Sunday, 17 February 2008

TED Talks: Ben Dunlap - The story of a passionate life

Wanted to share this with you (transcript of the last seconds of Ben Dunlap's talk), it is an amazing talk:

"And I realized in that moment of revelation that what these two men were revealing was the secret of their extraordinary success, each in his own right.

And it lay precisely in that insatiable curiosity, that irrepressible desire to know, no matter what the subject, no matter what the cost, even at a time when the keepers of the doomsday clock are willing to bet even money that the human race won't be around to imagine anything in the year 2100 - a scant 93 years from now.

"Live each day as if it is your last", said Mahatma Gandhi, "learn as if you live forever".

This is what I am passionate about. It is precisely this. It is this inextinguishable, undaunted appetite for learning and experience no matter how risible, no matter how esoteric, no matter how seditious it might seem. This defines the imagined futures of our fellow hungarians, Robeczeck, Teszler and Bartok. As it does my own, as it does, I suspect, that of everybody here.
"




The file for download as well as comments on the video can be found on the TED homepage, here.

Saturday, 16 February 2008

SemanticCampParis Atelier Découverte

Disclaimer: I'm live blogging, excuse lapses of grammar, spelling errors, and typos.

Session by Alexandre and Gautier. Excellent introduction to the subject! Thanks guys! I especially liked the part where you showed how search queries on Wikipedia could be done!

1. RDF Model

What is RDF?
"Resource Description Framework"

Idea: html relate documents between each other, via hyperlinks. RDS shall relate objects between each other.
Two concepts behind it: Triplet and URI

Triplet: (subject, predicate, object)
Example: "Gautier" "a pour nom" "Gautier"; "Gautier" "est de type" "personne"; "Gautier" "connait" "Alex"
Could be described as a type of underlying grammar.

URI: (Universal Resource Identifier)
To identify and localize objects in a network (an URL for example locates an object with an http address, an URN could locate a book in the isbn "framework")

Namespace (In XML, a namespace is a collection of names, identified by a URI reference, that are used in XML documents as element types and attribute names. )

What is missing? A syntax. There are different syntax models available. n3, nTriples, turtle, rdfxml.


2. RDF Vocabulary
- About RDF ontology: read.

3. SPARQL. Language & protocol for databases requests.

Good example: www.dbpedia.org (to build a request: http://dbpedia.org/snorql )
Gautier shows an example. Detailed description of the use case here (http://www.lespetitescases.net/sparql-maillon-essentiel-web-of-data) on his blog

4. RDFa
RDFa (express triplets in HTML) vs. Microformat (... is a piece of mark up that allows expression of semantics in an HTML web page. Programs can extract meaning from a web page that is marked up with one or more microformats).
Gautier has an article on the topic (here) on his blog.

Great to have learned a bit about the technological basis of the semantic web. Thanks again!

Semanticcampparis: presentation of participants and planning the program


Semanticcampparis @ La Cantine.

In the end around 20 people showed up (participant list here). Charles Nepote did a quick introduction and after, participants introduced themselves and explained their interest in the semantic web. (I explained that I'm especially interested in the possibilities for web apps, the use rather than the technological base and what the implications are for the future of the internet).

This camp shall help to start coordinating the efforts for the french speaking community of people interested in the semantic web.

The areas of interest that emerged:

1. Animation (coordination of efforts within France and french speaking regions of the world)
2. Translation & Ontology (how to organise and coordinate the translation efforts)
3. Classification and structure by themes/topics/subject
4. Atelier decouverte (getting to know the models behind the semantic web)
5. Websem Zen Garden, Showroom (similar to the idea of an css zen garden: helping to make the topic more understandable by using concrete examples)
6. Knowledge databases
7. Semantic wiki
8. DOAP equivalent for research projects
9. Distributed social networks

Friday, 8 February 2008

LIFT08: The Hedonistic Company

Holm Friebe and Phillip Albers from the socialistic-capitalist joint-venture Zentrale Intelligenz Agentur in Berlin will show new forms of cooperation and collaborative work in todays' society.

"A good name, website and business cards" is all a company needs.

Seven (or so) rules of working together

Rule 1: The 7 NOs
No ... office, employees, fixed costs, pitches, exclusivity, working hours, bullshit.

Rule 2: Work-Work Balance
Engage in client work and self induced projects. "Don't let your own projects be corrupted by commercial work"

Rule 3: Instant Gratification
Use money as an incentive. Distribute immediately after job.

Rule 4: Pluralism of Methods
Find technical solutions for social problems and use online tools of collaboration instead of meetings. Skype, Google Calendar & Docs, Lesemaschine (collaborative editing tool)

Rule 5: Fixed ideas
Live up your intellectual ideas and dark desires at work.
Powerpoint Karaoke


Rule 6: Responsibilities without hierarchies
Each project has to have one person in charge but it can be anybody. Organize majorities for your ideas.

