Sebastian Thrun leaving Stanford to focus on Education Start-up

Sebastian Thrun, one of the Stanford professors who recently taught the online Artificial Intelligence course course with over 100000 students, has decided to leave his tenured job at Stanford to focus on his education startup Udacity.

On his web site he states: 

"One of the most amazing things I've ever done in my life is to teach a class to 160,000 students. Volunteer students translated some of our classes into over 40 languages; and in the end we graduated over 23,000 students from 190 countries. In fact, Peter and I taught more students AI, than all AI professors in the world combined. This one class had more educational impact than my entire career."

"Now that I have seen the true power of education, there is no turning back. It's like a drug. I've just peeked through a window into an entire new world, and I am determined to help bring education to everyone out there." 

Incidentally, one of the first courses offered by Udacity uses exactly the approach I have long been pushing at the Universities I have worked, teaching computer science and software development holistically and building courses around ambitious real world challenges: 

Udacity's CS 101 course is described as follows:

"CS 101: BUILDING A SEARCH ENGINE: Learn programming in seven weeks. We'll teach you enough about computer science that you can build a web search engine like Google or Yahoo!" 

(P.S. Thrun retains a non-tenured research professor at Stanford and appears to keep his job at Google)

New York City gets a Software Engineering High School

Even though this post is about software engineering it strongly relates to our new Software Entrepreneurship course at Oxford University. The MBA's we are going to teach at Saïd Business School this summer would be a lot more tech savvy if they had gone to a high school that teaches software engineering early on. Maybe we would have more MBA's interested in Software Entrepreneurship than Banking ...

This from Joel Spolsky's blog: 

"New York City gets a Software Engineering High School
by Joel Spolsky
Friday, January 13, 2012
This fall New York City will open The Academy for Software Engineering, the city’s first public high school that will actually train kids to develop software. The project has been a long time dream of Mike Zamansky, the highly-regarded CS teacher at New York’s elite Stuyvesant public high school. It was jump started when Fred Wilson, a VC at Union Square Ventures, promised to get the tech community to help with knowledge, advice, and money."

More here. See also Fred Wilson's blog. 

Finally, as suggested by Fred Wilson check out Mayor Bloomberg's State of The City Address.

"On January 12, Mayor Bloomberg delivered the 2012 State of the City Address at the Morris High School Campus in the Bronx. In 2012, New York City will lead the way by pushing progress in city schools to the next level, making the economy a global capital of innovation, and making the government the most innovative of any in the world:

• Citywide economic growth will be facilitated by expanding industries, creating jobs, connecting New Yorkers to job opportunities, and increasing the minimum wage.
• Innovative solutions to government will be implemented across all city agencies in order to streamline operations and better serve New Yorkers."
• New York City will improve schools by attracting, rewarding, and retaining effective teachers with programs that include loan forgiveness, increased salaries, and new methods for teacher evaluation. Successful charter systems will be expanded, students will be better prepared for college and careers, and the City will help students claim federal financial aid for college.

Software Entrepreneurship at Oxford University

Later this year Jyoti Banerjee and I will be teaching a new Software Entrepreneurship course as part of Oxford University's MBA program at Saïd Business School. Today we presented the course to prospective students and are now waiting to see how many will sign up. 

Here is a short excerpt from the course description:

"The Software Entrepreneurship course focuses on the intersection of software and business, and investigates the theory and practice of creating and growing successful software firms. The class combines topics from innovation, entrepreneurship, business planning, finance, business management, design, software technology and engineering to provide students with a holistic understanding of software business in general and the software start-up process in particular.

The course is designed for students interested in understanding and practicing innovation in the fast-changing software landscape. Lessons from this course will not only be valuable to budding software entrepreneurs but to everyone who wants to understand and influence how software transforms business and society. The course will in equal parts explore business and technology aspects of software - this is in realisation that successful software firms are often characterized by a close alignment of technology and business innovation."

The Oxford Software Entrepreneurship course is based on previous courses that Jyoti and I have taught independently at various occasions, but this is the first time that we will be teaching a course together. 

Course details are not yet available online so please contact me if you want to know more.

New Project on Behaviour-Driven Computing

Today the University signed the final paperwork for GAMBAS, a soon-to-be-starting FP7 project (http://www.gambas-ict.eu/) I will be involved in. Our work will focus on mobile context-aware user interfaces.  Project partners are Universität Duisburg-Essen, National University of Ireland, Galway and The Open University (+3 industry/public organisations).

