The SMIL 3.0 language is the W3C's standard language for web-based multimedia. This book provides a comprehensive guide to all of SMIL's elements and attributes for all versions, from the complete SMIL Language profile to SMIL Tiny. Driven by dozens of examples and illustrated code fragments, every aspect of SMIL is treated in this book.
Starting with essential background on XML and the streaming media infrastructure, this book considers recent features such as smilText and smilState, as well as the structuring, layout and timing aspects required to master SMIL. In addition, the book describes advanced SMIL functionality, including transitions, animation and embedded metadata, all from an application-oriented perspective.

The SMIL 2.0 Language is the Web's solution for multimedia. This book provides a comprehensive guide to all of SMIL 2.0's elements and attributes for both desktop and mobile multimedia. Driven by dozens of examples and illustrated code fragments, every aspect of SMIL is treated.
This book considers the document structure, media integration, presentation layout and detailed timing constructs available in a wide range of SMIL compatible players and browsers.The book also considers advanced SMIL topics of linking, content control and SMIL animation, all from a practical, implementation-oriented perspective.

The book is the edited proceedings of the 2001 Multimedia Modelling conference, held at CWI in Amsterdam.

 

 

Readings in Multimedia Computing and Networking captures the broad areas of research and developments in this burgeoning field, distills the key findings, and makes them accessible to professionals, researchers, and students alike. For the first time, the most influential and innovative papers on these topics are presented in a cohesive form, giving shape to the diverse area of multimedia computing. The seminal moments are recorded by a dozen visionaries in the field and each contributing editor provides a context for their area of research by way of a thoughtful, focused chapter introduction.

In this presentation we present how low-level metadata extraction tools have been applied in the context of a pan-European project called Together Anywhere, Together Anytime (TA2). The TA2 project studies new forms of computer-mediated social communications between spatially and temporally distant people. In particular, we concentrate on automatic video analysis tools in an asynchronous community-based video sharing environment called MyVideos, in which users can experience and share personalized music concert videos within their social group

In this paper we introduce the design and implementation of seamless playback for video/audio in the Ambulant Player. The Ambulant Open SMIL Player is an open-source media player that supports SMIL 3.0. A typical SMIL multimedia presentation consists of a set of declarative references to video/audio clips, which are relative to each other in terms of temporal and spatial relationships. Unfortunately, the declarative nature of SMIL often imposes performance delays, as individual items are fetched and presented. In this paper, we discuss the design and implementation of a caching and prefetching scheme that avoids service interruption and eliminate switch delay among these clips. A collection of videos is thereby rendered as if they were continuously rendered from one media container on one media source. Experiments are carried out to validate that our techniques can significantly lower the start delay of media rendering and therefore realize the seamless playback of SMIL multimedia presentations.

Inclusion of content with temporal behavior in a structured document leads to such a document gaining temporal semantics. If we then allow changes to the document during its presentation, this brings with it a number of fundamental issues that are related to those temporal semantics. In this paper we study modifications of active multimedia documents and the implications of those modifications for temporal consistency. Such modifications are becoming increasingly important as multimedia documents move from being primarily a standalone presentation format to being a building block in a larger application. We present a categorization of modification operations, where each category has distinct consistency and implementation implications for the temporal semantics. We validate the model by applying it to the SMIL language, categorizing all possible editing operations. Finally, we apply the model to the design of a teleconferencing application, where multimedia composition is only a small component of the whole application, and needs to be reactive to the rest of the system. The primary contribution of this paper is the development of a temporal editing model and a general analysis which we feel can help application designers to structure their applications such that the temporal impact of document modification can be minimized.

This paper introduces a multimedia document model that can structure community comments about media. In particular, we describe a set of temporal transformations for multimedia documents that allow end-users to create and share personalized timed-text comments on third party videos. The benefit over current approaches lays in the usage of a rich captioning format that is not embedded into a specific video encoding format. Using as example a Web-based video annotation tool, this paper describes the possibility of merging video clips from different video providers into a logical unit to be captioned, and tailoring the annotations to specific friends or family members. In addition, the described transformations allow for selective viewing and navigation through temporal links, based on end-users' comments. We also report on a predictive timing model for synchronizing unstructured comments with specific events within a video(s). The contributions described in this paper bring significant implications to be considered in the analysis of rich media social networking sites and the design of next generation video annotation tools.

This paper discusses two upcoming challenges for model-based user interfaces: multimedia and the Web of Things. One the one hand, multimedia content and real-time media transmission have their own temporal model Ð state Ð that has to be seamless integrated into the user interface event model. On the other hand, thousands of interconnected objects that can act as input and output devices impose unforeseen challenges for the user interface mapping into the real world.

This paper discusses the architectural challenges of transitioning from services to experiences. In particular, it focuses on evolution from traditional IPTV services to more social scenarios, in which groups of people in different locations watch synchronized multimedia content together. In addition to the multimedia content, the shared experiences envisioned in this article provide a real-time communication channel between the participants. Based on an implemented architecture, this paper identifies a number of challenges and analyzes them. The most important challenges highlighted in this article include: shared experience modeling, universal session handling, synchronization, and quality of service. This article is the first stone paving the way for a truly interoperable ecosystem, which can offer cross-domain experiences to the users.

This paper introduces a community-based video sharing environment to support asynchronous communication among heterogeneous participants within a restricted social community. Unlike other community sharing efforts, our work is predicated on the desire to strengthen existing strong ties among group members, in which existing relationships can be nurtured. Using the example of a high school concert as a starting point, this paper discusses a sharing framework in which highly personalized music videos are constructed from a collection of independent parent-made recordings. The environment addresses a series of parent needs for producing tailored presentations with custom features, based on Ôsafe sharingÕ of common assets. We report on the user needs determined by a number of focus groups and on a web-based environment that can be used to manage the complex inter-personal relationships and time-variant social contexts with a community of diverse (but related) users.

The migration of media consumption to personal computers retains distributed social viewing, but only via nonsocial, strictly personal interfaces. This article presents an architecture, and implementation for media sharing that allows for enhanced social interactions among users. Using a mixed-device model, our work allows targeted, personalized enrichment of content. All recipients see common content, while differentiated content is delivered to individuals via their personal secondary screens. We describe the goals, architecture, and implementation of our system in this article. In order to validate our results, we also present results from two user studies involving disjoint sets of test participants.

 

The objective of this work is to define a document model extension that enables complex spatial and temporal interactions within multimedia documents. As an example we describe an authoring interface of a photo sharing system that can be used to capture stories in an open, declarative format. The document model extension defines visual transformations for synchronized navigation driven by dynamic associated content. Due to the open declarative format, the presentation content can be targeted to individuals, while maintaining the underlying data model. The impact of this work is reflected in its recent standardization in the W3C SMIL language. Multimedia players, as Ambulant and the RealPlayer, support the extension described in this paper.

Over the last years we have witnessed a rapid transformation on how people use digital media. Thanks to innovative interfaces, non-professional users are becoming active nodes in the content production chain by uploading, commenting, and sharing their media. As a result, people now use media for communication purposes, for sharing experiences, and for staying in touch. This paper introduces a user-centric authoring tool that enables common users to transform a static photo into a temporal presentation, or story, which can be shared with close friends and relatives. The most relevant characteristics of our approach is the use of a format-independent data model that can be easily imported and exported, the possibility of creating different storylines intended for different people, and the support of interactivity. As part of the activities carried out in the TA2 project, the system presented in this paper is a tool for end-users to nurture relationships.

