don't drive. After a while, the laugh disapppears. You really mean it. You don't drive. Can't drive. _Choose_ not to drive. You become some anarchist tendency, some freak of nature. Forget pacifism, homosexuality, liberalism, those can be worked around, forgiven, forgotten. But you're offending gasoline. You're offending the US.
In San Diego, whilst walking around for three and a half days, maybe 10, 12 miles in total, I passed about 80-100 other walkers. I can do that in a minute in London. In Britain, driving somewhere is a major event. You drive somewhere and then walk in that area, maybe even take public transport from then on. Whenever we think about walking, Americans think about jumping in a car. The system is set up this way. Every shop, office, restaurant, cinema... whatever... has it's own parking lot. Have you remembered to get your parking validated? Violators will be towed.
And so I get to LA, a city that sprawls from the air, and makes less sense on the ground. I had positioned myself very carefully - opposite the Farmer's Market in mid-Wiltshire. A lot of things I wanted to see were walkable (i.e. under 3 or so miles), and public transport connections weren't far.
The Farmer's Market is like crack for me - tens of places to grab good food, both ready to eat and cooked. It reminds me a bit of Borough Market and Spitalfields, but slightly sanitized, and catering more for (driving) passersby - the local working and living population. Damn good french dip sandwich though, and a fantastic slice of pecan pie from Du-pars.
Next door is the Grove mall. It's the opposite of the San Diego malls, welcoming, friendly, engaging. The most crazy feature is a trolley that runs from the mall to the Farmer's Market - at most 150 metres away. People queue and use it. This is not a walking culture.
Mall design appears to be what theme park imagineers do when they grow up. Even the most understated has loud quirks. The Grove has the tram and a huge model of a wind up soldier. The Beverly Center has a Pompidou-like escalator to take you to the 5th floor before you even get to any shops. The mall on Hollywood is, well, what you'd get if Elton John's interior decorators were forced to create a family retail experience.
None of them are really what I'd call destinations. Many have less than 30 shops. But still, people drive and park, shop and get validation. The parking lot is often larger than the mall. It's the same at individual stores, people park up, walk across the sidewalk, shop, get back in the car, drive to the next store, even if it's just up the street.
This drive-consume-drive culture is difficult for museums and galleries. LA has some outstanding collections, but it's hard to pursuade people to spend more time thinking and less time driving. There's also quite a hard class culture barrier. Art is seen as high art, appreciated by the few. I'm not quite sure why this is, as the institutions themselves are falling over backwards to be accessible, exciting and inspirational.
The Getty Center is a bold statement. It's art as a destination. Something you have to actively visit and take part in, from its remoteness to the monorail forming a mental barrier between visitor and car. it's a varied collection, from National Trust recreations of rooms, through British Museum ethnography to V&A collections of design and National Gallery Old Masters. There's not much modern work, apart from the photography collection, so the most interesting to me was the design elements - illuminated manscripts, stained glass, furniture.
I think I enjoyed the collection at LACMA more. Again it's a very liberal collection (arts rather than art), but there was a great modern selection, and the Chinese and Korean sections were fascinating. The most striking aspect of both museums is the architecture. The Getty is a huge construction that's very sympathetic to the art (there's some very good curation going on), and manages to hold together as design even though it is 6 large square building blocks.
The Japanese pavilion at LACMA is just beguiling. It's extraordinary from the outside, just about connecting with the right-angled main museum buildings, but the inside is just amazing. A series of ramps and curved spaces display the art, the walls appear to be tatmai screens, diffusing the light. Most amazingly, water and waterfalls providing a distinct serenity that binds the architecture together and benefits the exquisite detailed Japanese pieces.
