An Interactive Video System for Learning and Knowledge Management Michael Langbauer
Franz Lehner
Chair of Information Systems II University of Passau Passau, Germany
[email protected]
Chair of Information Systems II University of Passau Passau, Germany Franz.
[email protected]
Abstract— Interactive videos have high potential in a wide range of learning scenarios in both private and professional life. Due to their unique combination of characteristics they can be used to create situational, exploratory and personalized learning experiences. As we are currently investigating the applicability, costs and benefits of interactive videos, we learned that interactive videos are a useful instrument especially in several areas of knowledge management, E-Business and E-Learning as well as entertainment. This article aims at providing a comprehensive understanding for the context, functionality, technical implementation and suitable application scenarios of interactive video as well as provide experience gathered in several application projects in practice. Keywords—interactive videos; hypervideo; knowledge transfer; learing experience
e-learning;
I. INTRODUCTION A. Growing importance of videos Traditional forms of media like newspaper and TV are still dominating as sources of information but new media types are gaining ground and are changing the rules of the game. Video platforms and other new types of media have gained much attention recently – both from private persons [1] and with some delay from enterprises of all types and sizes [2–4]. As we focus on videos, this chapter will briefly motivate the importance of videos as a medium and on video platforms to illustrate the scope of the potentials of (interactive) videos we seek to address in this article. The video platform market is contested intensively and therefore is still in constant state of flux. YouTube as the flagship of said services for instance has seen spectacular growth rates in viewership and user generated content since its foundation in 2005. This success story [4, 5] also states the market potential as Google’s investment of 1.65 billion USD to acquire YouTube. But the service providers are not the only stakeholders profiting from new possibilities. With web-based platforms companies posting content can benefit from viral marketing effects which descend from users’ positive word-of-mouth advertisement. For these and more efforts to involve customers into the internal processes via social media platforms [3] companies can achieve major popularity gains and indirectly generate high
profit by using such new channels that try to attract new content providers with large user pools that can be reached. And the gateway to these channels is literally very open as well. Companies can invest by uploading and disinvest by their liking. This way, companies can harness all sources of information at-hand – from their own employees [6] as well as form their customers [3]. More video platforms have appeared since and forced traditional media companies to react. Newspapers, TV Channels and completely new players use videos to amplify their online appearance. Some even establish their own online channels (IPTV). This shift in power – the user in control, the provider relegated to permanent competition – also benefits the consumer as he is liberated from traditional broadcasting technology, can choose from a wider range of suppliers and can make further use of his inherent creativity (User Generated Content). [5] This synopsis of the significance of the medium video states that it’s prime value lies currently in entertainment and marketing. The concept of predefined and fixed storylines in movies and commercial works fine for entertainment and marketing purposes, yet it has prominent flaws when considering a learning environment. Taking another close look on YouTube, there is substantial demand for of tutorial videos to be recognized and also met by mostly semi-professional videos with little complexity. Albeit we find hidden complexity, low budget production and ease of use for both professionals and laymen to be desirable features for any media product with learning purposes, ideally the author of a learning video should also be enabled to cover complex topics nonetheless. Whether it is tying a specific knot into a tie or operating heavy machinery correctly in a production facility, video has the potential to illustrate motoric, sequential or nonverbal skills better than other types of media (e.g. texts or also motionless pictures) [7–10]. Regardless whether in form of continuous motion pictures as in films or computer-based graphic animations [8], video has high potential for e-learning and knowledge transfer both in private and professional life [11]. Still, with traditional video (e.g. movies) the viewer remains to be a passive observer. In order to conquer e-learning and knowledge management as well as to expand its use in the
aforementioned areas of entertainment and marketing, videos need a higher level of responsiveness to prior knowledge and situational interests of the individual by adapting content, scope and succession accordingly. B. Interactivity in videos As soon as the user should be allowed to explore learning material in his own pace and scope while still being guided by the content creator, traditional video is no longer sufficient on its own. This is where interactive videos excel in comparison to their traditional counterparts [6, 12]. Analogue to other forms of hypermedia, the video-based interactive videos can be enriched, or be “annotated”, with other media types (i.e. texts, pictures, audios, and videos), but the basis for the structure of an interactive video is video snippets or animations. By adding two more unique characteristics to the dynamic content (1st characteristic), interactive videos excel for learning purposes in comparison to their traditional counterparts [7–10]. “Interactive video”, or also called “hypervideo”, is the transference of the organizational concept of hypertexts like wikis to an audio-visual medium like video. [13] As all forms of hypermedia result in a non-linear structure (2nd characteristic) of a certain type of media, interactive videos are an arbitrarily complex construct of video scenes. Figure 1 shows a variety of storylines that can be modelled within the non-linear structure of an interactive video. Already during the creation of the storyline, the author decides where to branch or bound and still keep is core information intact. E.g. after the core information has been delivered, a non-linear structure offers the chance to let the user decide whether to proceed with additional information or a new subject. The resulting structure allows the user to navigate relatively free within defined boundaries [12, 13].