 

This paper reports on an architecture, and a working implementation, for using secondary screens in the interactive television environment. While there are specific genres and programs that immerse the viewer into the television experience, there are situations in which people perform as well a secondary task, whilst watching. In the living room, people surf the web, use email, and chat using one or many secondary screens. Instead of focusing on unrelated activities to television watching, the architecture presented in this paper aims at related activities, i.e., to leverage the user impact on the content being watched. After a comprehensive literature review and working systems analysis, the requirements for the secondary screen architecture are identified and modeled in the form of a taxonomy. The taxonomy is divided into three high-level categories: control, enrich, and share content. By control we refer to the decision what to consume and where to render it. In addition, the viewer can use the secondary screen for enriching media content and for sharing the enriched material. The architecture is validated based on the taxonomy and by an inspection of the available services. The final intention of our work is to leverage the viewersÕ control over the consumed content in our multi-person, multi-device living rooms.

The research area of interactive digital TV is in the midst of a significant revival. Unlike the first generation of digital TV, which focused on producer concerns that effectively limited (re)distribution, the current generation of research is closely linked to the role of the user in selecting, producing, and distributing content. The research field of interactive digital television is being transformed into a study of human-centered television. Our guest editorial reviews relevant aspects of this transformation in the three main stages of the content lifecycle: content production, content delivery, and content consumption. While past research on content production tools focused on full-fledged authoring tools for professional editors, current research studies lightweight, often informal end-user authoring systems. In terms of content delivery, user-oriented infrastructures such as peer-to-peer are being seen as alternatives to more traditional broadcast solutions. Moreover , end-user interaction is no longer limited to content selection, but now facilitates nonlinear participatory television productions. Finally, user-to-user communication technologies have allowed television to become a central component of an interconnected social experience. The background context given in this article provides a framework for appreciating the significance of four detailed contributions that highlight important directions in transforming interactive television research.

This paper presents a new approach for media presentation continuity in playback mode. We use the term presentation continuity over session transfer since our solution is at the presentation layer. Previous research on this topic has focused on transferring a particular stream or set of related streams at the sessions layer. We argue that in the realm of service oriented architectures, such as telecom operator networks, this approach is not the best solution for media playback. Our mechanism presents an alternative to the traditional approach, recognising the fact that a user is connected to a media presentation, which, may be composed of multiple sessions. The advantages of our system are i) Lower network control plane overhead, thus reducing chances of presentation consistency loss ii) Lower network data overhead due to lesser need for transcoding iii) Delegating presentation consistency issues, such as inter-media synchronisation, to the media player iv) Dynamically adapting the presentation to the new target devices without transcoding. At the end of this paper we present experimental results, showing a comparison with previous approaches.

Media consumption is an inherently social activity, serving to communicate ideas and emotions across both small- and large-scale communities. The migration of the media experience to personal computers retains social viewing, but typically only via a non-social, strictly personal interface. This paper presents an architecture and implementation for media content selection, content (re)organization, and content sharing within a user community that is heterogeneous in terms of both participants and devices. In addition, our application allows the user to enrich the content as a differentiated personalization activity targeted to his/her peer-group. We describe the goals, architecture and implementation of our system in this paper. In order to validate our results, we also present results from two user studies involving disjoint sets of test participants.

In this paper, we look at how television genres can play a role in the use of social interactive television systems (social iTV). Based on a user study of a system for sending and receiving enriched video fragments to and from a range of devices, we discuss which genres are preferred for talking while watching, talking about after watching and for sending to users with different devices. The results show that news, soap, quiz and sport are genres during which our participants talk most while watching and are thus suitable for synchronous social iTV systems. For asynchronous social iTV systems film, news, documentaries and music programs are potentially popular genres. The plot structure of certain genres influences if people are inclined to talk while watching or not, and to which device they would send a video fragment. We also discuss how this impacts the design and evaluation of social iTV systems.

In this paper we examine adaptive time-based web applications (or presentations). These are interactive presentations where time dictates the major structure, and that require interactivity and other dynamic adaptation. We investigate the current technologies available to create such presentations and their shortcomings, and suggest a mechanism for addressing these shortcomings. This mechanism, SMIL State, can be used to add user-defined state to declarative time-based languages such as SMIL or SVG animation, thereby enabling the author to create control flows that are difficult to realize within the temporal containment model of the host languages. In addition, SMIL State can be used as a bridging mechanism between languages, enabling easy integration of external components into the web application.

This paper demonstrates the advantages of using structured multimedia documents for session management and media distribution in ubiquitous environments. We show how document manipulations can be used to perform powerful operations such as content to context adaptation and presentation continuity. When consuming media in ubiquitous environments, where the set of devices surrounding a user may change, dynamic media adaptation and session transfer become primary requirements. This paper presents a working system, based on a representative scenario, in which multimedia content is distributed and adapted to a movable user to best suit his/her contextual situation. The implemented scenario includes the following scenes: content selection using a personal mobile phone, content distribution to the most suitable device according to the user's context, and presentation continuity when the user moves to another location. This paper introduces the underlying document manipulations that turn the scenario into a working system.

This working session will be an interactive discussion about multimedia content transformation. The basic assumption is that content transformation activities should be provided as non-destructive operations. The final goal of the panel is to gather researchers within the community interested in manipulating multimedia content for providing rich user experiences. The organizers of the panel will moderate and shape the discussion; nevertheless, position papers from the participants are expected.

This paper investigates a number of techniques and services around a unifying concept: the secondary screen. Far too often television consumption is considered a passive activity. While there are specific genres and programs that immerse the viewer into the media experience, there are other times in which whilst watching television, people talk, scan the program guide, record another program or recommend a program by phone. This paper identifies four major usages of the secondary screen in an interactive digital television environment: control, enrich, share, and transfer television content. By control we refer to the decoupling of the television stream, optional enhanced content, and television controls. Moreover, the user can use the secondary screen to enrich or author media content by, for example, including personalized media overlays such as an audio commentary that can be shared with his peer group. Finally, the secondary screen can be used to bring along the television content. This paper reviews previous work on the secondary screen, identifies the key usages, and based on a working system provides the experiences of developing relevant scenarios as well as an initial evaluation of them.

Despite most of the authoring systems for digital TV assume the author to be seated in front of a computer on the broadcaster side, current research is interested in the new role of the viewer in producing and distributing content. The goal of this paper is to identify a number of research directions around the authoring from the couch paradigm, an entertainment-oriented approach in which the authoring task is performed incidentally.

This demo presents a new approach for media presentation continuity in playback mode. We use the term presentation continuity over session transfer since our solution is at the presentation layer. Previous research on this topic has focused on transferring a particular stream or set of related streams at the sessions layer. Our approach presents an alternative, recognising the fact that a user is connected to a media presentation, which, may be composed of multiple sessions. The advantages of our approach are i) Lower network control plane overhead, thus reducing chances of semantic presentation loss ii) Lower network data overhead due to lesser need for transcoding iii) delegating presentation semantic issues, such as inter-media synchronisation, to the player iv) dynamically adapt the presentation to the new target devices without transcoding.

