September 08, 2003

The topic of my PhD thesis: Individual Service Provisioning

I successfully defended my PhD thesis on February 7th, 2003. This text describes its topic. The full thesis is available online under the link Publications and Documents.

The Candidates work applies to the broad area of human-computer interaction as well as to the areas of software engineering and service provisioning. The results are applicable as a fairly non-debatable description of a theoretical base line for the work ahead within ubiquitous computing.

Electronic services are the tools for interaction between humans and computer systems in a new model of human-computer interaction. In this new model services are omnipresent and the devices with which to access them vary from stationary PCs, to mobile phones, to embedded computers. The service providers also change. Traditional providers are slow moving and large software houses, whose large monolithic software applications have been installed on countless PCs. In the new model there is a need for more flexible and dynamic providers who can adapt to shifting currents and who are able to profit in smaller niche markets which open up in the future.

The mismatch between the supply and demand of software in this new world is already highly evident in the failure of the large corporations to profit during the first steps into this new market place; mobile services distributed over telco channels have hitherto been a great disappointment.

The work of the Candidate pertains to this domain of a future world of electronic services and can be summarized in three points: first, the Candidate has designed and implemented a software platform for humans to store, execute, and interact with electronic services, second, the Candidate has designed and implemented a sub-system to the main platform, which allows users to create their own services from smaller building blocks, and third, the Candidate has designed and implemented a sub-system to the main platform, which allows user-created services to be provisioned among other users. The resulting system allows individual users to create for themselves fitting electronic services which may be used in the personal platform but which are also published and provisioned for the benefit of other users. This concept fundamentally changes our perception of who is the service provider and who is the service consumer. It also allows for unlimited flexibility and adaptability for a changing service market space.

In the field of Ubiquitous Computing the future computing environment is described both in terms of underlying infrastructure and requirements on hardware and networking technologies but also in terms of the users' experiences. In this vision, all computing relevant to mainstream users will be permeated in the fabric of every day life - its situations and its artifacts. The current state of the art describes, among other things, wireless networking technologies, context gathering infrastructure, miniature hardware solutions, etc. What is missing is the underlying and encompassing software layer which will tie together all appliances and access devices and services. The Candidate's work describes one possible such software infrastructure which although it may not be the final answer, goes a long way to bridge the gap between the personal computer interaction model and the ubiquitous computing interaction model.

Posted by fredespi at 05:47 PM

September 05, 2003

The STES project final report

The Sharing and Trusting Electronic Services (STES) project ended this spring and the final report is due. The following text (in Swedish) has been submitted to Vinnova as the final report. First comes the short description of the project (in English). If anyone is interested in an english version of the final report, please contact me.

Computer usage is moving towards a mobile, distributed, and service centric model. Many computer systems including the World Wide Web, broadband networks, and digital cable and TV systems are evolving into provisioning platforms for interactive electronic services. In a previous project, sView, we explored, designed, and implemented solutions for a personal service briefcase in which a user can keep her personal electronic services. The purpose of the future STES project is twofold: first STES will enable users to create and share their own services and thereby act as individual service providers. Second, STES will explore and design for trust mechanisms that will enable users to trust these individually created services.

The current project is a pre-study. We intend to tie to the project several external commercial and non-commercialpartners to take part in the main project.

Slutrapport för Sharing and Trusting Electronic Services (STES):

Fredrik Espinoza har slutför sitt avhandlingsarbete, disputerat, och tagit ut sin doktorsexamen. Detta moment är viktigt för STES eftersom projektet delvis baseras på resultaten som beskrivs i avhandlingen. Titeln på avhandlingen är "Individual Service Provisioning" och den handlar om den teknik som behövs för att göra det möjligt för individer att skapa och dela med sig av elektroniska tjänster. Detta moment är viktigt för STES eftersom projektet delvis baseras på resultaten som beskrivs i avhandlingen: delningssystemet Briefcase Connectivity som kopplar ihop alla användare i ett peer-to-peer nätverk och delningstjänsten SharedServicesLoader som möjliggör enkel utdelning av tjänster. Briefcase Connectivity har under projekttiden genomgått en andra utvecklingscykel och vi har kunnat demonstrera enkel utdelning av tjänster mellan användare av tjänsteplattformen sView. För mer detaljer se avhandlingen.