"The overall objective of the GAMBAS project is the development of an innovative and adaptive middleware to enable the privacy-preserving and automated utilization of behavior-driven services that adapt autonomously to the context of users. "

Some links on IT and Computer Science Education in the UK

For anyone who cares about computer science education, here a few links: 

"Michael Gove admits schools should teach computer science."

http://www.guardian.co.uk/technology/2011/dec/06/michael-gove-computer-science

"On Monday 28 November UKIE announced the launch of the Next Gen Skills campaign to call for fundamental changes to the education system to drive hi-tech growth."

http://ukie.info/content/next-gen-skills-campaign-launched

"New Gen Skills is a major new campaign formed from an alliance between the biggest names from the UK digital, creative and hi-tech industries and the UK’s leading skills and educational bodies to improve the computer programming skills needed for the future growth of the UK’s economy."

http://www.nextgenskills.com/

"This landmark report sets out how the UK can be transformed into the world’s leading talent hub for video games and visual effects."

http://www.nesta.org.uk/publications/assets/features/next_gen

Animal-Computer Interaction

Clara Mancini, a colleague of mine at the Open University, recently published a Manifesto on "Animal-Computer Interaction" (ACI). The manifesto describes the scientific aims, methodological approach and ethical principles of ACI and proposes a research agenda for its systematic development.

What is ACI? A quote from the abstract:

"Although we have involved animals in machine and computer interactions for a long time, their perspective has seldom driven the design of interactive technology meant for them and animal-computer interaction is yet to enter mainstream user-computer interaction research. This lack of animal perspective can have negative effects on animal users and on the purposes for which animal technology is developed. Not only could an Animal-Computer Interaction (ACI) agenda mitigate those effects, it could also yield multiple benefits, by enhancing our inter-species relationships with the animals we live or work with, leading to further insights into animal cognition, rendering conservation efforts more effective, improving the economical and ethical sustainability of food production, expanding the horizon of user-computer interaction research altogether and benefiting different groups of human users too." Advances in both our understanding of animal cognition and computing technology make the development of ACI as a discipline both possible and timely, while pressing environmental, economic and cultural changes make it desirable. But what exactly is ACI about and how could we develop such a discipline? This Manifesto describes the scientific aims, methodological approach and ethical principles of ACI and proposes a research agenda for its systematic development."

Check it out and let me know what you think about ACI.

Mancini, Clara (2011). Animal-Computer Interaction (ACI): a manifesto. Interactions, 18(4), available at http://oro.open.ac.uk/28857/

Helsinki Design Lab: "Recipes for Systemic change"

New book by Helsinki Design Lab: "Recipes for Systemic change". Describes design process for Strategic Design. http://helsinkidesignlab.org/instudio/

"Helsinki Design Lab helps government see the 'architecture of problems.' We assist decision-makers to view challenges from a big-picture perspective, and provide guidance toward more complete solution..."

Leaving Lancaster University for The Open University

Effective 1 Sept 2011 I have taken a new positon at the Open University (www.open.ac.uk) as Professor of Ubiquitous Computing. I will be setting up a ubiquitous computing resarch lab around the concept of 'ubiqutious computing for a sustainable society'.   

Lancaster Open-Data Workshop (Mar 21, 2011)

Together with Julian Tait of FutureEverything, I am organizing an Open Data workshop on March 21, 2011 from 13:00-16:00 at Lancaster University. If you are interested in attending please sign up. 

Overview

This workshop will explore initiatives around the recently founded Manchester Open Data Store http://datagm.org.uk/ (see also a recent article in the Guardian). The data store already contains some details about public sector spending across Greater Manchester, transport, crime data, and locations of recycling centres, schools and GP surgeries. It will also contain data about education, the economy and health. 

This workshop is an opportunity to gain hands-on experience with data and to collaboratively explore ideas and innovations around open data.

Workshop Facilitators

Julian Tait (FutureEverything)

Gerd Kortuem (Lancaster University, www.kortuem.com) 

Prelimnary schedule:

1. The Manchseter Open Data Store. Presentation by Julian Tait.
2. A closer look at open data. Participants will collaboratively explore concrete data sets from the Manchester Open Data Store to identify opportunities and discuss potential issues (data quality, privacy, ownership etc.)
3. Idea Generation. Participants will collaboratively generate ideas for data-driven products and services and discuss to what extent current data sets are suitable or appropriate to support these products.  

Who should attend?

Anyone interested in open data.

Connected Marketplaces for the Internet of Things

A while back I asked: Can market-based mechanisms encourage user innovation in the IOT? My answer is yes, even though we have not yet seen many examples. Details can be found in a recent paper:

Gerd Kortuem and Fahim Kawsar. Market-based User Innovation for the Internet of Things. Internet of Things 2010 Conference (IoT-2010) Nov 29 - Dec 1, Tokyo, Japan. (A corresponding presentation is available on slideshare or as pdf.)