This paper introduces an architecture, together with an implemented scenario, capable of dynamically adapt the way mobile users consume and interact with multimedia content. The architecture is based on a representative scenario identified by the European project SPICE, in which multimedia content is distributed to a user independently of his/her contextual situation. The implemented scenario includes the following scenes: content selection using a personal mobile phone, content distribution to the most suitable device according to the user's context, and presentation continuity when the user moves to another location. This paper reports on the defined architecture and the current status of its implementation. It shows the initial results in the form of screenshots.

This paper describes an approach to defining, manipulating and sharing state variables between a web browser and a multimedia presentation engine in functionally compound XML-based documents. This framework, which we call smilState: the SMIL XML State Architecture, is a fully declarative approach to sharing state without the need for extensive scripting. The state variables in smilState are defined using standard Web technologies such as XPath, XForms and XSLT, which have been integrated with a SMIL-specific temporal component. The smilState architecture enables interactive, user-centered applications to be created that allow temporal semantics that extend beyond the facilities currently available for integrating a conventional (X)HTML browser interface with SMIL, SVG or HTML+Time content. The primary benefit of this work is that it adds a controlled temporal dimension to existing state architectures that is free of document scheduling side effects within the multimedia content. This paper provides a use cases for temporal sharing of state across documents, it describes the smil- State architecture in detail, and it describes an implementation of smilState within the open-source Ambulant SMIL player.

In order to make multimedia a first-class citizen on the Web, there is a need for major efforts across the community. European projects such as Passepartout (ITEA) and SPICE (IST IP) show that there is a need for a standardized mechanism to provide rich interaction for continuous media content. CWI is helping to build a framework that adds a temporal dimension to existing a-temporal Web browsers.

For this special issue on Multimedia Systems, Languages, and Infrastructures for Interactive Television the four best papers on multimedia systems and infrastructures were invited to extend their conference contribution in the form of a journal paper. These papers cover a wide range of current challenges for multimedia systems: content recommendation, participatory multimedia genres, evaluation of mobile media usage, and digital media narratives.

Coming soon...

Coming soon...

Coming soon...

Coming soon...

Coming soon...

Coming soon...

SMIL 3.0 is the next version of the W3C Synchronized Multimedia Integration Language. This version supports a wide range of fundamental enhancements in the areas of metadata, state, animation, timing, layout and media handling. It also includes a new internal timed text datatype.
The first public draft version of the language allows interested users from within and outside W3C to comment on this evolving language.

Coming soon...

Coming soon...

The focus of much of the research on providing user-centered control of multimedia has been on the definition of models and (meta-data) descriptions that assist in locating or recommending media objects. While this can provide a more efficient means of selecting content, it provides little extra control for users once that content is rendered. In this article, we consider various means for supporting user-centered control of media within a collection of objects that are structured into a multimedia presentation. We begin with an examination of the constraints of user-centered control based on the characteristics of multimedia applications and the media processing pipeline. We then define four classes of control that can enable a more user-centric manipulation within media content. Each of these control classes is illustrated in terms of a common news viewing system. We continue with reflections on the impact of these control classes on the development of multimedia languages, rendering infrastructures and authoring systems. We conclude with a discussion of our plans for infrastructure support for user-centered multimedia control.

Coming soon...

Coming soon...

SMIL 3.0 is the next version of the W3C Synchronized Multimedia Integration Language. This version supports a wide range of fundamental enhancements in the areas of metadata, state, animation, timing, layout and media handling. It also includes a new internal timed text datatype.
The first public draft version of the language allows interested users from within and outside W3C to comment on this evolving language.

Coming soon...

Coming soon...

Coming soon...

Coming soon...

Coming soon...

Coming soon...

SMIL 2.1 Layout provides two classes of changes to SMIL 2.0 layout. First, the SMIL 2.0 HierarchicalLayout module has been replaced by the SubRegionLayout, AlignmentLayout, and OverrideLayout modules; this allows differentiated features to be implemented in profiles without necessarily requiring support for all of the functionality in the HierarchicalLayout module. Second, several new elements and attributes have been added to SMIL 2.1 Layout to provide expression for common functions in an authoring-efficient manner; these functions include the short-cut notations for media positioning now available in the AlignmentLayout module and support for background image tiling in the BackgroundTilingLayout. In addition, new support for simple audio positioning has been added that allows audio placement to be supported by those players that allow audio 2-D imaging. The new OverrideLayout module groups existing support for per-media-object overrides of BasicLayout attributes.

The SMIL 2.1 Mobile Profile is a collection of SMIL 2.1 modules that provide support for the SMIL 2.1 language within the context of a mobile device. Such a device is expected to have sufficient display, memory, and processor capabilities to render basic interactive multimedia presentations in SMIL. The SMIL 2.1 Mobile Profile is a super-set of the SMIL 2.1 Basic Profile and a sub-set of the SMIL 2.1 Extended Mobile Profile. The SMIL 2.1 Mobile profile is largely compatibility with the SMIL profile that third Generation Partnership Program (3GPP) has defined for the multimedia messaging (MMS) and the enhanced packed switched streaming (e-PSS) mobile services in its own specification ([3GPP26.246R6]).
The functionality of the SMIL 2.1 Mobile Profile may be further extended by using the SMIL 2.1 Scalability Framework. When extending the functionality of this profile, it is highly recommended to include functionality from the SMIL 2.1 Extended Mobile Profile first.

The Media Object Modules for SMIL 2.1 introduces a facility to predefine sets of common param element values in the document head section and a facility to refer to these definitions from media object references within the body section. This change is made to reduce the size of a SMIL document containing many similar parameter definitions and to ease the authoring and maintenance of SMIL 2.1 documents that use the elements and attributes in the MediaParam Module.

Authoring context sensitive, interactive multimedia presentations is much more complex than authoring either purely audiovisual applications or text. Interactions among media objects need to be described as a set of spatio-temporal relationships that account for synchronous and asynchronous interactions, as well as on-demand linking behavior. This paper considers the issues that need to be addressed by an authoring environment. We begin with a partitioning of concerns based on seven classes of authoring problems. We then describe a selection of multimedia authoring environments within four different authoring paradigms: structured, timeline, graph and scripting. We next provide observations and insights into the authoring process and argue that the structured paradigm provides the most useful framework for presentation authoring. We close with an example application of the structured multimedia authoring paradigm in the context of our own structure-based system GRiNS.

This article provides a review of the state of metadata architecture and application. It starts with a short story that serves as a background to the article, the goal being to discover whether metadata is, in fact, the greatest thing since sliced bread.

The TabletPC is an example of a new generation of user interface device where pen-based manipulation of information is integrated directly into a user’s workflow. Using the TabletPC's existing pen and electronic ink systems, a wide range of static documents can be created or annotated. While the facilities of the TabletPC are useful for creating virtual images containing ink that can be overlaid on text or picture context, there is little support for creating annotations of time-based content such as video.
This article describes an annotation authoring model and interface for creating peer-level annotations to video media. Peer-level annotations allow existing content to be enriched with additional content annotations that can be co-presented with the original media. A system for creating a SMIL language document containing SVG-based annotations that exist along-side the visual content is described, along with a discussion of the needs and limitations of supporting video markup in a web context. An example using peer-level annotations in a medical context is provided.