If the course of a video is not regulated by a fixed sequence of scenes, the viewer is needed to interact. Interaction (3rd characteristic) derives from Latin inter (between) and “agere” (to act) and means mutual influence and interdependence. Accordingly, interaction in controlling the flow in a video does not necessarily require active choice, e.g. clicking buttons in a selection panel, but can be made automatically on the basis of previous behavior, i.e. log data. Interactive control functions can also be dynamic if a temporal dimension is involved and the time to react to an offered anchor is limited [13]. This article aims at providing a comprehensive understanding for the context, functionality, technical implementation and application scenarios of interactive videos. The remainder is structured as follows: after having covered the unique characteristics of interactive videos, we will provide a detailed description of research in this field and the requirements and technical specifics for a prototype as our object of interest. Subsequently we will discuss practical application scenarios for interactive videos and close with outlining past and future research in our project. II. RESEARCH ON INTERACTIVE VIDEOS Interactive videos have been used as a research object in various research and development projects that addressed different aspects and reflect different approaches to the topic of interactive videos. In a semiformal collection process of related research projects, we found plenty of institutions active in this field. In this chapter we seek to give a synoptic view of the research projects, systems and technologies encircling the research field of interactive videos. We start by describing a selection of research projects in the German IS community to provide a representative overview with research agendas and where appropriate the utilized interactive video system. In addition, this also allows demarcating our own research projects. The collection below has no claim to completeness. It serves as an overview and gives hints at concepts shared with other projects and approaches to the topic. A broad clustering for approaching paths might be the division into technology and usability, cognition, learning and application in general, and outcome. At the university of Lübeck [14] researchers developed an interactive video system to foster joined-up thinking as a part of a bigger research project focuses on ambient learning with focus on different technologies and devices. The resulting hypermedia platform entails the loosely coupled hypervideo system HyperVid. This web-based prototype illustrates that a single environment for creating, editing and management of interactive videos is feasible. Furthermore, the researchers in this project stress that this prototype allows for collaborative and location independent content creation which we deem a necessity for creative work in the era of Web 2.0.