The Walt Disney Concert Hall is the latest piece of art to arrive in LA. I liked it a lot more than I expected, mainly because the inside works almost as well as the exuberant outside. Originally this building was going to be built in stone, but after the success of Bilbao, the project team demanded that Gehry redesign the building in metal (this building uses stainless steel rather than the titanium of the Guggenheim). The building is quite spectacular in bright sunlight, even producing hotspots and sun traps on the sidewalks. The part of the design that doesn't really work is Bruce Mau's wayfinding. The use of one font and weight throughout is pretty but quite useless, compounded by the decision not to have any obvious signs (all wayfinding is through inscriptions painted directly on walls) and the donor branding of every space and staircase.
The concert hall was all I saw of downtown - I've learnt my lesson. Petula Clark was wrong. Downtown in the US means a place to work, and leave as quickly as possible at night. I got on the first bus out (the Speed-inspiring number 10 to Santa Monica) and wandered down the beach to Venice.
There was nothing that particularly inspired me about Santa Monica, apart from being able to reach a beach from most of the city in under an hour, and being able to walk along the beach with a little bit of sun, even though it was cloudy, and February. Sun and warmth, available in a city year round, always gives me pangs of jealousy.
Time was running out - there was just enough for a visit to the Museum of Jurassic Technology, and then a trip along Hollywood to check I wasn't missing anything (I wasn't).
The Museum of Jurassic Technology is just the most crazy, inspirational, insane, creative, freakazoid, well designed errr... 'thing' I have seen for a very long time. Pictures do not do it justice (as it's very dark in there), words can only brush the surface, although the book Mr Wilson's Cabinet of Wonder has given me the same gleeful smile as going to the museum itself. The museum is a testament to serendipity (and missed chances), fanatical extremes of human mundanity, and unknown/unpopular science, coupled with complete belief, authority, and irony. It's slightly Fortean and slightly true.
Do explore the online texts of most of the current exhibitions, and if you're in LA, ever, move hell or high water to get there. The biggest weirdness is that the Museum is settled between an In n Out and a vacant row of shops, as though it crashed down from space, and no one much noticed.
Nearby is the Center for Land Use Interpretation, which unfortunately had just closed, but is meant to do some great work looking at human geography, and also provides an example of a place publicly stating their location.
And that was it. Yes, I have a similar pang as last year; California does have some attractive attributes (which, I'm sure, would disappear quickly once you live there). I enjoyed LA as much as SF, even though everyone said I was insane to try visiting without a car. Heck, I'd even go as far as saying LA has a pretty efficient public transport system - and I admit that's probably because I didn't try to go really long distances, or to anywhere really difficult. So, all in all, I'm looking forward to going back at some point. I feel I've only just scratched the surface of LA, and I'd love to explore more.
link | Comments (2) | TrackBack (0)
February 19, 2004
mass amateurisation of commercials
In the UK, we've had the rathergood Switch Maestro adverts, and the weebl-and-bob-as-cows Anchor Butter ads, both originating from b3ta.
In the US, there are the Flash-copying Quiznos Subs ads*. Has the American media been farked, or are these a conniving ripoff?
(click for less moire)
*with a great offer - take in a coupon for absolutely anything from any store, and get a buck off a sandwich.
link | Comments (1) | TrackBack (0)
February 13, 2004
35 ways to find your location
Well, I think the talk went well.
The presentation is here (Powerpoint), but in the next few days I'll annotate it will all my notes that I didn't use, and turn it into something HTML or more accessible.
Many thanks to all those I have hassled for help or criticism over the last 6 months, and thanks to those that asked questions and provided comments.