Projektets första delresultat är att knyta till projektet ett antal samarbetspartners som kan ingå i den senare ansökan om mer finansiering. Vi har framgångsrikt knutit till oss två partners, Telia Research (Bruce Horwitz) och Appear Networks (Xavier Aubry). Båda företag är positiva till att ingå i det planerade projektet och har varit delaktiga i att skriva den nya projektansökan för det nya projektet Enabling and Promoting Trust in Micro Service Environments (EPTMSE).

Projektets andra delresultat är den framgångsrika ansökningen i 2003 års utlysning för Nätverksbaserad Programvaruteknik. Vår nya ansökan för projektet EPTMSE beviljades och därmed kan STES projektet ses som framgångsrikt.

Projektbudgeten för STES har använts till två saker: dels den andra implementationen av Briefcase Connectivity och den första implementationen av SharedServicesLoader som tillsammans med sView behövdes för att kunna demonstrera att skapande och delning av tjänster överhuvudtaget är möjligt, och för det andra till knytandet av kontakter och deras introduktion till projektet samt till framtagningen av den nya projektansökan.

Projektets 4 publikationer:

Fredrik Espinoza and Lucas Hinz. Generic Peer-to-Peer Support for a Personal Service Platform. In Sumi Helal, Yuji Oie, Carl Chang, and Jun Murai, editors, 2003 Symposium on Applications and the Internet (Saint 2003), pages 218--225, Los Alamitos, California, January 2003. IEEE Computer Society.

Fredrik Espinoza. Towards Individual Service Provisioning. In W. Lewis Johnsson, Elisabeth André, and John Domingue, editors, In 2003 International Conference on Intelligent User Interfaces (IUI 2003), pages 239--241. ACM Press, January 2003.

Fredrik Espinoza and Ola Hamfors. ServiceDesigner: A Tool to Help End-Users Become  Individual Service Providers. In Proceedings of the Thirty-Sixth Annual Hawaii International Conference on System Sciences. IEEE, January 2003.

Fredrik Espinoza. Individual Service Provisioning, Ph.D. Thesis, Dept. of Computer and Systems Sciences, Stockholm University and Royal Institute of Technology, Kista, Sweden, 2003.

Posted by fredespi at 12:30 PM

September 02, 2003

Fredrik Espinoza awarded 3MSEK Vinnova funding for new project EPTMSE

From Vinnova's 2003 call in the Network based software (Nätverksbaserad Programvaruteknik) program I was in May awarded a 3 year research grant of 3 MSek. The project is called Enabling and Promoting Trust in Micro Service Environments (EPTMSE) and it is a cooperative effort with Teliasonera and Appear Networks. The project started Juli 1st 2003. This is the short description of the project (in English and Swedish):

The purpose of the project "Enabling and Promoting Trust in Micro Service Environments" (EPTMSE) is to study, design, and implement methods and technology for ensuring a trustworthy experience between users and services in micro service environments. Micro services are services that are created and distributed by small to medium sized producers (sometimes even individual users can provide services) and which aim to complement a user's personal service sphere. The micro services environment forms the basic foundation of the users future service cloud, in which the electronic service is the basic unit and the individual user's needs are catered for by a plethora of such services instead of a few more traditional and monolithic applications. With a great number of possible producers and distribution channels for services in this scenario (i.e. wireless position based distribution) the ability for service consumers to employ the usual methods to ascertain a perceived level of trust (for example brand name recognition) decreases. Therefore, in micro service environments, the need for explicit systems that enable and promote trust increases. The EPTMSE project will build these trust systems and the project is a collaborative project between SICS, Appear Networks, and TeliaSonera.

and a short description in Swedish:

Projektet Enabling and Promoting Trust within Micro Service Environments (EPTMSE) syftar till att studera, utforma, och implementera metoder och teknik för att skapa användarupplevd tillit i system av mikrotjänster. Med mikrotjänster avses tjänster som skapas och tillhandahålls av medelstora eller små producenter och leverantörer - i många fall så små som individuella användare. Dessa mikrotjänstemiljöer utgör grundstommen i ett framtida användarcentrerat tjänstemoln, där den elektronisk tjänsen är grundenheten och användarens behov uppfylls av en mängd sådana tjänster i stället för ett fåtal traditionella och monolita applikationer. Med många fler tjänsteproducenter och annorlunda distributionskanaler (t ex trådlösa positionsberoende hotspots) försämras möjligheterna för tjänstekonsumenterna att lita till vanliga metoder, såsom varumärkeskännedom, för att skapa sig en uppfattning om tillgängliga tjänster.I mikrotjänstemiljöer finns alltså ett större behov av explicita tillitsystem än vad det gör i dagens system och projektet EPTMSE ämnar bygga dessa tillitsystem. Projektet är ett samarbete mellan SICS, Appear Networks, och TeliaSonera.