Fig. 1. Non-linear structures
The university of Jena [15] developed OSOTIS, or relabelled Yovisto, a search engine with tagging and annotation functionality for video material of different format. By address-
ing not only whole scenes but also fragments of scenes. Selectable and linkable anchors can be added to a scene both automatically by its inherent descriptors (start and end) but also manually and even collaboratively by annotations. This shows an interesting concept where annotations can be made recognizable by search engines and allow for direct linking to discrete time stamps within in a scene. This feature, as well implemented into Google’s YouTube, opens possibilities to an even more flexible use of given set of scenes. This stresses and in a way adds to the aforementioned dynamicity of the medium video. A similar but differently focused approach was followed to develop Advene [16, 17]. With this dedicated interactive video system viewers are asked specifically to contribute via comments. An outstanding role in the research field of interactive videos is occupied by learning situations. Chambel, Zahn & Finke [9] developed an environment for individual and collaborative learning where every single user can share his knowledge whenever he chooses. Braun & Finke [18] developed an interactive video system that uses an avatar and a set of different annotation types to create interactive video resembling a conversation, thus investigating a computer-based learning situation very close to person-to-person contact. MOVieGoer an example for a concrete interactive video in a learning environment and was produced with Riva VX. It combines videos and texts were about the subject „Eco system lake“. This and other systems were introduced and evaluated at the University of Münster. MOVieGoer for instance was assessed for varied numbers and positions of the embedded hyperlinks for the impact on learning outcome (comp. [19], 359362). To address other aspects in the application of interactive videos, many more systems have been developed in Academia and practice. Hyper-Hitchcock allows for the user to call up additional external information and the usage of existing links between scenes. Shipman, Girgensohn & Wilkox [20, 21] call this limited utilization of a non-linear structure Detail-onDemand. In contrast, users of Advene can enrich video content by commenting it [16, 17]. In addition to research projects plenty specialised tools have been developed, which with the help of a few mouse clicks interactive elements as well as textual comments can be embedded in available videos at a minor scale (Hotspots, Links, Buttons). Through easy operations, these programs target especially unschooled users which can quickly reach presentable results despite with little prior knowledge. Programs such as Riva Producer and Videoclix belong to this product category, but also rather community centred online-video tools like Jumpcut, Viddix or the annotation functions of YouTube. The tool VideoClix enables the user to click upon objects within a video, in order to get contextual, multimedial additional information. This tool is especially designed for the marketing area. Integrated functions offer detailed metrics and vital marketing data of the users’ behaviour. VideoClix represents a proprietary tool, whose future security and extensibility cannot be verified. The video platform YouTube also provides a tool which enables the users to enrich their videos to a limited ex-
tend with music, text and images. However, this tool can only be used online and provides no open interfaces. More information to these and further authoring tools are to be found in Lehner et al. [22]. Technologic questions deal with the development of efficient codecs and components or entire frameworks for the creation and distribution of interactive videos. Many of the technologies necessary for the development of the planned tool already exist. Especially in the internet rich media applications play an increasing role and facilitate, among other things, the creation of interactive videos. Traditional authoring tools are equipped with high performance functions for the processing of videos, whereby mostly appealing to media developers with advanced programming skills. Most existing technologies and projects known until 2008 have in common that they did not support alternative plot lines in contrast to the planned authoring tool. Commercial providers do not pay much attention to extensibility or reuse based on open interfaces. And tools available are often designed for a special group of users and therefore restricted to them. Due to the complexity of the tools laymen would not be able to produce or create interactive videos for the vast number of possible applications. Since then, the topic of interactive videos has seen reasonable attention and resulted in more content design-oriented articles. However, the possible impact and outcome interactive videos might have on technology-based knowledge management and learning has hardly been investigated properly. Consequently we are currently conducting an evaluation and adaptation of said prototype in practice in various case studies. III. GENERAL AND PROCEDURAL REQUIREMENTS Depending on the respective manifestations of the main characteristics of interactive video, i.e. different forms of interactivity, dynamicity of content and non-linearity, interactive videos can take numerous shapes in order to serve different purposes. The possible manifestations range from designs for passive content reception with VCR/ DVD controls to systems with bidirectional content editing at runtime [23]. Existing prototypes emphasize the breadth and continuity of possibilities offered by the interactive video concept, especially in a knowledge transfer context. All these mentioned prototypes in the previous chapter have in common that they consist of subsystems or modules for the creation and the use of interactive videos. An interactive video system ideally is modularized yet integrates both the “producer” module for authors to assemble interactive video applications and the “player” module for the audience to use them. With an additional server application, provision and distribution of interactive videos can take place online. However, further details depend on the particular system and can hardly be generalized reasonably. This chapter provides detailed information on a prototype we have developed that carries its main goals already in its name – the SIVA Suite was designed as a comprehensive toolset for Simple Interactive Video Authoring. In a long-term series of research projects, we have developed a prototype for the creation, distribution and use of interactive videos. For depicting the architectonic and functional principles of our
prototype in the following, we focus on possible application its individual aspects facilitate.
one video file. This concept can be transferred to other media types.