It's been a great conference, and I'll need a bit of time to decompress and start making sense. There's lots of great stuff going on, and it's mindblowing to be part of it.
link | Comments (0) | TrackBack (0)
February 11, 2004
Social software apps using mobile phones
Social software apps using mobile phones
ITP
dodgeball.com
text messaging app - 5 cities in the US
location based service
tell us where you are, and we'll tell you who and what is around
focus on building things that real people can use now
(uses sms to email??)
dodgeball guide - user generated content on venues
circles - friend finder application
go onto the website to setup circles of friends
send your location by text, finds people nearby and send them a message that you're there
scout - relationships based on proximity
listen to events in the area (e.g. 10 blocks)
why is it different?
opt-in location tracking
focus on landmarks, not street addresses
take advantage of near ubiquity of text messaging
only push content that people can act on / always relevant
"start thinking of phone as more than just a voice tool" - [heh, we're now trying to go back the other way]
***
fiasco
street game for mixed physical/digital play
interests:
mixing on and offline communities
opportunities for revitalising public space
fiasco:
people take over street corners by participating in small pieces of street theater
make alliances and enemies as jostle for space
object is to dominate the map by putting one's dignity on the line!
design research
'turf' - street as territory
tagging - personal mark making, identity in public places
street games - hopscotch
psychogeography - meaning and places, street as narritive
public performance - guerilla theatre
fun - santacon
alert given - 20 mins, Tompkin Sq Park
the stunt - player has to top previous player's stunt
elements from website: object from city, urban situation, action
friend takes photos and uploads
website acts as amplifier, and sends out
other players then rate against reigning player
after a week, the challenge is closed, and map is updated
being built atm, roll out in spring
will people actually play the game?
how does conventional mapping change with social dynamics
capture richness/riskyness of physical world in game design
***
modus
problem - sharing personal media
http://stage.itp.tsoa.nyu.edu/~dc788/spring2003/netobjects/modus/
music selection in public spaces is not in the hands of the people
traditional jukebox is too passive
digital distribution systems are short-sighted
expedience, discovery and distribution
website - upload songs to a bar
username attached
personalities created over time
and how taste of music evolves
bar code reader on jukeboxes, given users cards, and user can swipe cards if they hear something interesting
how can you get people the music and still be legal?
tied into dodgeball
when you say you're somewhere, jukebox pulls up your songs
***
interactive telejournalism
low cost live interactive news tv production
utilises 'freely available' broadband wireless networking
streams from telejournalists
comments from audience relayed back to journalists
push the meaning of interactive tv
tv isn't going to go away
utilise stregths of internet and strengths of television
anytime, anywhere news gathering
http://www.walking-productions.com/itj/
****
Q: how do you deal with nonstandard vocabulary - e.g. names of places?
street addresses were really hard to parse
strips out spaces and punctuation to try and match
refines with user if there are similar results
Q: how will the interactive tv deal with quality?
more a homegrown CSPAN than a news show
more human may make it more engaging with viewers
role of the anchor becomes something different - good interfaces for the moderator
link | Comments (0) | TrackBack (0)
Enabling 'Life Goes Mobile'
Enabling 'Life Goes Mobile'
Pertti Korhonen
CTO, Nokia Corporation
3 themes:
mobility
- and what makes mobility different?
groups
- how does peer to peer relate to mobility?
openness
- need developers to help innovate
Mobility
Tied to a screen - living through a screen much more than we would liek to
Interacting with the world vs. interacting with the screen
devices:
More flexible, more immersive displays
Bandwidth increasing
Multimedia capabilities
Small screen usability is key - very hard to see an overview. New era of complexity and functionality - and opportunities.
Tangible interaction - bring touch into the picture
People used to tactile and audio feedback. Touch will bring new interactions.
Mobile means travelling through multiple domains of life - work, home, travelling, socialising. Context of use is a great oppurtunity.
User expection of simplicity. More complex the systems, more important to simplify for the end user.
Has to work - cannot crash. Expectations of always being available. This relates to security - need ubiquity for trust.
How open and hackable can we make the mobile platform and still be reliable?
Two ways to approach this -
take a PC and shrink it into a small form factor
(our way) take a mobile device and grow the functionality and platform
can't/mustn't lose simplicity and intuitiveness plus flexible and personal
don't forget the creativity of end users in getting things done - self-expression (covers, ringtones), downloads, adding skins
encourage human creation
Don Norman - social interaction is the killer application
'personal life recorders' - how can we share very personal media?
elements that drove PC inductry - hard drives, displays etc. is now applied to mobile
ubiquity of bandwidth benefits the end user - multi-radio - wi-fi, cell service, close proximity
plan for it or not - people will do amazing things with tec
hnology - hard to forcast, so just enable groups at the edges of networks. Don't make them ask for permission or hinder them.
