Abdallah Namoun

Service Composition for Non-programmers: Prospects, Problems, and Design Recommendations

Although Service Oriented Architecture technologies are becoming widespread, user research to support the exploitation and composition of services by non-programmers is still in its infancy. This paper presents three user studies conducted during the user-centered design of the Serv Face Builder, a service composition tool for non-programmers. Results suggest that end users do not realize that services can be connected together and do not easily understand that information can flow between services. Further conceptual and usability problems of service composition for end users and some guidelines to overcome them are discussed.

Assisted Service Composition for End Users

Involving people who do not have programming background in assembling and tailoring service-based applications promises to open up access to the creativity of millions of users. An increasing number of development environments aim to do this by offering drag-and-drop visual representations connecting different service components into an assembly. In contrast to the majority of these, we did not start with the technology but with the users - service producers and consumers, and studied the core issues which should be resolved before people who are not programmers can start to assemble services into meaningful applications, over and above the presentation-level integration offered by current mash-up environments. The result is an assisted approach to service composition for end users, which uses semantic technologies to shield users from the irrelevant complexity of service technology and from the need to manually resolve dependencies between services. The approach was evaluated by a focus group of non-technical users, who ranked it highly and provided valuable suggestions for further improvements and supporting features.

Toward process mashups: key ingredients and open research challenges

Over the last few years, the mashup community has grown significantly, and mashup development has matured substantially compared to the initial hacking practices. Mash-ups as applications have specialized into data mashups, service mashups, or user interface mashups -- although these terms lack a common agreement on definitions -- while other types of mashups can still be identified. In fact, recently the term process mashup emerged, yet, again, its meaning is everything but clear. Intrigued by this latter idea, in this paper we try to understand what process mashups are. We identify three dimensions that distinguish process mashups from most of the current types of mashups and we show that exploring them leads to a set of new types of mashups, which are the actual basis for the development of process mashups. For each of these new types of mashups, we provide a discussion, discuss suitable application scenarios and show tool support, so as to highlight challenges and open issues.

End-user development of service-based interactive web applications at the presentation layer

Lightweight service composition approaches are gaining a fast momentum in the integration landscape, among which is the integration/composition at the presentation layer where software components are integrated using their frontends, rather than application logic or data. This paper presents a new approach for composing web services through their user interfaces (UI) to form composite web applications in a purely graphical manner without the necessity to write any programming code. Unlike existing approaches, our service composition approach is shaped by a set of iterative user based evaluations to ensure no modeling or programming skills are required for web application development. Indeed our approach is tailored towards non-programmers. This paper provides an in-depth description of the general concepts and fundamental principles of our UI-centric design time approach, a brief description of our prototype, namely the ServFace Builder which serves as a proof of concept, and evaluation results.

Assisted Composition of Services on Mobile Devices

Composing software services on mobile devices is especially challenging when attempted by non-programmers. In this paper we compare two alternative supporting strategies: using generic task templates and scripting together condition-response fragments. The first is exemplified by a prototype called ACOM (Assisted Composition on Mobiles), the second by a commercially available alternative called IFTTT (IF This Then That). The paper uses a comparative observational study to highlight the benefits and drawbacks of both approaches, and to derive lessons for their improvement.

Designing and evaluating a system of document recognition to support interoperability among collaborative enterprises

Abstract A common understanding of business documents is important for realizing interoperability among collaborating enterprises. In this paper we report on the design and evaluation of a system that can help collaborating enterprises to efficiently recognise and align business documents (e.g. incoming emails and attachments) according to the enterprises’ local document repository. The system relies on the interplay between automatic recognition of business documents and human intervention by means of an assistive mapping tool. Our findings show that the balance between automatic recognition and human intervention ensures greater levels of accuracy in the system. We evaluate our system from both usability and accuracy perspectives and argue that some of the lessons learned and design decisions we took are applicable to the general mixed initiative tools for end user support.

Empowering end-users to develop service-based applications

The 2nd International Workshop on End User Development for Services (EUD4Services) focuses on the issues encountered when people who are not educated as software developers attempt to create and compose software services, and on approaches and theories aiming to support such activities. The aim is to establish a community of academics and practitioners and facilitate the production of a coherent body of work related to this area.

A Comparative Study: Service-Based Application Development by Ordinary End Users and IT Professionals

Service-Oriented Architecture enables users, both ordinary end users and IT professionals, to be part of the development cycle of interactive service-based systems in order to fulfil their desired needs. In this paper we explore and compare the mental model of two different categories of users towards the idea of “service composition by end users”. Participants’ responses are concluded from 5 separate focus groups, including a total of 64 participants. Results have shown that both groups of users are highly interested in the composition of service-based systems; however, privacy and security concerns and technical complexity of current approaches and service composition environments hinder the diffusion of service-based technologies among users. In this respect, this paper proposes a preliminary model of service composition uptake by end users and discusses user views and requirements to facilitate service composition.

User-centered design of a visual data mapping tool

Understanding the meaning and the type of a business document received by a company is important in order to determine an appropriate response. We have developed a visual tool allowing ordinary users to express mappings between arriving documents and their elements on one side and the different document types on the other. The tool is used to set up and continuously update an automatic semantic analysis mechanism which determines the document type from a set of information items contained in the document, thus allowing automatic processing associated with the types to be applied to the arriving document instances. The activities performed by end users within the visual data mapping tool are quite complex and require user-centric design to ensure tool is useful and usable. In this paper we describe the user-centric process informing the design of the tool. We conducted two workshops to discover the mental models of our users regarding the information content in their documents and their requirements towards the design of the tool. Subsequently, these findings were used to design and implement the visual data mapping tool. The resulting system was evaluated by target end users who proficiently demonstrated the usability of the developed concepts and features.

Tool support for realising a common understanding of business documents in collaborating enterprises

A common understanding of business documents is important for realizing interoperability among collaborating enterprises. In this paper we explore the mental model of administrative staff in business enterprises to unravel how they map information items from business documents to the elements of an ontology of core business concepts. The data for our study has been gathered by organizing three focus groups. The results provide an insight into how users map and classify business documents. The findings of the user studies are fed into the design and development of a visual data mapping tool that enables users to create mapping between sets of information items and their own document taxonomy, thus classifying the incoming documents according to their local practices and needs.