In a current research project we develop and evaluate a prototype that should cover a wide range of interactivity levels, various complex non-linear structures and dynamicity in an interactive video depending on the scenario. The project goal is not only to develop a prototype but to identify and validate application scenarios too. This intention involves several disciplines, e.g. information systems, media science, information science and also law. Goal of the aforementioned research project which stands to be the trigger for this article is
Step (4) – Configuring the scene graph: A basic concept of toolset is to display the so-called scene graph. This gives the possibility to connect the loosely existing scenes from the repository and in this way to establish a graph of scenes. A scene can be connected to one or more scenes, which next to linear (classical videos) also makes alternative scene sequences possible. Through user interaction, applying branch and bound operations to create a dynamic tread of action is realized.
Through the modular architecture (see. chapter IV) of the SIVA-producer as of the player, the functions radius can be changed at all times and thus can be adapted for special requirements. The authors have hardly any limits as to the functional aspects. Consequently, the content of interactive videos can be constructed by the user (for example through the form of any arrangement of scenes and sequences) and also manipulated (for example through forks or leaps in the video).
Step (5) – Annotating scenes: The user has different functions at his disposal to enrich the video with interactivity. A scene can be annotated in different points and areas in a sequence of time. The possibilities range from simple variations such as texts and diagrams known from TV sport broadcasting or movie subtitles to complex annotations such as composed Rich-Media-contents (Websites, Videos, etc.) or widely spread business document formats like PDF or PowerPoint. The author has different possibilities to call the attention of the user to the existing extras. This can be realized, for example, through the superimposing of buttons in the video clip. Through the same mechanism menus can be realized for branched out action sequences or table of contents. Step (6) – Saving projects. Ongoing operations can be saved in a specific format, so that the projects can be picked-up at any time. This steps also signifies the end of a creation session in Step (7) – Exporting projects and saving of interactive video data. The (temporary) termination of creation process of an interactive videos constitutes the export into an application consisting of all media files and the meta-data in an XMLformat to enable interpretation by an HTML5 player. As mentioned before, the export does not prevent the author(s) to revise his work at will and export multiple versions from on interactive video project, e.g. to address multiple target audiences with different content requirements and restrictions. Depending on the openness of the production process, user generated content or collaborative creation is also realizable within this process.
Fig. 2. Workflow of the production of interactive videos
As shown in figure 2, a typical production process of an interactive video begins with (1) starting a new or opening an existing interactive video project within from a repository. Th illustration and the remaining steps explained below demonstrate a main objective for our software which is simplicity that allows for usage by laymen. It also shows, that interactive video projects can be revisited at any time which provides the author with high flexibility with producing and editing content. Step (2) – Importing media objects: The basis of the content of an interactive video is formed through individual files, e.g. videos, pictures, texts. Especially bigger projects need an extensive collection of multimedia objects that are organized in repositories to facilitate control and flexible access. Step (3) – Editing media objects: The cutting of imported videos objects for example is a very useful function for interactive video production in order to define multiple scenes from
IV. TECHNICAL IMPLEMENTATION AND PROTOTYPE FUNCTIONALITY Figure 3 represents the natural trisection of a software environment in regard to its basic purposes, i.e. creating, distributing and consuming interactive videos. In addition, the three resulting roles have separate yet integrated software modules for their individual tasks. Authors produce videos in the producer module with its various editing tools and repositories. Operators manage finished videos, users and user data on the server module. Users consume available interactive videos with the player module. These distinctive roles are however combinable and facilitate a manifold of applications scenarios that are presented in the subsequent chapter.