This paper provides an overview of the Ambulant Open SMIL player. Unlike other SMIL implementations, the Ambulant Player is a reconfigureable SMIL engine that can be customized for use as an experimental media player core. The Ambulant Player is a reference SMIL engine that can be integrated in a wide variety of media player projects. This paper starts with an overview of our motivations for creating a new SMIL engine, then discusses the architecture of the Ambulant Core (including the scalability and custom integration features of the player). We close with a discussion of our implementation experiences with Ambulant instances for Windows, Mac and Linux versions for desktop and PDA devices.

This paper discusses the work-flow control features provided by the GRiNS editor for creating SMIL presentations. We start with an overview of the generic presentation creation process workflow and then introduce the general features supported by GRiNS. We then follow with a detailed set of examples of how these features can be used to create a simple slideshow of the type that can be played in mobile devices. We then provide an analysis of the use of GRiNS’s feature set and contrast these with features found in other SMIL editors. We close with a set of directions for future work in supporting a presentation workflow.

The SMIL 2.0 Language profile support several mechanisms for controlling interactivity in a SMIL 2.0 presentation. Unfortunately, the SMIL standard testset does not verify complex interactions of linking/interaction behavior of SMIL players and applications. This paper describes a linking and interaction test suite that was developed as part of the Ambulant SMIL Player project. We begin with a short review of SMIL’s linking and interaction facilities, then describe aspects of the test suite that have proven to highlight faults in current SMIL players.

This paper discusses applying facilities in SMIL 2.0 to the problem of annotating multimedia presentations. Rather than viewing annotations as collections of (abstract) meta-information for use in indexing, retrieval or semantic processing, we view annotations as a set of peer-level content with temporal and spatial relationships that are important in presenting a coherent story to a user. The composite nature of the collection of media is essential to the nature of peer-level annotations: you would typically annotate a single media item much differently than that same media item in the context of a total presentation.
This paper focuses on the document engineering aspects of the annotation system. We do not consider any particular user interface for creating the annotations or any back-end storage architecture to save/search the annotations. Instead, we focus on how annotations can be represented within a common document architecture and we consider means of providing document facilities that meet the requirements of our user model. We present our work in the context of a medical patient dossier example.

A new generation of tablet computers has stimulated end-user interest on annotating documents by making pen-based commentary and spoken audio labels to otherwise static documents. The typical application scenario for most annotation systems is to convert existing content to a (virtual) image, capture annotation mark-up, and to then save the annotations is a database. While this is often acceptable for text documents, most multimedia documents are timesensitive and can be dynamic: content can change often depending on the types of audio/video data used. Our work looks at expanding the possibilities of annotation by integrating annotations onto timed basis media. This paper discusses the AMBULANT Annotation Architecture. We describe requirements for multimedia annotations, the multimedia annotation architecture being developed at CWI, and initial experience from providing various classes of temporal and spatial annotation within the domain of medical documents..

This document provides background information to create slideshows for the RealOne, IE-6’s HTML+TIME and 3GPP/Mobile players using the GRiNS Pro Editor for SMIL 2.0 software (hereafter called simply: GRiNS/ Pro). We discuss how to create simple slideshows and how to integrate transitions, animations and links for several SMIL players, as well as creating links to the RealOne media browser and related info panes. You will also learn how to publish your presentation for use with the RealOne player and a streaming server and various other SMIL 2.0 media players.

The article is the second part of a two-part series on SMIL 2.0, the newest version of the World Wide Web Consortium's Synchronized Multimedia Integration Language. Part 1 looked in detail at various aspects of the SMIL specification and the underlying SMIL timing model. This part looks at simple and complex examples of SMIL 2.0's use and compares SMIL with other multimedia formats. We focus on SMIL's textual structure in its various implementation profiles.

The World Wide Web Consortium's Synchronized Multimedia Integration Language format for encoding multimedia presentations for delivery over the Web is a little-known but widely used standard. First released in mid-1998, SMIL has been installed on approximately 200,000,000 desktops worldwide, primarily because of its adoption in RealPlayer G2, Quicktime 4.1, and Internet Explorer 5.5. In August 2001, the W3C released a significant update with SMIL 2.0. In a two-part report on SMIL 2.0, the author will discuss the basics of SMIL 2.0 and compare its features with other formats. This article will focus on SMIL's basic concepts and structure. Part two, in the January-March 2002 issue, will look at detailed examples of SMIL 2.0, covering both simple and complex examples. It'll also contrast the facilities in SMIL 2.0 and MPEG-4.

This document specifies the second version of the Synchronized Multimedia Integration Language (SMIL, pronounced "smile"). SMIL 2.0 has the following two design goals:
* Define an XML-based language that allows authors to write interactive multimedia presentations. Using SMIL 2.0, an author can describe the temporal behavior of a multimedia presentation, associate hyperlinks with media objects and describe the layout of the presentation on a screen.
* Allow reusing of SMIL syntax and semantics in other XML-based languages, in particular those who need to represent timing and synchronization. For example, SMIL 2.0 components are used for integrating timing into XHTML [XHTML10] and into SVG [SVG].

As end-user bandwidth increases to a level where the (re)distribution of audio/video material via the Internet becomes attractive, XML-based standards that help broadcasters migrate their existing content to the Web are becoming richer and more powerful. SMIL 2.0 – developed by the World Wide Web Consortium (W3C) – is the newest version of the Web’s most popular multimedia format. This article provides an introduction to the concepts and facilities of the SMIL 2.0 language, in the context of the work flow requirements for taking existing broadcast content and making it available for a Web-centric audience.

This essay presents a brief discussion of combining temporal aspects of multimedia presentations with hypertext links. Three ways of combining linking with temporally synchronized components of a presentation are described. We describe work that has been done to incorporate both temporal and linking information within the W3C language SMIL (Synchronized Multimedia Integration Language). We conclude with a discussion of future directions, namely providing support for linking within and among non-linear presentations and the ability to add temporal information to existing XML document languages.

The W3C has recently released a language for Web-based Multimedia presentations called SMIL: the Synchronized Multimedia Integration Language. GRiNS is an authoring and presentation environment that can be used to create SMIL-compliant documents and to play SMIL documents created with GRiNS or by hand. This paper discusses GRiNS as a tool for creating multimedia education materials on the Web.

SMIL is the W3C recommendation for bringing synchronized multimedia to the Web. Version 1.0 of SMIL was accepted as a recommendation in June. Work is expected to be soon underway for preparing the next version of SMIL, version 2.0. Issues that will need to be addressed in developing version 2.0 include not just adding new features but also establishing SMIL's relationship with various related existing and developing W3C efforts. In this paper we offer some suggestions for how to address these issues. Potential new constructs with additional features for SMIL 2.0 are presented. Other W3C efforts and their potential relationship with SMIL 2.0 are discussed. To provide a context for discussing these issues, this paper explores various approaches for integrating multimedia information with the World Wide Web. It focuses on the modeling issues on the document level and the consequences of the basic differences between text-oriented Web-pages and networked multimedia presentations.