Rule 7: The Power of Procrastination
Don't try to be too efficient. Good ideas will adapt and catch on, even if you neglect them for a while.

Rule 7.5: Marketing by Feuilleton
... changing a company to a hedonistic format

Links:
ZIA's Collective Blog: Riesenmaschine

LIFT08: Kevin Warwick

From the program:
"Kevin Warwick, the British "cyborg" researcher who implanted a microchip in his body will tell us about his pioneering experiments. He had a neuro-surgical device implanted into the median nerves of his left arm in order to link his nervous system directly to a computer. The aim of these experiment series is to assess the use of latest technologies for disabled people."

Kevin Warwick on Neural Implants
"Why not link our brains together and transmit ideas, colors, ... "




Thursday, 7 February 2008

LIFT08: Rafi Haladjian

From the LIFT program: "Rafi Haladjian, founder of Minitel start-ups and internet/wi-fi providers in France will tell us about his latest venture: Violet, a french tech company, has designed the Wi-Fi and RFID enabled rabbit called Nabaztag."

"Violet's Strategy"

  • affordable products with one at a time buying decisions
  • fun & interpersonal communciation are as powerful propositions as usefulness
  • design matters
  • make it simple
  • a new image of technoloy
  • open standards
  • empower the user
  • build a community to provide services and help generate killer apps
2003: the DAL Lamp
2005: Nabaztag

Why a rabbit?
With a rabbit on the desk the day they decided to do the new product (and have ears, no voice, they multiply, etc). In the end they wanted to make a statement: if they could connect a rabbit, they could connect anything.

What does he do?
  • Reads short reports (weather), RSS feed headlines, etc
  • emotional messenger, physical avatar of friends
  • ear movement chat
Nabaztag and Gallimard Jeunesse (a french editor) joined to let theNabaztag read books. This is done by letting the rabbit "smell" the Ztamp (RFID tag) on the book, whereafter he starts reading it.

2008: connecting the rest of the 10.000 things at home ("the strategy is: first to connect rabbit, then connect everything else).

LIFT08: User experience session

11:30 at LIFT Conference.
From the program: "We will have two anthropologists, one working for Nokia in Tokyo (Younghee Jung) and another for Intel in Seattle (Genevieve Bell). Both will show us why observing users of technologies is important and how this brings insights for designing future products.

Paul Dourish, a researcher from University of California Irvine will then criticize how such results from anthropological studies are often translated into "implications for design", missing relevant insights from anthropologist's works."


Younghee on Nokia Open Studio: Vision by Relevance, Design for Use.

Younghee asked the questions "What does it take to host a design competition"?

The design competition was about imagined uses and designs of mobile phones in poor and emerging countries.

The design competition forms part of and supports ethnography research and street surveys.

> Local teams
> 3 locations: India, Brazil and Ghana
> Challenges: weather, security, electricity


Some of the participants' ideas:

India (Mumbai)

  • Cloudy-Buddy Mobile (a mobile that you just have to point to the sky to forecast the weather). Why was that entry chosen? Very special was that it was a woman and the high weather dependency
  • Eco-Cell (measuring air pollution so that people are more aware of environmental problem)
Ghana
  • Four Star (host up to four separate SIM Cards, because of the four mobile carriers in Ghana).
  • Combi (loudspeaker, to share media)
  • Shape plays a major symbolic role. Durable, rugged mobile phones as phones are used for a long period of time in these communities.
Brazil (Rio)
  • Mothers checking on kids as there are many underage mothers leaving their kids at home to go to school.
  • Relying on solar charging.
  • Pen drive phone (the only device to store electronic personal documents and information)

Genevieve on "Secrets, lies & digital perceptions"

A couple of my takeouts from Genevieve's speech:
History and cultural background of lies & secrets:
  • Part of the job of the 3o sociologists at LIFT is to "find out what people could do with technology and how to make sense of it in their lives".
  • For every mechanism that was created to e.g. locate someone with GPS, services were invited to deceive those services.
  • Trends in people's habits and daily lives change far slower than technology.
  • Cultural practices on lies and secrets: we tell somewhere between 6 and 200 lies per day. Men tell 20% more lies than women.
  • "All information being available equally" as a fairly new notion of the internet. However many communities are not willing to give up on secrets.
  • What appears to be in the public domain needs a lot of knowledge to actually fully understand the meaning behind an image of a neighborhood, a piece of art, etc.
  • Online, people are celebrating secrets and lies

Paul on etnography & design
Got a bit tired, a lack of coffee, fresh air and something to eat. So just two links to researchers Paul talked about

Nancy D. Munn: aboriginal navigation
Liisa Malkki: national identity




Wednesday, 6 February 2008

LIFT: Tool to Manage PFFAP (and GTD)

The am workshop I attended today (actually already yesterday now). A more detailed description can be found here.