fied users by limiting access to certain paths and nodes of the video. To realize such scenarios, a conditional fork limits the available paths and only denies access to those with unfulfilled requirements, e.g. prior knowledge obtained in another scene or a certain number of points in a quiz. Another kind of fork uses randomized path choosing to provide dynamicity of content succession. Another navigational element available for the scene graph, technically not forking a path, pauses the video feed and affords an active decision by the user to resume with the consecutive scene. Fig. 3. prototype architecture
A. The producer module No aspect of the creation of new interactive videos with our prototype affords a single line of code by the author. Within the producer module and its graphical user interface, new and existing interactive videos are handled as projects. The center of each interactive video is the scene graph as the graphical representation of the interactive video itself. In a new project, the scene graph starts with both a defined start and end node. Between these unique nodes the author(s) of the interactive video are free to place more nodes, i.e. video scenes or special interactive objects, e.g. quizzes, and link them with edges. As a result, the scene graph allows creating alternative storylines in which the user chooses with which part of the content to proceed by providing hotspots or buttons to navigate.
The video scenes as the central content objects that make up the video structure are stored in a repository and can be placed in the graph via drag and drop. All scenes base on a media file in several on of the supported video formats like MPEG. With the built-in video editor, video files from the media repository can be cut and partially or wholly defined as scenes – while the original media file remains unchanged. To support aforementioned multi modal annotation of additional information to the primary video, the media repository contains not only videos but also texts, pictures, animations, audios and complex media formats like PDFs. This allows companies to reuse existing material in the interactive video applications. In order to display any further information in addition to the primary video, the author can edit and annotate files of the various formats from the media repository to each scene individually. To ensure the integrity of the intended message in each scene, the author can decide on the positioning and scheduling of each annotation individually. The prototype provides both a function for free positioning the object as an overlay to the main video and into the collapsible annotation area. For interests like corporate identity and branding, annotations can also be defined and positioned as global which relieves the author from annotating objects redundantly to each node throughout the entire structure.
Fig. 4. user interface producer module showing a scene graph
The scene graph as shown in figure 4 comprises special navigational objects to implement forking and looping. As nonrule-based splits demand user based decision, this is the first aspect where interactivity is facilitated as a core concept of both interactive videos and this prototype in specific. By placing decision forks, conditional forks, or quizzes authors l allows for individualized user experiences in accordance to the application’s function. By providing free choice of how to succeed from a certain point in the video structure, the author subordinates his authority in deciding on content order to the situational needs and desires of the users. A user centric approach however should not require sacrificing the video’s informational integrity, but augment the user experience with individuality and spontaneity. Reacting to the informational need of a single user might also mean that the author withholds information from unquali-
Furthermore, authors can create a multilayered table of contents (TOC) that is detached from the scene graphs structure. E.g. in learning scenarios the TOC can provide an overview of lessons by only linking a first scene of a sequence of integrated content, thus hiding the extent of the interactive video if need be. Another feature allows tagging each scene and the annotated media objects to create another easy access method to jump between different segments of the video. Before releasing a version of the interactive video project, a graph validation mechanism ensures compliance with the constraints of a directed graph with a single start and a single end node. The export function results in single file repository comprising all media and meta data ready for distribution. As mentioned previously, the main videos are never altered. A released interactive video contains the video footage separately from the additional media objects and the meta information in XML-format about how the video is structured and all the media objects are interconnected [24]. B. The server module The server module has a web interface and is the main working environment for the operator of interactive videos as
it provides functionality to distribute videos to its intended audience by limiting access to videos individually for each user. The user management section includes all basic functionality from suspending users, contacting them via E-Mail and changing their master data. In the video management section the operator handles versions of an interactive video project and provides multiple channels for access to specific users. Figure 5 shows the web interface to these sections for both roles of operator and users. Other than the operators, logged in users can only access their own master data and videos made available to them. With a built-in web player they can also watch without downloading any data.
Fig. 7. user interface player module
Fig. 5. user interface server module
Is an annotation not positioned as an overlay, the user can find it in a designated annotation area. It is collapsible so the user can choose if and when he wishes to receive links to additional information. Supported media types for user annotations are videos, pictures, texts or complex media types like PDFs. As illustrated in figure 7, annotations can be used as nonclickable ribbons that fade in and out at the beginning of a scene, e.g. to make the scene title more prominent. In this example, the annotation area contains both a video scene and picture gallery that both expand and pause the main video when clicked on. Other than the title ribbon, they may contain complex additional information, that demand the users full attention.
