The relation between humans, computers and electronic devices has evolved rapidly defining technology eras with shifts in the related IT business leadership. From the "one to many" relation (computer vs. human users) in the enterprise domain during the 60s they moved to the family environment with the launching of the personal computer in the 80s. With the mobile phone they established a personal "one-to-one" relation ten years later.
We are currently moving to a "one to many relationship" with phone complements, Wifi routers, gaming consoles, mp3 players, STBs, TVs and infrastructures with impressive computing and storage capabilities. A new concept of global and transversal platform is emerging for exploiting the real potential of the network and affecting all business areas.
OSAMI-Commons targets open source common foundations for a dynamic service-oriented platform which is able to personalise itself in large diversity of co-operating Software Intensive Systems (SISs). The platform will make easy what is difficult now, it facilitates seamlessly:
- Service retrieval from external centralised or distributed repositories
- Devices to connect and exchange information and services
- The connection between various vertical markets in order to allow for new business solutions
OSAMI-Commons consortium shares a common vision of an Internet infrastructure emerging from a community process. The evolution of the logical infrastructure (i.e software) is part of a higher level roadmap that also includes the content. The emerging infrastructure is progressively embedding the users in a virtual world in which, at long term, every one will contribute as individual or community (e.g. social network) playing different roles; proving physical containers (i.e. devices), logic (i.e. software), content (i.e. music, photos, videos) or just as user providing information of needs, life habits, behaviour and preferences as digital content consumer, Ami digital liver at home, at work, in the car or in the public space, defining with this the directions of the platform to evolve.
Important challenges derive from this platform to become open (at least to some extent) as it will provide the perception of the physical world in a network context allowing massive collaboration of computing systems. Nevertheless, an open service platform is not a direct consequence of an open source base. Open software and content can help to create a concentration of impressive amount of knowledge (derived from massive automated processing of links among information and user interactions) which, in many cases, could go against the spirit of those contributing and block a fair business playground.
In the convergence process driven by Ambient intelligence (AmI) can be defined as an automated service provider embedded in devices. The Ambient Intelligence platform will be able to personalise itself dynamically in devices according to the context (i.e. physical container, user needs and environment).
"People who are really serious about software should make their own hardware" -Alan Kay-.
Was one interesting sentence in the slides of Steve Jobs (Apple CEO) kick-off Macworld Conference & Expo 2007 during the outstanding iPhone presentation. This is not new as we have seen relevant movements in this direction in the past from big software industries.
In the AmI roadmap, the platform for development and target system are getting closer. This is a necessary step and previous phase for reaching the critical mass before it becomes alive and finally intelligent.
The health of the market segments relying on embedded systems depends on its presence in the high level roadmap. The success is strongly linked to its capability to collaborate for defining its identity in the global platform contributing to common foundations. This is a significant challenge when the experience from the past shows that trends in non-embedded systems are introduced in the embedded world 5-7 years later (which is a significant delay when things are evolving fast).
In economic terms, according to Research and Markets report, SOA Market will reach US$ 18.4 billion by 2012, although this figure probably does not foresee the indirect impact through the different segment markets of a paradigm shift derived from the combination of Open Source and Service Oriented Architectures (SOA) allowing run-time extensible systems or building systems on demand.
In addition to the transversal market relevance the project is highly for the vertical domains. Europe is facing significant demographic changes often subsumed under the term "ageing society". This requires cost-efficient ways to support the supply of informal care and expand formal health and care provisions as well as progress in aids and assistive technologies. As a consequence of this, there is a strong need for interoperability, interfaces and standa rds to allow for a seamless information exchange and cooperation amongst the different care takers from the general practitioner, care institution and clinic up to the patient. This will allow extending the health infrastructure from the hospitals to the home or even to a personal health care on the move. The OSGi technology provides a unique building stone for such an infrastructure. No other software technology provides so many of the crucial aspects that are needed to implement large scale personal healthcare. The portability, security, reliability, integrity, and manageability are all prerequisites for a health care infrastructure. Similar benefits could also be identified for other OSAMI domains ( mobile, home, public services )
Technical and strategic relevance
Relevant progress have been achieved during previous projects with relatively limited resources, leading the industrial support for open source frameworks targeting embedded Java devices and starting relevant repositories. Following, similar approaches have been adopted by IBM. Similar architectural solutions between desktop developments and equivalents in Eclipse have also shown a commonality in the vision. The movement of the OSGI framework involved in ITEA Osmose project to Apache community has initiated a discussion process among other platforms such as Apache directory, Harmony and Cocoon together with links with Maven project. In the mean time the refactoring of Eclipse for OSGi compliance has been fast due to a strong IBM leadership and a relevant set of assets are currently available confirming the previous project vision of the strength of this platform once reinforced its architectural openness though open source implementations.
