Smart-M3 training

Overview

The training session gives an overview of the Smart Space concept using the Smart-M3 platform as an open-source implementation (http://sourceforge.net/projects/smart-m3/) of this concept. We study different available APIs for the Smart-M3 platform and simple programming models for Smart-M3 applications.

Smart-M3 is an interoperability platform that provides an infrastructure to share information expressed in RDF. The platform consists of a Semantic Information Broker (SIB) that stores the information and Knowledge Processors (KP) that can insert, remove, query, and subscribe to information. The Smart Space Access Protocol (SSAP) is the protocol that a KP uses to access its SIB. The information is stored in SIB as an RDF triple store according to standardized or otherwise agreed ontology. Information sharing helps the participating KPs to gather relevant context information, leading to more efficient operation and innovative multi-device use cases. For more information about Smart-M3 please check http://en.wikipedia.org/wiki/Smart-M3 and http://fruct.org/m3.

The developers of the application logic of KPs use a knowledge processor interface (KPI) to access information in smart space. There are two classes of KPI, depending on the code abstraction level.

1. Low-level KP programming in terms of RDF triples.
2. High-level KP programming in terms OWL ontology.

Low-level access requires the user code to operate with RDF triples, directly following the SSAP operations. In contrast, high-level KP development is based on an ontology library, and the code deals with ontology classes, properties, and instances.

The focus is on step-by-step developing a simple “Hello World” application consisting of several KPs. For KP programming we show two KPIs: KPI_low (low-level, ANSI C, low-performance devices) and SmartSlog ontology library (high-level, ANSI C and C#, low-performance devices). For simple ontology manipulation we use Protégé.

The training agenda

1. Overview of the Smart Space concept, Smart-M3 platform, Smart-M3 programming models, and available KPIs.
2. Smart-M3 KPI: different code abstraction levels. Case studies: KPI_low and SmartSlog.
3. Hello world application using KPI_low
4. Hello world application using SmartSlog. Web Ontology Language and the Protégé tool.
5. Cross-platform applications: demos for Linux, Maemo5, and MeeGo.

Prerequisites

The number of PC-based work places is limited. We recommend you to take your own laptop. It must have the following software installed in advance.

• Smart-M3 platform (http://sourceforge.net/projects/smart-m3/)
• KPI_low (http://sourceforge.net/projects/kpilow/)
• SmartSlog (http://sourceforge.net/projects/smartslog/)
• Protege (http://protege.stanford.edu/)

Another option is own laptop with installed Oracle VirtualBox (http://www.virtualbox.org/). In this case you will be supported with the image with installed Linux, Smart-M3, and other tools.

Recommended skills for the audience

• Basic skills of C
• Basic skills in using GNU/Linux platform as development environment (make, gcc)

Registration to training

All registered participants have to fill the attached form and send it from email used in registration to info<at>fruct.org within 3 days after online registration. If we don't receive your registration form, after 3 days your registration will be automatically cancelled.