In another section, operators exclusively can evaluate usage data for each interactive video they manage. For evaluation purposes we designed a few standard reports that serve different interests. For example figure 6 shows a so-called sunburst graph to illustrate the most common paths through an interactive video. If needed, the operator can also define new reports based on SQL due to the integration of the log database.
Fig. 6. usage statistics
C. The player module The user interface of the player is split up into two navigational bars and two content areas. The main video area contains the momentarily playing scene and may be overlayed with (time triggered) annotations.
The micro navigation bar contains standard VCR functionality and a timeline for precise navigation within a scene. Additionally, markers show the starting point of time triggered annotations. Next to the timeline users find sound control and other settings, e.g. language. The macro navigation bar primarily serves the purpose of providing different navigational functions for changing between different scenes and even annotations within the interactive video. With our prototype we try to give as much control to the viewer by providing suitable tools and limit these possibilities to the intent of the content’s author by making them available and configuring them in the producer module. For example, the table of content is a multi-layered structure of the video that is not bound to the scene graph but is configured and linked to content objects by the author separately. As a result the user is ideally provided with a quick overview of the scope of the interactive video and with elegant access points to other parts of his current path or completely different sequences within the interactive video. Forward and backward skip buttons next to the scene title allow jumping through the current path through scenes according to the scene graph. A search function compares the search term to the tags of the various content objects, i.e. both scenes and annotations. Lastly, there are non-navigational buttons to be found that are positioned here in order to not disturb the appearance of the player as a whole. These additional functions triggered here are the toggle to full screen and a link to a user specific usage calendar.
these instructions can even be responsive to prior knowledge and interests of the individual user without producing and deploying multiple versions for minor changes in content, scope or succession. B. Scenario type 2: Marketing and E-Commerce Google’s YouTube showcased how videos can be used for successful viral marketing campaigns mostly without severe financial investment. Viral marketing is the idea of solidifying positive attitudes towards a product in a person and using his positive word-of-mouth to reach out to other people in a snowball effect. Fig. 8. Interaction concept of the player module
Figure 8 depicts the interaction concept for the player module and illustrates the high density of clickable elements within the user interface. Most prominent are the triggers for various interactional and navigational possibilities in the top and bottom bar as mentioned before. The figure also displays the collapsible area for annotations and different types of annotations that can be set either responsive to clicks or non-clickable by the author. With the permanently visible interface objects outlined above, there are some event triggered objects left for explanation. For example, whenever the video reaches a branching point in the scene graph that needs a decision by the user of how to succeed, an additional side bar with the granted options within the structure of the interactive video appears. Furthermore, when paused or reaching the end of a scene, a prominent (re)play button is superimposed. V. APPLICATION OF INTERACTIVE VIDEOS IN BUSINESS Existing prototypes emphasize the continuity of possibilities offered by the interactive video concept especially in a knowledge transfer context. As stated before, the different manifestation of the three core characteristics of interactivity, non-linearity and contentual dynamicity facilitate far more diverse shapes and purposes of interactive videos than those described above. Thereby different yet combinable forms of interactive videos can be created to address the needs of the three primary application fields, i.e. Knowledge Management and E-Learning as well as E-Commerce and entertainment [24]. A. Scenario type 1: Training and E-Learning A more and more obvious exploitation angle for an audiovisual medium like (interactive) video lies in knowledge acquisition and sharing as well as e-learning [25]. Especially for professional and less academic subjects, video material fits best to display complex motoric procedures [10], e.g. assembling or operating a machine when put into an industrial context. The same idea can be transferred to households as a private person may be in need of a more seamless visual orientation compared to standard print instructions when assembling furniture. With the unique structural superiority of interactive videos provided by the unique combination of its core characteristics