Posted by fredespi at 06:23 PM

The ThinkWearable Course as a Prototype Bridge Between Academia and Industry

One of my final doctoral courses was the thinkWearable; course at Stockholm University. Please see ThinkWearable for more details. The interesting thing about this course was the format and its coupling to the academic research world. I wrote the following commentary about this coupling:

The thinkWearable project served two purposes: first, it was a graduate course for graduate students in the Stockholm area and mainly the Royal Institute of Technology and Stockholm University. Second, it was a prototype for a bridge between the academic research world and industry. ThinkWearable as a course, is described elsewhere in this book. This text will briefly describe thinkWearable as a possible link between research and industry.

Approximately half of the course participants of the thinkWearable course were at the time of the course employees at the Swedish Institute of Computer Science, a Swedish national research institute with strong (ownership) ties to major Swedish industry parties such as Ericsson and Telia. The purpose of SICS is to leverage the computer science research done in the institute to benefit Swedish industry in general and the owners in particular. Usually this happens through dissemination of results in academic papers, publicly open seminars, joint research projects, staff migration from SICS to other companies, and through spin-off companies building on results created in SICS. SICS has a very good record of performance in terms of disseminating results and creating industry value in all of the aforementioned ways. Despite this, however, many of the research results that may have strong potential as products or services in the national and international market place, never reach that full potential. The reasons for this vary, but contributing factors are: failure to communicate the value and purpose of the results, failure to appreciate the value and purpose of the results, failure to methodically evaluate and nurture results, and failure to encourage researchers to exploit the results. These characteristics are not particular to SICS, they are general to many research institutions that aspire to do applied research for the benefit of their sponsoring environments. In the light of this, the setting, process, and result of the thinkWearable course become interesting.

The thinkWearable course brought together a set of talented and ambitious individuals (the students), a well connected and visionary leader (Prof. Zary Segall), a group that contributed various competences (the sponsoring organizations), and a structured method to create from this mix the seeds for bringing research results to market. The process began with brainstorming sessions concerned with singling out the best alternatives on which to base the further work. Next, the two main project ideas were fleshed out and put in a theoretic and practical perspective: did the ideas have the novelty necessary for research and did they appear feasible in a reasonable time frame? The criteria for the course were to be able to build a working prototype within the project period but to still allow for future improvements and feature enhancements as technology evolves. The ideas then underwent several phases of development, both as ideas, as implementations in hardware and software, and as designs. Throughout this process each idea was refined and the project participants became very well acquainted with each idea's possibilities and limits. The result of the iterations were two sets of working prototype systems implemented in software and hardware with a very elegant and suggestive graphical and industrial design. The project group, being staffed by a graduate student from the Stockholm School of Economics (Handelshögskolan), was also able to produce credible business models for both prototypes. At each step of the way the sponsors contributed with their various competences and with funding and equipment but mainly with their behind the scenes commitment to the project. At the end of the project the results were at a qualitative level at which the sponsoring companies were willing to commit more funding and the Museum of Science and Technology in Stockholm agreed to an exhibition.

What are the conclusions that we wish to draw from this? There are four: 1) It is likely that research results can be brought to the commercial arena by purposeful effort and the execution of a structured method. 2) It is likely that many of the existing results in organizations such as SICS could be brought to market through such a process. 3) It is likely that an organization whose purpose is to develop business ideas in much the same way as was done in the thinkWearable course could benefit by getting close to where the action is, i.e. by cooperating directly with the organizations that produce the research results. 4) The participating individuals are a strong component of success in a project to develop research results into commercially viable products. Therefore it is likely to be beneficial to such a project if the originating researchers are encouraged to participate.

Posted by fredespi at 05:13 PM