Here the gist of the argument: the iPhone powerfully demonstrates how user innovation can be fostered by open markets and market-based mechanism. By combining programming tools, application platform and distribution channel, Apple has created an environment that effectively supports user innovation networks in the sense of von Hippel*, in which innovation development, production, distribution and consumption are performed by users (or more precisely by user/developers and micro software firms). The user innovation network supported by the iPhone ecosystem is horizontal, where innovation – in the form of iPhone apps – is created by and for users. In contrast to von Hippel’s original notion, which refers to open-source development and the ability to replicate and adapt a product, the iPhone innovation network does not compel users to make their innovations openly accessible to other users. Instead, transfer of innovation among users is facilitated by a two-sided market (realized by the App Store), with user/developers on the one side and users-only on the other. 

However unlike the iPhone ecosystem, the Internet of Things cannot be confined to a single device platform and a unified distribution channel. Instead, the IoT ecosystem will necessarily consist of a heterogeneous collection of hardware, software and data components. This greatly complicates user innovation as it introduces dependencies and compatibility issues, which make it harder to share and reuse artifacts. Thus in order to foster user led innovation in the IoT space I argue for a connected set of marketplaces, each one addressing a particular innovation touchpoint. I define marketplaces as connected if products of one marketplace can be used to enhance, control or interact with products of another marketplace.

A simple example of this concept is www.liquidware.com, an online shop for open-source DIY hardware. Liquidware not only sells hardware but also offers an App Store for software that runs on this hardware. Following our definition, the Liquidware hardware store and the Liquidware App Store are connected. In order to be traded in connected marketplaces, products need to be compatible: in this example software in the one marketplace needs to be compatible to the hardware in the other. Another form of connection can be envisioned between a marketplace for sensor devices and a marketplace for sensor data produced by these devices. However, there is no example yet for such a link (even though simple forms of data marketplaces exist, for example as part of Pachube). Connection is a one-way relationship and connections between three or more marketplaces can be complex. For example, two sensor device marketplaces could be linked to the same data marketplace. Marketplaces may also be chained: a marketplace for electronics components may be connected with a marketplace for sensor devices built from these components, which in turn could be connected with a data marketplace.   Connected Marketplaces for the Internet of Things

Applying the concept of connected marketplaces to a future smart buildings ecosystem we can envision a set of connected marketplaces as follows: 

1. Smart Object Marketplace
2. Application Marketplace
3. Configuration Marketplace
4. Data Marketplace
5. Data Manipulator Marketplace

The details of these marketplaces are described in a recent paper: Gerd Kortuem and Fahim Kawsar. Market-based User Innovation for the Internet of Things. Internet of Things 2010 Conference (IoT-2010) Nov 29 - Dec 1, Tokyo, Japan. 

 A corresponding presentation is available on slideshare or as pdf.  Connected marketplaces create an open ecosystem that supports innovation and diffusion across multiple levels of complexity.  An innovator can use lower-level marketplace to acquire devices and tools to build something more complex and use higher-level marketplace to share (or sell) his/her creations with others, who in turn can use them as a starting point their for their own innovations. This mechanism not only supports an innovation chain from low complexity to high complexity, it also allows for a distribution of ownership and control of marketplaces. 

---

*von Hippel, E. (2002). Open Source Projects as Horizontal Innovation Networks - By and For Users (June 2002). MIT Sloan Working Paper No. 4366-02)

Interaction across Objects, Time and Space

Ubiquitous computing and the Internet of Things represent a radical departure for the design of interactive systems. We are moving away from the 1:1 interaction model where one user interacts with one device at a time, to a n:m model where one or more users are in constant and seamless contact with many devices. The 1:1 model is exemplified by devices like a PC and a mobile phone. The new n:m model is exemplified by interactive public displays, smart furniture and future smart physical objects.

Two question now arise:

• How do we design user interfaces that support interactions that cross objects, time and space? 
• How do we design interactive systems that support interactions that cross objects, time and space?  

In a recent paper entitled Supporting Interaction with the Internet of Things across Objects, Time and Space and presented at the Internet of Things 2010 conference at Tokyo, my colleague Fahim Kawsar and I tackled the second question. In the paper we describe a software architecture for flow-based interactive systems. A flow, or more precisely a situated flow, is a high-level activity model that is linked to physical entities and digital devices.

Situated Flow

At runtime the model maintains the state of the interaction between one user and one or more devices, making it possible to carry interaction state from device to device. A corresponding software framework provides the foundation for flow-based interaction and device co-ordination. 

The work was done in a hospital context, so our application examples relate the use of smart medical devices in a future care scenario. Details can be found in the paper: 

Fahim Kawsar, Gerd Kortuem and Bashar Altakrouri. Supporting Interaction with the Internet of Things across Objects, Time and Space. Internet of Things 2010 Conference (IoT-2010), Nov 29 - Dec 1, Tokyo, Japan. 