Making hypermedia adaptable for multiple forms of presentation involves enabling multiple distinct specifications for how a given collection of hypermedia can have its presentation generated. The Standard Reference Model for Intelligent Multimedia Presentation Systems describes how the generation of hypermedia presentation can be divided into distinct but cooperating layers. Earlier work has described how specifications for generating presentations can be divided into distinct modules of code corresponding to these layers. This paper explores how the modules for each layer of a presentation specification can be exchanged for another module encoded for that layer and result in the whole specification remaining well functioning. This capability would facilitate specifying presentation generation by allowing for the use of pre-programmed modules, enabling the author to focus on particular aspects of the presentation generation process. An example implementation of these concepts that uses current and developing Web standards is presented to illustrate how wide-spread modularized presentation generation might be realized in the near future.

Most hypermedia documents don't incorporate time explicitly. This prevents authors from having direct control over the temporal aspects of a presentation. In this paper we discuss the concept of presentation time - the timing of the individual parts of a presentation and the temporal relations among them. We argue why time is necessary from a presentation perspective, and discuss its relationship with other temporal views of a presentation. We derive the requirements and present our solution for incorporating temporal and linking information in a model of time-based hypermedia.

Having the content of a presentation adapt to the needs, resources and prior activities of a user can be an important benefit of electronic documents. While part of this adaptation is related to the encodings of individual data streams, much of the adaptation can/should be guided by the semantics in and among the objects of the presentation. The semantics involved in having hypermedia presentations adapt can be divided between adaptive hypermedia, which adapts autonomously, and adaptable hypermedia, which requires presentationexternal intervention to be adapted. Understanding adaptive and adaptable hypermedia and the differences between them helps in determining the best manner with which to have a particular hypermedia implementation adapt to the varying circumstances of its presentation. The choice of which type of semantics to represent can affect speed of the database management system processing them. This paper reflects on research and implementation approaches toward both adaptive and adaptable hypermedia and how they apply to specifying the semantics involved in hypermedia authoring and processing. We look at adaptive approaches by considering CMIF and SMIL. The adaptable approaches are represented by the SGML-related collection of formats and the Standard Reference Model (SRM) for IPMS are also reviewed. Based on our experience with both adaptive and adaptable hypermedia, we offer recommendations on how each approach can be supported at the data storage level.

This paper describes document modelling con­structs that support alternate content choices for generalized hypermedia presentations. While there has been much work done on adaptive hypermedia documents in the context of low-level quality-of-service adaptation, little atten­tion has been paid to support of user-level adap­tation of multimedia content. Taking examples from the domains of information accessibility for the visual/hearing impaired, multi-lingual information presentation, and content adapta­tion in distance learning, we show how simple interfaces to rich hypermedia documents can give decided benefits to the user community. We discuss our work in terms of experiments from the CWI CMIF project and indicate how these solutions have been integrated with the W3C SMIL language in the GRiNS editor and player for Web use.

In order to facilitate adaptability of hypermedia documents a distinction is often made between the underlying conceptual structure of a document and the structure of its presentation. This distinction enables greater variety in how a presentation can be adapted to best convey these underlying concepts in a given situation. What is often confusing for those applying this distinction is that although both levels of structure often share similar components: transformation form the storage of a document to its presentation sometimes occurs directly between these similar components and sometimes does not. These similarities typically fall in the categories of space, time and relationships between document portions. This paper identifies some primary similarities between the structure of hypermedia storage and presentation. It also explores how the transformation from storage to presentation often does not follow these similarities. This discussion is illustrated with the Fiets hypermedia application, which addresses the issues of storage, presentation and transformation using public domain formats and tools. The intention is to help authors who separate storage from presentation to better understand this distinction.

This paper discusses the implementation of adaptability in environments that are based on the Standard Reference Model for Intelligent Multimedia Presentation Systems. This adaptability is explored in the context of style sheets, which are represented in such formats as DSSSL. The use of existing public standards and tools for this implementation of style sheet-based adaptability is described. The Berlage environment is presented, which integrates these standards and tools into a complete storage-to-presentation hypermedia environment. The integration of the SRM into the Berlage environment is introduced in this work. This integration illustrates the issues involved in implementing adaptability in the model.

In order for multimedia presentations to be stored, accessed and played from a large library they should not be encoded as final form presentations, since these consume storage space and cannot easily be adapted to variations in presentation-time circumstances such as user characteristics and changes in end-user technology. Instead, a more presentation independent approach needs to be taken that allows the generation of multiple versions of a presentation based on a presentation-independent description.
In order for such a generated presentation to be widely viewable, it must be in a format that is widely implemented and adopted. Such a format for hypermedia presentations does not yet exist. However, the recent release of SMIL, whose creation and promotion is managed by the World Wide Web Consortium, promises to become such a format in the short term and be for hypermedia what HTML is for hypertext.
The technology for enabling this presentation-independent approach is already available, but requires the use of large and unapproachable standards, such as DSSSL and HyTime. In this paper we show that these two standards can be used with SMIL, and by concentrating on a particular application, illustrate the use of publicly available tools to support the generation of multiple presentations from a single presentation-independent source.

This document specifies version 1 of the Synchronized Multimedia Integration Language (SMIL 1.0, pronounced "smile"). SMIL allows integrating a set of independent multimedia objects into a synchronized multimedia presentation. Using SMIL, an author can
1. describe the temporal behavior of the presentation
2. describe the layout of the presentation on a screen
3. associate hyperlinks with media objects
This specification is structured as follows: Section 1 presents the specification approach. Section 2 defines the "smil" element. Section 3 defines the elements that can be contained in the head part of a SMIL document. Section 4 defines the elements that can be contained in the body part of a SMIL document. In particular, this Section defines the time model used in SMIL. Section 5 describes the SMIL DTD.

This paper investigates the main issues related to the translation of SMIL into MHEG
documents This is driven by the more general objective to achieve interoperability between the domains of Web and digital TV where MHEG
is used in the digital TV environment and SMIL is the Web format to specify interactive synchronized multimedia presentations A summary of both formats is presented on the basis of which it is shown how SMIL translates into MHEG
 Although the formats have dif ferences such translation appears to be feasible Aspects of authoring for both domains and other interoperability issues are discussed.

The content and structure of an electronically published document can be authored and processed in ways that allow for flexibility in presentation on different environments for different users. This enables authors to craft documents that are more widely presentable. Electronic publishing issues that arise from this separation of document storage from presentation include (1) respecting the intent and restrictions of the author and publisher in the document's presentation, and (2) applying costs to individual document components and allowing the user to choose among alternatives to control the price of the document's presentation. These costs apply not only to the individual media components displayed but also to the structure created by document authors to bring these media components together as multimedia.
A collection of ISO standards, primarily SGML, HyTime and DSSSL, enable the representation of presentation-independent documents and the creation of environments that process them for presentation. SMIL is a W3C format under development for hypermedia documents distributed on the World Wide Web. Since SMIL is SGML-compliant, it can easily be incorporated into SGML/HyTime and DSSSL environments.
This paper discusses how to address these issues in the context of a presentation-independent hypermedia storage. It introduces the Berlage environment, which uses SGML, HyTime, DSSSL and SMIL to store, process, and present hypermedia data. This paper also describes how the Berlage environment can be used to enforce publisher restrictions on media content and to allow users to control the pricing of document presentations. Also explored is the ability of both SMIL and HyTime to address these issues in general, enabling SMIL and HyTime systems to consistently process documents of different document models authored in different environments.