What's PFFAP...
Presence, Flow, Feeds, Activity and Peripheral Awareness (and Getting Things Done).

And what about the tool?
Trying to use the tools that are "out there on the net" (see further below for review of prior art) for use in a work environment

Kick-off
Lee Bryant, from Headshift, kicks off with his introduction presening his teams knowledge pyramid (the social "stack", see a detailed description of the social stack here) and the six goals that the tool (or an integrated solution combining several tools) might help to achieve:

Picture taken from the Headshift website.

1. "How do we break the inbox addiction"
2 ."More flow, less sequential processing"
3. "More human scale, informal sharing"
4. "A better personal radar"
5. "Negotiating shared meaning and language"
6. "Use attention feedback to drive relevance"

Review of prior art:
Tom Taylor & Colin Schlüter give an overview on different tools that are currently being used (Twitter, Tumblr, iGTD, NetNewsWire, E-mail).

Discussion then moves on to the topic of "How the decision making process is affected when we use social tools and when we don not use social tools". This is just an unsorted list of thoughts people uttered.

With social tools:
includes marketing noise,
synchrony of discoveries (e.g. digg),
group think,
possibility to discuss in an asynchronous way,
good for information gathering and backing up the decision,
logged conversations allow others to be involved

Without social tools:
using personal criteria makes it easier (faster),
involves more perceived risk,
is based on trust.

Several groups agreed on the following: Decisions are influenced most by the person who's "there" the longest (especially in corporate context), although there might be people using social tools who could (and try to) add relevant information to the decision process. In the end the decision however is often taken on a gut level. Complexity of the decision process and time constraints play a big role as well.

Talking about complex situations and the related analysis:

Recognition Primed Decision Model

  • Situation analysis: The decision maker interpretes cues from the environment.
    Currently forms that are used by social tools to gather information restrict the input of much of the side information (that however is really important to understand the situation) gets lost.
  • Pattern recognition: The decision maker analyses the cues from the environment and compares them against stored cues.
    Cognitive abilities are not supported by the tools at the moment.
  • Idea generation: The decision maker generates workable solutions to the problem at hand
  • Solution selection: The first workable solution is selected. The solution is defined, as being workable if the simulation generated poses no problems.
  • Simulation generation: The decision maker constructs a mental scenario of the proposed solution. Any problems are then addressed.
  • One more point that I unfortunately missed.

Developing a feature list
(that we can derive from the goals for social tools in the enterprise)

4 questions and 4 groups:
  1. What do we want to pipe through?
  2. How do we wan to access and transform this info?
  3. What sort of global actions do we need to support?
  4. What are the key features of the tool?
ad 1. "Work / play" switch or master switch (on/off)
"Context" switch"

ad 2. "Ranking by peers" (not digg style but implict, by reacting on it)
"Emerging peer groups": people who have read this item also subscribe to this RSS Feed

ad 3. Collect, process, organise by priorities, postpone, delete, relate, delegate, involve, follow the info.

ad 4. Resilient and agile
Context specific
Multiple interfaces ... and some more

Design brief and wrap-up
Several propositions on the how the tool could look like:
  • just a search box. Searching lenses around topics, created by discussions by people you know, in the background. Those lenses can be made up of feeds, discussions via Skype, documents, twitter updates, social bookmarks, e-mails. Basically everything that is currently used for searching and organising information,
  • "workstream"(in comparison to lifestream, timeline)
  • flexibility of formats - converting e.g. e-mail to SMS
  • leverage e-mail as de facto standard for conversations
A couple of good posts over at Ewan's Edublogger:

LIFT: First Day Workshops

Good morning,

today, two workshops on the program here at LIFT:

As another option for the afternoon there's Dave Brown's "Generation Y and technology" workshop, he asked the LIFT students to join his session.

More on the workshops later on here.
Looking forward to it!

Sunday, 3 February 2008

LIFT 2008: Way around Geneva

Next week will start off quite intense:
travelling to London on Monday morning, back to Paris Tuesday midday with 45 minutes to change from Gare du Nord to Gare de Lyon, catch the train to Geneva (arriving there at 21:30). Once there, I'll try to find my way to City Hostel Geneva, drop my stuff and meet the other students... for a glass of wine (something to eat - as I'll probably be starving after all that travelling) and to get into LIFT mood.

And then ... well, just looking forward to all that's coming up at LIFT.

I found great hosts in Geneva (via Couchsurfing), they will be my home from Wednesday through Saturday.

As there's not one map on the LIFT homepage where all the venues are in one place, I just created one to get a better overview where the event will be going down.
Blue's for Wednesday, Red for Thursday and Green for Friday.


View Larger Map