This fact proposes new challenges due to fragmentation in related but independent R&D projects. To overcome this fragmentation a "syndication mechanism" was explored between Cosi and Osiris projects in the so called Cosiris cluster. The syndication provides a framework for a joint virtual management infrastructure keeping independent project identity. Following the positive experience a survey of related projects is being carried out for exploring collaboration opportunities. Augmented information flow through a closer cooperation across projects increases the generated knowledge that is the strength of the new paradigm versus the property in the conventional approach.
"If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas." -George Bernard-
The competitive advantage of the current Internet supported software development scenario in which high skilled software people can work together independently of the geographical location for a flexible collaboration is recognized by venture capital firms (several scope overlaps took place during previous/running ITEA projects). The opportunity of sharing development effort in self-organizing merit based organization is difficult to beat with conventional models.
Many organisations are currently establishing strong links with these communities for recruiting or sponsoring personnel, influencing the global direction, leading technology or simply to avoid the risk of being left out or involved too late.
The incubator role of recognized communities is helping in the re-factoring of the whole software industry while Open Source is emerging as a new global paradigm challenging the conventional approach for software development and related business models.
Many industries consider open source as a commodity. Something that can be taken for free and integrated in the products. Nevertheless, when developed in the context of well driven communities "real business value is created" due to the strength of the new paradigm for improving developers knowledge and skills learning form each other and driving the technology as a consequence. In this context, Open Source code is just written ideas improved through the interaction and with lower development cost due to the sharing of effort. Due to the strengths of the new model, future software related business will be strongly based on open source cooperation.
"The purpose of words is to convey ideas. When the ideas are grasped, the words are forgotten. Where can I find a man who has forgotten words? He is the one I would like to talk to" - Chuang Tzu-
Starting 2007, C/C++ is still the preferred language for real-time systems. Surveys indicate that the general feeling of Java as big and slow for real-time persist. Nevertheless, significant progress seems achieved to overcome these aspects. Standards aligned with various implementations are there (JSR-000001: Real-time Specification for Java, JSR 302: Safety Critical Java Technology) and several products seem close to support them. This fact combined with the availability of a relevant base of developers proposes Java as a promising candidate for embedded systems development. The recent selection of Java by Lockheed Martin for several critical subsystems for the Aegis Weapon System reinforces the above. Decisive factors in this decision has been productivity and portability strengths combined with the performance requirements. In the meantime, Java has become the dominant language in the university with 4 million developers outside and one billion worldwide shipments of embedded virtual machines (annual growth rate of 24.5%).
The trend in lower level layers is helping higher level standards in gaining momentum. The OSGi Service Platform is the optimal Java-based application server for networked devices, however small or large they are. This non-proprietary service platform spans in a broad range of domains: Digital mobile phones, Vehicles, Telematics, Embedded appliances, Residential gateways, Industrial computers, Desktop PCs and High-end servers, including mainframes.
One of the surprising aspects of the OSGi specifications is that they were targeted at small, embedded devices but have become popular in high-end servers, up to mainframes.
However, the OSGi specifications do not address many of the issues that were addressed in J2EE like databases, transactions, messaging, etc. -- APIs that are crucial for most enterprise applications. Interestingly, in this area J2EE was competing with Spring. Both Spring and OSGi technology are very orthogonal; there is little overlap. Interface21, the company behind the Spring Framework and the Spring family of products, was named a "Cool Vendor in Integration and Platforms, 2007" by Gartner. As Rod Johnson (Interface21) says: "OSGi technology will strengthen Spring's value as a basis for server infrastructure and offer benefits to users in the area of componentization, versioning and dynamic deployment."