The W3C working group on synchronized multimedia has developed a language for Web-based Multimedia presentations called SMIL: the Synchronized Multimedia Integration Language. This paper presents GRiNS, an authoring and presentation environment that can be used to create SMIL-compliant documents and to play SMIL documents created with GRiNS or by hand.

A number of different systems exist for creating multimedia or hypermedia applications—each with its own internal document model. This leads to problems in comparing documents created by these systems, and in describing the information captured by a document for long-term (system independent) storage and future playback. We discuss the components which should be considered for a hypermedia document model. These include the hierarchical and linking structure of a document and the spatial and temporal relations among components of the document. Other aspects, such as transportability of documents and information retrieval, are also addressed briefly. We present the Amsterdam Hypermedia Model which, while expressing only a subset of all possible structures, has been used as a basis for a comprehensive authoring environment.

The World-Wide Web has been used extensively to present hypertext documents that have a limited mixture of text and simple graphics which are distributed via the public Internet. The performance characteristics of the Internet have made the delivery of complex multimedia documents (that is, documents that include time-based components) difficult. An effort is currently underway by members of industry, research centers and user groups to define a standard document format that can be used in conjunction with time-based transport protocols over Inter- and Intranets to support rich multimedia presentations. This paper outlines the goals of the W3C’s Synchronized Multimedia working group and presents an initial description of the first version of the proposed multimedia document model and format.

Being able to author a hypermedia document once for presentation under a wide variety of potential circumstances requires that it be stored in a manner that is adaptable to these circumstances. Since the nature of these circumstances is not always known at authoring time, specifying how a document adapts to them must be a process that can be performed separately from its original authoring. These distinctions include the porting of the document to different platforms and formats and the adapting of the document s presentation to suit the needs of the user and of the current state of the presentation environment. In this paper we discuss extensions to our CMIF hypermedia authoring and presentation environment that provide adaptability through this distinction between authoring and presentation specification. This extension includes the use of HyTime for document representation and of DSSSL for presentation specification. We also discuss the Berlage architecture, our extension to HyTime that specifies the encoding of certain hypermedia concepts useful for presentation specification.

Being able to author a hypermedia document once for presentation under a wide variety of potential circumstances requires that it be stored in a manner that is adaptable to these circumstances. Since the nature of these circumstances is not always known at authoring time, specifying how a document adapts to them must be a process that can be performed separately from its original authoring. These distinctions include the porting of the document to different platforms and formats and the adapting of the document s presentation to suit the needs of the user and of the current state of the presentation environment. In this paper we discuss extensions to our CMIF hypermedia authoring and presentation environment that provide adaptability through this distinction between authoring and presentation specification. This extension includes the use of HyTime for document representation and of DSSSL for presentation specification. We also discuss the Berlage architecture, our extension to HyTime that specifies the encoding of certain hypermedia concepts useful for presentation specification.

We consider the generic hypermedia structure of a document to be a means of representing the document that allows it to be processed into a wide variety of presentations. Representing a document in this manner requires additional specification and resources to render it into any presentation. In this paper we discuss the relationship between the generic hypermedia structure of documents and the processing of this structure into presentation. Our discussion is expressed in terms of existing models for hypertext and hypermedia systems and also in terms of ISO standards for text and hypermedia document formatting and processing. The discussion and the resulting formalisms and the resulting formalisms are illustrated with extension designs for the hypermedia authoring and presentation environment developed at our laboratory.

The standard reference model (SRM) for intelligent multimedia presentation systems describes a framework for the automatic generation of multimedia presentations. This framework, however, lacks an explicit document model of the presentation being generated. The Amsterdam hypermedia model (AHM) describes the document features of a hypermedia presentation explicitly. We take the AHM and use it as a basis for describing in detail the stages of generating a hypermedia presentation within the SRM framework, which we summarise in a table. By doing so the responsibilities of the individual SRM layers become more apparent.

In this position statement, challenges in three areas of information and data interface devel­opment are considered. These are: interfaces to the technology that are used to deliver commu­nication primitives; interfaces to the presentation of a single information projection on a particu­lar system; and interfaces to the abstract informa­tion containing the ultimate message that the technological messenger is trying to present. Since the nature of this statement makes the presentation cursory, I will focus on the needs of the user rather than the potential of the infrastructure.

Coming soon...

As the availability of digital video becomes commonplace, a shift in application focus will occur from merely accessing video as an independent data stream to embedding video with other multimedia data types into coordinated hypermedia presentations. The migration to embedded video will present new demands on applications and support environments: processing of any one piece of video data will depend on how that data relates to other data streams active within the same presentation. This article describes presentation, synchronization and interaction control issues for manipulating embedded video. First, we describe the requirements for embedded video, contrasted against other forms of video use. Next we consider mechanisms for describing and implementing the behavior of embedded video segments relative to other data items in a document; these relationships form the basis of implementing cooperative control among the events in a presentation. Finally, we consider extending the possibilities for tailoring embedded video to the characteristics of the local runtime environment; this forms the basis for adaptive, application-level quality of service control of a presentation. In all cases, we describe a mechanism to externalize the behavior of hypermedia presentations containing resource intensive data requirements so that effective control can be implemented by low-level system facilities based on application-specific requirements. We present our results in terms of the CMIFed authoring/presentation system.

The integration of audio, video, graphics and text on the desktop promises to fundamentally challenge the centuries-old model of the printed document as the basis for information exchange. Before this potential can be realized, however, systems must be devised that enable the production and presentation of complex, inter-related media objects. These systems are generically called multimedia authoring tools. In this article, we consider the development of multimedia authoring tools, examine the current state of the art, and then discuss a set of research challenges that need to be addressed before the full potential of multimedia output technology can be effectively utilized to share information.

We give a short description of the Amsterdam Hypermedia Model followed by examples of its use in a number of existing and planned applications. The main application to date has been as a basis of the multimedia authoring system, CMIFed, along with its ability to specify trade-offs for resource use. We discuss the model's potential for generating differing document formats, followed by future work on using it as a goal format for generating multimedia documents.

There are two major problems with the current ways of creating interactive multimedia presentations. Firstly, authoring a multimedia presentation is a non-trivial task, requiring a range of skills such as creating individual items in each medium as well as combining these into a coherent presentation. Secondly, after having devoted a large amount of time and effort to the creation of a presentation, there is no guarantee that it can be played back on a platform other than the one for which it was created, let alone whether it can be played back by future systems.
To tackle both these problems, we first present an information model for interactive multimedia, so that information can be stored independently of the system that creates or plays it. We then investigate a number of authoring systems. By differentiating four authoring paradigms we classify and describe a selection of both research and commercial systems. These provide examples of the types of support that can be given to authors, and how this support can be provided in practice. Using the approaches and features supported by these systems as a base, we give an analysis of the facilities desired in an ideal authoring system.

We present an approach for generating hypermedia presentations from multimedia information items distributed around a network. Our goal is to create a media-independent description of a presentation, from which multiple final presentations can be generated, taking into account the user's information need, the user's task and network and end-user platform resources.
In order to generate the structure of a hypermedia presentation from existing media items we need to define a way of grouping similar items and making links among the groups. This grouping can be based on semantic annotations attached to the media items. Current approaches to video annotation, as a complex example, are analysed. A number of research questions arising from our approach are discussed.