Modish demo - mobile distribution and sharing
Openness - more devices, more developers, better products, open interfaces, easier integration
Convergence of mobile world and Internet world
Symbian - developing open standards based OS and middleware
Importance of usability and user experience
1500 applications for Series 60 this year
100 million devices with development platforms - Java or Symbian
Examples of apps
Worldmate - 5000 licences sold, 15000 downloads a month
Futurice PhotoBlog - we believe in blogging - uses Atom API
Support through Forum Nokia
1.3 million developers
Java is the prime end-to-end platform
working on scripting languages for Series 60 - making it really easy for fast app development
Python is the first, may extend to Perl.
(python demo of weather app)
Simplicity is the key theme. Have to address usability with more complexity.
Invisible experiences - hide complexity.
Build in open standards, not islands.
link | Comments (0) | TrackBack (0)
Catalyzing Collective Action on the Net
(and so, notes begin)
Catalyzing Collective Action on the Net
Marc Smith
Microsoft Research
monthly tree map of Usenet (last 50 months)
Social cyberspace is large and burgeoning environemnt, distinctly biological.
Ideas from sociology - collective acction through computing
What do people really need to interact successfully?
Finding your place in the hierarchy, and communicating that to other people.
social neighbours:
Social - ologist
Social - worker
social - izing
social - disease
social - problems
social - psychologist
social - netowrk
social - software
other names:
community - were they really communities?
group(ware)
networks
worlds
These get us into trouble. Even groups - what is a group? Probably not what we're finding online. (small <10, rather than online >1000)
Robert Axelrod - The Evolution of Cooperation
Elinor Ostrom - Governing the commons
we don't celebrate the commons or collective project any more
a successful culture lets people get things done collectively
Erving Goffman - The presentation of self in everyday life
Edward T Hall - The hiddden dimension
? - Social Network Analysis
Tufte (the church of Tufte - I'm a sinner in that church)
Marc Smith - Communities in Cyberspace
When is a group not a group? - Brian Butler
small population of people who often know who else is in the group
know that they are in the group
online aggregations resemble voluntary associations more than groups (Shriners, PTSA, AxAs)
boundaries are more porous and memberships less demanding
'virtual schelling points' - places where people congregrate, each with different affordances.
If you're 1 in a million, there are 768 of you on the Internet.
Yphrum's Law - systems that shouldn't work, but sometimes do. There are enough of them for some to work.
e.g. ebay, with relatively flimsy reputation system
Ostrom's design principles for groups to organize:
5 is the kicker - a system for monitoring members' behaviour exists; this monitoring is done by community members themselves
the goal: plumbing for collective action
what affordances or furniture do you need?
waste, resource networks.
scaffolding to support collective action
key tasks for participants:
discovery
selection
evaluation
motivation
clean data (e.g. A-Z restaurant names) isn't used, you'll use extra data - how does it look, is it busy, how does it smell etc. etc.
Usenet, is that still around?
2003 - 240 million messages
8.6 million unique authors
151 thousand newsgroups
the message - at the nub of newsgroups, authors and threads
threads do not exist within a newsgroup, they veer off
http://netscan.research.microsoft.com
Rich clients - treemaps, scatterplots, histograms, piano roll, soon thread visualisation
3 million identities , 10 million messages in the microsoft.* public hierarchy
a growing environment
related visualisation work -
shneiderman
doanth
viegas
sack
ducheneaut
venolia
binaries groups do not have regular authors, discussions groups (e.g. fan groups) do
in microsoft.*, chinese hierarchy has appeared and grouwn to be as big as other languages.