In the appendix can be found more detailed information concerning OSGi State-of-the Art and its adoption in mobile devices, home products such as the Siemens serve@home, Eclipse development environment and BMW 5 series. The OSGi teaming up with EclipseCon 2007 is a step ahead from the enterprise world that reinforces the OSAMI vision concerning platform and tools convergence (as the tools are personalisation of the platform for a developer user).
Technological innovation and achievements
OSAMI-Commons targets open source common foundations for a reference platform building-up from the embedded system domain. Based on dynamic service-oriented programming principles it will be able to personalise itself in large diversity of co-operating Software Intensive Systems (SISs). This is achieved through technologies allowing virtualisation of SISs with a high level of abstraction (e.g OSGi and Web Services).
From the above it seems clear that we need to change the way we do software related business as we cannot expect many Ambient Intelligence platforms outside. As a consequence this will entail changes in the R&D approach too.
The future networked scenario proposes relevant challenges for embedded software industry due to the merging of the devices and software, the leading role of server based technologies and the lack of a common reference platform due to the large diversity of physical containers (i.e. devices or hardware platforms in general) The health and business potential of the market segment depending on embedded software will be strongly linked with its presence and identity in defining the AmI roadmap.
The appendix includes summary information (outlined considering space constraints) about working groups in the OSGi Alliance which addresses topics within the core ITEA competences and OSAMI-Commons scope. Some of these groups have been launched recently. OSAMIcommons will contribute to a joint positioning reinforcing synergies across participating European actors. The project will also support a cooperation network of OSGi users groups in Europe.
The project will Demonstrate the potential of combining Product Line engineering (www.sei.cmu.edu/productlines/, www.esi.es/Families/), Service Oriented Architectures (www.oasis-open.org, www.osgi.org) and Open Source (www.opensource.org) to reduce development costs and increase product qualities.
OSAMI-Commons will demonstrate derived benefits in the services domain and will extend service oriented solutions in vertical domains, such as healthcare.. The common architectural approach will support not only the reuse of assets across different domains but also (in run time) a cooperation framework of embedded devices for providing added value services. Deployed applications will allow seamless access to services across devices in the context of interaction domains (e.g., personal, family, social network, equipment & service provider, industrial systems...). This will be achieved through an open source security middleware for defining boundaries of distributed virtual systems. Some technical areas strongly related are; open source & product line (OSPL) engineering, (open source assets selection/maintenance/evolution, product/services derivation; development platforms, product, services), OSGi/Web services combined architectures, OSGi/Web services related repositories/directories, architecture re-factoring towards OSAMI-commons architecture, dynamic SOA, security and intelligent browsers & interfaces.
Due to the fact that OSAMi-Commons impacts in the approach to the business it will entail innovation in different dimensions. The project will help in shaping a common vision of the approach to the business, organisation and process, building on results from previous ITEA projects (e.g. ITEA-COSI) but bringing the architecture dimension as a core asset to orchestrate the others (bringing together results from OSIRIS, SIRENA... and may be other ITEA projects too, ref. appendix on leverage possibilities). A common architectural approach based on a combination of SOA principles within the same physical container (i.e. OSGI) and interoperability (e.g. SOAP, REST) will provide essential inputs for the business, process and organisation related activities.
The open source community has achieved a significant progress in middleware, this is why the approach in OSAMI-commons will be very much from the domains towards driving the technology. This is possible once a common agreement in high level architectural principles has been achieved during the proposal preparation phase.
Major innovation from OSAMI-commons will derive from the effort of combining domain driven architectural approach with a transversal platform. This will have a significant impact on business, process and organisation work too. All together will help in a global positioning with respect the Ambient Intelligence Platform.
The project will contribute to open an innovation framework for increasing the global competence of markets relying on embedded systems through optimizing transversal synergies. There are no known efforts with a similar objective as OSAMI-commons (i.e. building-up from embedded systems core technologies taking advantage of transversal synergies and linking benefits in software development with services) Under a software engineering perspective OSAMi-commons proposes a disruptive paradigm shift from state-of-the-art software Product Line development towards some sort of run-time Service Lines around a transversal and evolving platform.