The increasing availability of high-speed networks has served as an enabling technology for applications such as networked multimedia, where a mix of information types can be retrieved from decentralized information stores and presented at a user's workstation. Such transfers work best when a guaranteed amount of network capacity is presented to the application fetching the data. Unfortunately, as the storage of information becomes more decentralized, and as the load on individual information sources and sinks increase, it is often difficult to obtain service guarantees for the duration of a lengthy information exchange-especially when the information access is determined by the dynamic behaviour of users, such as is present in a hypermedia application environment. This project defined an experiment in providing adaptive, runtime control over hypermedia information in a wide-area network. The purpose of the experiment was to determine the adaptive control mechanisms required to provide decentralized access to complex data in an unpredictable networked environment. The unpredictability of the environment may be caused by transient reallocations of network bandwidth, overloading of network servers or reliability problems within the communications infrastructure. In the following sections, we describe the planned and encountered: environment, method and expected results of a series of quality-of-service experiments conducted over moderate-bandwidth links between various European network organizations. We then discuss problems in producing the planned results. We conclude with a travel/expense summary for the project.

On the surface, hypermedia is a simple and natural extension of multimedia and hypertext: multimedia provides a richness in data types that facilitates flexibility in expressing information, while hypertext provides an elegant way of navigating through this data in a context-based manner. One popular approach to supporting hypermedia is to take an existing hypertext model and augmenting it with multimedia data types within the storage model. Although such a `marriage of convenience' can provide some immediate results, the underlying control assumptions of the hypertext model make this approach unsuitable for describing and supporting generalized hypermedia documents. In particular, conventional hypertext cannot adequately support complex temporal relationships among data items, specifications that support high-level presentation semantics or a notion of `information context' that specifies global behavior when following links-all elements that are of fundamental importance in supporting multimedia.

The term multimedia conjures up visions of desktop computers reproducing digital movies, high-resolution images and stereo sound. While many current systems support such functionality, none do so elegantly-especially when data is fetched and synchronized from dissimilar sources distributed across resource-limited networks. Our research investigates general approaches for managing the flow of multimedia information in a distributed computing environment, providing adaptive support for time-sensitive retrieval and presentation based on multimedia document specifications. The benefit of our approach is that it provides flexible, content-based utilization of resources without overburdening the application author/developer.

The creation of a multimedia presentation is a non-trivial task. It involves skills that are not readily available to users and it requires support not generally available from authoring software. In order to understand the basic problems of multimedia authoring, this article considers the requirements for defining interactive, dynamic presentations. When contrasted against the facilities available in current-generation commercial authoring systems, we can see that their focus is often on low-level details rather than high-level structure. The prospects for future editing systems are somewhat brighter: support for high level editing can be provided. As an example, we describe the CMIFed authoring environment; CMIFed not only supports authoring at a high level but also incorporates most low-level features found in current systems.

CWI has been investigating the placement and use of high-speed networks for use in international, national and local communication. As part of a multi-year grant from the Dutch Government, we are currently upgrading our local network infrastructure to make use of ATM technology, providing an environment to support research in areas such as multimedia and scientific visualization, and to improve general network service to our users. We have conducted several performance tests during the past year to familiarize ourselves with the technology and its potential in servicing user needs. Our initial results indicate that while installation of network interfaces is relatively trivial, much work remains to be done before the full potential of high-speed communication can be enjoyed by general applications.

Supporting intelligent multi-media multi-modal systems is a broad problem that has many facets. One of these facets is defining the phrase `intelligent multi-media multi-modal systems': where one puts the intelligence, when one supports the multi-modality and how one provides multi-media capabilities depend greatly on one's perspective and assumptions about how information is to be manipulated and presented within a user/computer environment. From the perspective of the user, the entire problem may be paraphrased as follows: "how can information be presented (and extracted) to make the solution of a set of problems more accessible." From the perspective of the `system,' the problem can be paraphrased as being: "how can information be manipulated and presented to aid in the development of a solution." In both cases, an emphasis exists on transforming information so that it can be used to better suit the needs of a (phase of an) application. In general, we refer to these transformations with respect to various static and dynamic data types as adaptive multimedia.

Taking the concept of a link from hypertext and adding it to the rich collection of information formats found in multimedia systems provides an intuitive extension to multimedia that is often called hypermedia. Unfortunately, the implicit assumptions under which hypertext links work do not extend well to time-based presentations that consist of a number of simultaneously active media items. It is not obvious where links should lead and there are no standard rules that indicate what should happen to other parts of the presentation that are active. This paper addresses the problems associated with links in hypermedia. In order to provide a solution, we introduce the notion of context for the source and the destination of a link. A context makes explicit which part of a presentation is affected when a link is followed from an anchor in the presentation. Given explicit source and destination contexts for a link, an author is able to state the desired presentation characterisitics for following a link, including whether the presentation currently playing should remain playing or be removed. We first give an intuitive description of contexts for links, then present a structural-based approach. We go on to describe the implementation of contexts in our hypermedia authoring system CMIFed.

Accessing multimedia information in a networked environment introduces problems that don't exist when the same information is accessed locally. These problems include: competing for network resources within and across applications, synchronizing data arrivals from various sources within an application, and supporting multiple data representations across heterogeneous hosts. Often, special-purpose algorithms can be defined to deal with these problems, but these solutions usually are restricted to the context of a single application. A more general approach is to define an adaptable infrastructure that can be used to manage resources flexibly for all currently active applications. This paper describes such an approach. We begin by introducing a general framework for partitioning control responsibilities among a number of cooperating system and application components. We then describe a specification formalism that can be used to encode an application's resource requirements, synchronization needs, and interaction control. This specification can be used to coordinate the activities of the application, the operating system(s) and a set of adaptive information objects in matching the (possibly flexible) needs of an application to the resources available in an environment at run-time. The benefits of this approach are that it allows adaptable application support with respect to system resources and that it provides a natural way to support heterogeneity in multimedia networks and multimedia data.

We present the user interface to the CMIF authoring environment for constructing and playing multimedia presentations. The CMIF authoring environment supports a rich hypermedia document model allowing structure-based composition of multimedia presentations and the specification of synchronization constraints between constituent media items. An author constructs a multimedia presentation in terms of its structure and additional synchronization constraints, from which the CMIF player derives the precise timing information for the presentation. We discuss the advantages of a structured approach to authoring multimedia, and describe the facilities in the CMIF authoring environment for supporting this approach. The authoring environment presents three main views of a multimedia presentation: the hierarchy view is used for manipulating and viewing a presentation's hierarchical structure; the channel view is used for managing logical resources and specifying and viewing precise timing constraints; and the player for playing the presentation. We present the authoring environment in terms of a short example: constructing a walking tour of Amsterdam.

In this paper we discuss the architecture and implementation of CMIFed, an editing and presentation environment for hypermedia documents. Typically such documents contain a mixture of text, images, audio, and video (and possibly other media), augmented with user interaction. CMIFed allows the author flexibility in specifying what is presented when, using multiple simultaneous output channels. Unlike systems that use a timeline or scripting metaphor to control the presentation, in CMIFed the user manipulates a collection of events and timing constraints among those events. Common timing requirements can be specified by grouping events together in a tree whose nodes indicate sequential and parallel composition. More specific timing constraints between events can be added in the form of synchronization arcs. User interaction is supported in the form of hyperlinks. We place CMIFed in the context of the Amsterdam model for hypermedia documents, which formalizes the properties of hypermedia presentations in a platform-independent manner.