(bubble visualisation of a newsgroup)
shows a structure, sohws regulars, people who are going to be there in 4 years
without pictures, you could not seee the difference in how people interact in the groups
what does a person look like in cyberspace?
(histogram of an author's postings)
need to visualise to show things, such as trends in posting, hiatusses, initiator of threads or replier, flamer?
'answer person' - just replies once in each thread - member of tech support newsgroups
25,000 of 3 million in MS hierarchy are like this
Styles of participation are often stable across time.
Implicit vs .explicit
problems with explicit systems - 'are you my friend?'
statements about behavious and relationships
self-report data is bad - people have difficulty telling you what they had for breakfast
Implicit is better - observations about behavious and relationships
Now online is mainly inscription mechanisms, there's a lot more implicit data.
mobile devices as the new mouse
first waves of devices that integrate cameras, processor, network, display
the world is a web page - every object has a digital aura, a story to tell
applications: navigation, annotations ....
these things are getting small (e.g. rfid)
Swiss - spotme.ch
linux handheld for conferences, set what you like, and when like-minded people come together, they both ring
Boston - ntag
the display is for the other person (until you flip it up to read it)
hot stuff, dangerous stuff - 'i thought you said you didn't see david'
barcode route - semacode etc.
machine-readable tags
Aura: weaving threads into things
link online info to physical objects
create an app platform for tag-driven apps
a set of sample apps
social goals
link people through shared objects and places
2% of people will do a lot to help create systems
can get more info from tags, and start searching the net for more info
(e.g. tags in an art gallery)
easy to blog (harder to annotate - but the option is there)
isn't this just cuecat?
what's changed? mobile devices, open backend, public wireless broadband, enormous amounts of existing content to leverage
(most of the supermarkets in redmond have wifi)
use in supermarkets - get Google results from barcodes (e.g. food recalls)
labels are political (nutrition labels, art labels)
all this is stopped with things like Aura
collecting places where standard formatted data could be looked up (UPC numbers, VIM numbers)
Aura going public in a few weeks. (pocket PC w/wifi or phone edition)
easy to build a new resolution service
anyone could become a trsuted authority on a topic
create easily digested summary wisdom
annotation hosting and sharing
business model? sponsored links(?)
great for activism - and great for selling things!
next steps -
web apps to manage collections
address network speed/reliabilty
cameraphone barcode reading (hard) [no it isn't]
integrate photography into annotation tools
integrate more work on sensors and location detection
newsgroup
microsoft.pubblic.research.aura.discussion
try it yourself!
annotate the planet!
link | Comments (0) | TrackBack (0)
etcon: monday
Ok, slightly overtired... lots of others are taking great notes.
the grid
city of lights
UI for alien cowboys
collaborative mapping workshop
link | Comments (0) | TrackBack (0)
February 05, 2004
gaslamp
In preparation for Etcon, and to illustrate some of the location ideas I will babble about, I've hacked together gaslamp, a collaborative guide to downtown San Diego.
If you're a San Diegoan, or are there for Etcon, please explore, and add comments and places (if you've got a digital camera - take pictures of intersections and places too, I'd like to do something with them later).
It works on WAP phones too, so you can use it in the field. Try http://undergroundlondon.com/gaslamp/gaslamp.cgi in your WAP browser.
Let me know if you have any suggestions, improvements or ideas - there's lots I'd like to fix and implement, but it's just a quick and dirty prototype.
Data collected by gaslamp will be used as part of the collaborative mapping workshop.
link | Comments (0) | TrackBack (0)
February 03, 2004
dorkbot berlin
Subject: [dorkbotlondon-announce] dorkbot berlin tomorrow (monday)
>>Hello,
>>For those in Berlin, including those at the Transmediale festival,
>>there's the first dorkbot.bln at c-base (http://c-base.org) on Monday
>>from 7pm. Come over if you're in town.
>Strangely I am going to be in Berlin tomorrow.
