Accent français

Un de mes collègues à Empire N est l’heureux créateur d’une application qui vient enfin d’être rendue public: le Nokia Sport Tracker.

Vous pouvez désormais aller marcher, courrir ou faire du vélo, du roller, du ski ou quoique ce soit d’autre, et suivre le timer, la distance parcourue, votre vitesse en temps réel sur votre téléphone portable. Les informations sont conservées dans le journal d’entraînement. Vous pouvez même voir vos parcours avec Google Earth par la suite sur votre ordinateur.

Vous avez besoin pour celà d’un téléphone Nokia S60 3.0 et d’un module GPS Blutooth (ou d’un Nokia S60 3.0 avec le GPS intégré). Si vous avez par exemple un de ces modèles: N71, N73, N75, N76, N80, N81, N91, N92, N93, N93i, N95, E50, E60, E61, E62, E65, E70, E90, 3250, 5500 Sport, 6290, 6110 Navigator, downloadez l’application, c’est gratuit!

Si vous voulez acheter un module GPS, cherchez un Nokia LD-3W Module GPS Sans Fil (~100 euros) ou une autre marque devrait aussi faire l’affaire.

Je suis venue en Finlande il y a 2 ans 1/2 notamment pour travailler sur le projet MobiLife pour l’empire N. Le projet est maintenant fini. On a eu le dernier audit cette semaine qui consistait en 2 jours de présentations et de démos pour que les ‘reviewers’ du projet (3 professeurs et 1 personne coté commission EU) puissent faire leur rapport final.

Une des parties les plus visibles du projet a été peut-être l’application appelée Context Watcher. Celle-ci fonctionne sur les téléphone Nokia Symbian S60. Une fois démarré, Context Watcher envoie au server WEB les données de positionnement du téléphone – cette donnée est appelée cellid (c’est l’identification de la cellule GSM/3G du réseau où se trouve le téléphone). Si un GPS est connecté au téléphone, les coordonnées sont aussi envoyées. Que cela permet-il? De suivre avec plus ou moins de précision où vous êtes, et de partager l’information avec vos ‘buddies’ ou amis qui ont le Context Watcher installé aussi sur leurs téléphones. Vous pouvez aussi voir une carte des environs, avec les POI (Point of Interest) c’est-à-dire les restaurants, stations essence, supermarchés, etc. De nombreuses autres fonctions sont possibles: envoyer une photo sur votre compte flickr avec le titre, description et tag remplis en fonction d’où vous vous trouvez. Un Context Blog peut être activé, soit un blog dont les posts seront créés automatiquement par le serveur, et contiendront par exemple des statistiques de la journée: les endroits où vous avez été le plus fréquemment, les photos prises, les gens rencontrés.

Pour toutes les informations, télécharger le Context Watcher, voir des Context Blogs: http://portals.telin.nl/contextwatcher/
Présentations (en anglais):

• Context Watcher flyer
• Context awareness and management: présentation plus détaillée du Context Watcher
• Autre présentation du Context Watcher, ainsi que l’introduction du concept ‘context=data cluster’

There was a funny question from the EU commission reviewer this morning; it went more or less like that:

EU politicians are the ones to make the decision of investing EU money into research projects. After the projects are completed, these politicians want to know how fruitful the investment was, and so they ask project reviewers. Politicians want to know for example which research areas are not risk areas anymore, or became mature enough that the industry can alone take them further; these areas do not need public funding. Which are these areas? How do you evaluate the outcome of the investment? You can look at publications, patents that were made. But how can you connect really these to the work done in projects (and thus the money given by EU)? In the case of the telecommunication industry, more jobs were lost than created in the past couple of years. What kind of indicator is that? How do you measure overall the quality of research?

Well, good question! Unfortunately as easy to answer as whether believing in god is beneficial. Except maybe that for companies and state organisations the stakes and the pressure to find out how to make research useful are higher. I would bet every manager and employee asks himself the question every now and then. How do I know this research project is being useful to the company? How do we quantify the outputs of the millions euros invested? How do we identify performant projects and researchers?

Maybe a study of the organisation styles of private research organisations would confirm that there is hardly a right structure. If all possible systems are put on a circle, I think management ends up switching from one to the other. Research organisation structures can be defined in terms of:

• Matrix vs group-based structure: either pools of competences that are shared vertically and horizontally during the lifetime of projects, versus multi-disciplinary groups of people working together on projects
• Technology or user-driven: develop technology first then find user problems versus the opposite
• Short-term vs long-term research: the research supports current businesses versus focuses on 10 years ahead topics
• All internal vs internal+external: research and development carried out by research projects, versus for example research done internally (e.g. vision, theory) and development externally (e.g. outsourcing of product creation)
• Bottom-up vs top-down strategy: top management decides where to go versus groups and researchers do
• Democratic vs dictatorial decision making process: everybody has an opinion about everything versus decisions are made by one
• Quantitative vs qualitative measure criteria: performance is defined by numbers of xxx done versus e.g. peer review
• Centralized/hierarchical vs distributed channels of control: the top evaluates performance and quality versus peer review
• PhD vs engineer researchers: visionary people and high-level thinking versus hands-on people
• Experts vs entrepreneur-researchers: researchers are asked to focus on their specific domains versus asked to understand the whole chain technology-user-business.
• Focus on academic results vs focus on patent portfolio: work top ranked in the international research communities versus work protecting future businesses
• Academic vs industry style: research group organised around a senior professor with graduate and undergraduate students versus groups formed of peers
• Cheap vs world-class researchers: more and cheap versus few and famous people
• Managers vs leaders: focus on administrative tasks versus focus on where to go
• Out of vs use of public funding: in house only work or one-one cooperation versus national, european or international collaboration
• Defensive vs aggressive Intellectual Property Right (IPR) strategy: use patents to defend the company products and business, and negotiate patent portfolio with other companies versus use patents to prevent others from doing business, collect royalties or go to trials

The list could certainly grow up more. But focusing on the structural details of the whole organisation does not compensate for the lack of good vision.
I had read once that having 1% of successful projects should justify the investment in research. This question will keep being asked as long as it’s not 200%. The Harvard Review journal and other technology innovation consultanting firms have still good years to come. It’s easier to say how it should be done than doing it. But who could be surprised by that?

My notes from the workshop are on the wiki.

It’s now very likely that I will fly to Boston and participate to a workshop in Portland, Maine. I will leave Wednesday morning, and come back Sunday – provided that the administrative process goes well.

I have not gone to the US since August 2003. And I miss it a lot. But this is a work travel, not much time to visit. Maybe NY on Saturday, or should I stay in Boston, which I know already a bit after staying there 2,5 months around Havard square?