In this paper, we single out one of the problem aspects of multimedia: how can one efficiently store high-quality picture information in a manner that does not make its retrieval characteristics incompatible with the needs of a (distributed) multimedia application. These needs include: timely access to data, predictable access to common (i.e., network) resources and ease in user specification of information. Our approach to solving this problem is to define adaptive data objects that adjust the amount and type of information given to an application as a function of resource availability. The key to our approach is that we transparently adapt the information presented to the application based on a set of pre-specified conditions that were defined by the application at author time. We discuss this work in the context of a parallel JPEG image decoder that provides adaptive images (with respect to data content and image representation) based on a transparent client/server negotiation scheme. Our work is based on the Amsterdam Multimedia Framework (AMF), a partitioning of control operations for supporting distributed multimedia. The parallel JPEG algorithm, AMF and the negotiated control algorithm are explained.

We present a model of hypermedia that allows the combination of "hyper-structured" information with dynamic multimedia information. The model is derived by extending the Dexter hypertext reference model and the CMIF multimedia model. The Amsterdam hypermedia model allows the following, in addition to the model provided by Dexter: · the composition of multiple dynamic media, in order to specify a collection of time-based media making up a complete multimedia presentation; · the definition of channels for specifying default presentation information, allowing the specification of the presentation characteristics of nodes at a more general level than that for an individual node; · the composition of existing presentations into larger presentations, taking into account possible clashes of resource usage; · the inclusion of temporal relations while maintaining the separation of structure and presentation information, where time-based relationships are treated as presentation information; · the definition of context for the source and destination anchors of a link in order to specify the parts of a presentation affected on following the link. The Amsterdam hypermedia model enables the description of structured multimedia documents, incorporating time at a fundamental level, and extending the hypertext notion of links to time-based media and compositions of different media. The paper is organised as follows. The Dexter hypertext model and the CMIF multimedia model are summarised, and their limitations for use as a more general hypermedia model are discussed. The extensions included in the Amsterdam hypermedia model are described and a summary of the resulting model is given.

In this paper, we single out one of the problem aspects of multimedia: how can one efficiently store high-quality picture information in a manner that does not make its retrieval characteristics incompatible with the needs of a (distributed) multimedia application. These needs include: timely access to data, predictable access to common (i.e., network) resources and ease in user specification of information. Our approach to solving this problem is to define adaptive data objects that adjust the amount and type of information given to an application as a function of resource availability. The key to our approach is that we transparently adapt the information presented to the application based on a set of pre-specified conditions that were defined by the application at author time. We discuss this work in the context of a parallel JPEG image decoder that provides adaptive images (with respect to data content and image representation) based on a transparent client/server negotiation scheme. Our work is based on the Amsterdam Multimedia Framework (AMF), a partitioning of control operations for supporting distributed multimedia. The parallel JPEG algorithm, AMF and the negotiated control algorithm are explained.

Accessing multimedia information in a networked environment introduces problems to an application designer that don't exist when the same information is fetched locally. These problems include "competing" for the allocation of network resources across applications, synchronizing data arrivals from various sources within an application, and supporting multiple data representations across heterogeneous hosts. In this paper, we present a general framework for addressing these problems that is based on the assumption that time-sensitive data can only be controlled by having the application, the operating system(s) and a set of active, intelligent information object coordinate their activities based on an explicit specification of resource, synchronization, and representation information. After presenting the general framework, we describe a document specification structure and two active system components that cooperatively provide support for synchronization and data-transformation problems in a networked multimedia environment.

Multimedia is "in". In bijna iedere computerwinkel zijn multimedia-systemen te koop en multimedia software is inmiddels verkrijgbaar op toepassingsgebieden van reparatie en onderhoud tot muziekonderwijs. Toch hebben huidige multimedia systemen geen enkel probleem op het gebied van de informatietechnologie opgelost. Wel is een aantal problemen sterker naar voren gekomen. Willen die opgelost kunnen worden, dan is een gezamenlijke inzet van onderzoekers uit diverse vakgebieden nodig. Pas dan kan multimedia de doelstelling waarmaken dat mensen op een totaal andere manier met computers kunnen gaan communiceren - en mogelijk zelfs met elkaar.

One of the most important emerging developments for improving the user/computer interface has been the addition of multimedia facilities to high-performance workstations. Although the mention of multimedia I/O often conjures up visions of moving images, talking text and electronic music, multimedia I/O is not synonymous with interface bells and whistles. Instead, multimedia should be synonymous with the synchronization of bells and whistles so that application programs can integrate data from a broad spectrum of independent sources (including those with strict timing requirements). This paper considers the role of the operating system (in general) and UNIX (in particular) in supporting multimedia synchronization. The first section reviews the requirements and characteristics that are inherent to the problem of synchronizing a number of otherwise autonomous data sets. We then consider the ability of UNIX to support decentralized data and complex data synchronization requirements. While our conclusions on the viability of UNIX for supporting generalized multimedia are not optimistic, we offer an approach to solving some of the synchronization problems of multimedia I/O without losing the benefits of a standard UNIX environment. The basis of our approach is to integrate a distributed operating system kernel as a "multimedia co-processor." This co-processor is a programmable device that can implement synchronization relationships in a manner that decouples I/O management from (user) process support. The principal benefit of this approach is that it integrates the potential of distributed I/O support with the standardization provided by a "real" UNIX kernel.

This paper considers the role of UNIX in supporting multimedia applications. In particular, we consider the ability of the UNIX operating system (in general) and the UNIX I/O system (in particular) to support the synchronization of a number of high-bandwidth data sets that must be combined to support generalized multimedia systems. The paper is divided into three main sections. The first section reviews the requirements and characteristics that are inherent to multimedia applications. The second section reviews the facilities provided by UNIX and the UNIX I/O model. The third section contrasts the needs of multimedia and the abilities of UNIX to support these needs, with special attention paid to UNIX's problem aspects. We close by sketching an approach we are studying to solve the multimedia processing problem: the use of a distributed operating system to provide a separate data and processing management layer for multimedia information.

This paper presents a document structure for describing transportable, dynamic multimedia documents. Multimedia documents consist of a set of discrete data components that are joined together in time and space to present a user (or reader) with a single coordinated whole. Transportable documents are those in which the document structure can be accessed across system environments independently of individual component input or output dependencies; dynamic documents are those in which the synchronization of document components are not staticly defined as an integral part of the data definition but are dynamicly defined as attributes of the general document structure. The focus of this paper is the presentation of the basic building blocks of the CWI Multimedia Interchange Format (CMIF). CMIF is used to describe the temporal and structural relationships that exist in multimedia documents. In order to put our work in a concrete context, we start our discussion with a brief description of the portability requirements for documents used within the CWI /Multimedia Pipeline. We then provide a layered description of our document structure format; this format provides a means for expressing a document in terms of synchronization channels, event descriptors, data descriptors, data blocks and synchronization arcs, each element of which contains a set of appropriate descriptive attributes. The paper describes each of these concepts abstractly as well as in the context of a uniform example. The paper concludes with a discussion of our intended future direction in using the various attribute descriptors to control a broad range of activities within the CWI /Multimedia Pipeline.