The corresponding presentation slides can be found on slideshare.

Internet-connected Bike Sharing System by RAFAA

An exciting new design for a bike sharing system for Copenhagen was developed by RAFAA, a design and architecture firm based in Zurich:  

This Bike Share System is more than just a transporting system. It deals not only with the problem of stocks and flows of people, but adds extra value to its user and to the city itself. We suggest that the Bike Share System becomes an integral part of the city. The bicycles should function as censors [I suppose this should mean sensors GK] and inform the system about certain behaviors, so that the system can react according to the situation. To predict the performance of a system, the entities have to exchange information. An internet-based platform can analyze the different interests and could then manage possible conflicts. The bicycles are equipped with GPS and W-Lan, so they are connected to each order and can inform the system about their position and status. (Is a bike being used? Where is the bike and where is it moving to? Is there a reservation for the bike? etc.) Privacy protection is a matter that has to be taken into account in the process. [http://www.rafaa.ch/rafaa/Copenhagn.html]

A bit on the speculative side in terms of technology and functionality but I really do like the design. 

New Bike Share System in Copenhagen by RAFAA (photo copyright by RAFAA)

New Bike Share System in Copenhagen by RAFAA (photo copyright by RAFAA)

New Bike Share System in Copenhagen by RAFAA (photo copyright by RAFAA)

User Innovation for the Internet of Things

In May 2010 I a gave a talk at the CIOT workshop at Pervasive 2010 in Helsinki, which explored the questions: "What can the Internet of Things do for citizens?" I just uploaded my presentation slides (available as pdf and on Slideshare). The talk was based on a paper by my colleague Fahim Kawsar and myself. 

User Innovation for the Internet of Things | Gerd Kortuem

In my talk I discussed how user innovation and market-based innovation can be combined to create user-centered ecosystems that are open for and provide incentives for end-user innovation. The talk was motivated by the observation that traditionally the development of the Internet of Things has been driven by large commercial players, leaving not much room for small independent players or non-profits. This situation is changing, however, as the success of the open hardware movement and the work of design firms such as Tinker London and Berg London testifies. 

A workshop summary has been published in the IEEE Pervasive Magazine. 

Digital Object Memories

I recently was involved in organising a workshop on Digital Object Memories in the Internet of Things. Unfortunately I was unable to attend myself, but the position statements, technical papers, and design studies are interesting enough (available online). 

At the workshop participants created memories for objects they had brought to the workshop - by linking 'video memories', photos or text documents to these objects using QR codes. Memories created during the workshop are available at the Tales of Things website. 

Digital Object Memory Workshop in Copenhagen (DOME-IoT 2010)

The notion of object memories of course is not new. In his book Shaping Things Bruce Sterling elaborates his SPIME concept: "A SPIME is, by definition, the protagonist of a documented process. It is an historical entity with an accessible, precise trajectory through space and time." Information shadow is another term for essentially the same concept (for example described by Adam Greenfield in his book Everyware and related to Alan Westin's "data shadows" concept - hat tip to Mike Kuniavsky and his book Smart Things: Ubiquitous Computing User Experience Design.)

Digital Object Memories comprise hardware and software components that physically and/or conceptually associate digital information with real-world objects in an application-independent manner. Such information can take many different forms (structured data and documents, pictures, audio/video streams, etc.) and originate from a variety of sources (automated processes, sensors in the environment, users, etc.). If constantly updated, Digital Object Memories over time provide a meaningful record of an object's history and use.

QR CodeFrom a technical point of view, Digital Object Memories provide an open-loop infrastructure for the exchange of object-related information across application and environment boundaries. Besides fostering information reuse and reducing the risk of information inconsistencies, they allow for novel classes of applications in which rich object histories are created and exploited.

From the user's point of view, Digital Object Memories create a new design space for everyday interactions. Physical objects could become sites for their owners' personal stories, but also afford people the opportunity to explore an object's provenance and connections to other elements of physical and digital life. In this sense there is the potential for designers to augment or even transform our relationship with objects and the services that they mediate.

Tales of Things, an experimental website for people to upload object memories, is an attempt to realise the object memory notion and make it available to everyone. It will be interesting to see if sites like Tales of Things will one day become as popular as photo albums were years ago. 

My slides from the Internetome 2010 conference in London are now online

My talk from Internetome 2010 in London is now available online, at slideshare and as pdf. I discussed design issues of smart connected objects in the context of the Internet of Things. In particular I discussed three case studies, one from interaction design, one from contextual design and one from information design. 

Internetome 2010 was a great conference. I'll post more about it in a few days.