As an E-Commerce application, interactive videos can directly link promotional video footage with products in e-shops so that companies might open new channels to the customers and augment the effectiveness of their marketing [13]. C. Scenario type 3: User Generated Content and Social Media Lastly, there is high (economic) potential to be expected from interactive videos when considering user-driven content creation (user generated content). By tapping into a large amount of creative minds, new interactive videos for various purposes can be rapidly conceptualized and realized. Providing users with the necessary tools to (co)produce by themselves, the success story of YouTube tells us to expect both high demand and supply of interactive videos in the private sector. Taking entertainment purposes into consideration for the first time, interactive videos can take even more diverse forms than previously described, e.g. vide-based games or interactive movies. However, this potential of user generated content also holds for the aforementioned companies (knowledge transfer and EBusiness scenarios). Altogether the scenario types here should not be understood as distinct categories, but as overlapping, combinable focus points to show the broad applicability of interactive videos that we are exploring currently in our research project. VI. SHOWCASES OF PRACTICAL APPLICATION SCNENARIOS AND PRELIMINARY RESULTS As mentioned, the prototype described in detail before is the research object of a current research project and had been developed in predecessing projects. As these efforts created proof of the technical feasibility to prototypical implementation of an interactive video system, we now aim at validating the practical exploitability of interactive videos and therefore creates by testing the applicability in concrete application scenarios. So far, we gained detailed experience concerning scenario type 1 (Training and E-Learning) by two partially concluded field studies. The second case also provides insights into type 3 (User Generated Content and Social Media).
A. Interactive video as an instructional treatment measure for long term physical therapy In a first practical application, we partnered with a rehabilitation clinic that specializes on physical therapy for cancer patients. The partnership’s goal was to establish an interactive video that helps prostate cancer patients with their pelvic floor gymnastics to regain bladder control. The treatment plan for this condition states that actual bladder control can only be expected to be reached by executing appropriate exercises on a daily basis for six to 24 months. In order to learn these exercises, patients absolve three weeks of stationary rehabilitation at a clinic a few days after surgery. After this stay they return home and have to continue their training routine without professional help before and after returning to the clinic only once more. The interactive video is intended to fill the informational void between two stationary rehabilitations was conceptualized to mimic actual training sessions from the clinic. As such, the patient was provided with training programs consisting of five exercises that last for about 20 minutes in total, which is the advised duration of training for a single day. In order not to distract the patient unnecessarily he can choose to include a detailed explanation scene of every exercise before the actual instructional scene where the patient is supposed to execute while watching. Before and in between exercise sessions, the viewer can decide on which exercise to perform next, thus creating a situational and individual training experience. For a better understanding of their own physical condition, the patient is additionally provided with a few scenes that explain the anatomical effects of prostate removal for example. Figures 7 and 8 give an impression of the graphical user interface we presented the patients of the clinic.
that allows the user to configure the exercises he wants to perform individually for each training session. In the surveys, the patients stated high levels of satisfaction and appreciation for this extra measure of treatment. A majority of the patients stated that the interactive video in this configuration is providing the intended optical and audial orientation during training sessions as well as being a reliable pool of knowledge for coping with their physical condition. As such, it complements the treatment of the clinic as intended. Applying the content structuring qualitative content analysis in accordance to Mayring [27] ten face-to-face interviews with patients after one week of usage, we identified the following three main and twelve sub categories that explain how interactive videos benefits the individual1: Motivation and Attitude (M): This main category describes what persuades a potential user of an interactive video application towards usage and represents his anticipation of benefits for his task, (here: the pelvic floor exercises). So far, we identified the individual’s intrinsic motivation (M-1), his attitude towards the supported task (M-2), towards the used technology (M-3) and towards the operator of the interactive video (M-4). These dimensions of a person’s attitude determine his openness to try the interactive video which is imperative to actually experience benefits. In addition, we found that there are certain circumstances that impede using the interactive video regardless of personal attitude and cannot be controlled by the user (M-5). Applicability (A): The second main category comprises all benefit influencing experiences that emanate from first-hand experience. Therein, we distinguish the individual perception of how well the interactive supports the instructed task (“task fit”, A-1), the general perceived ease of use (A-2) with a special interest in the interplay between executing the task while operating the interactive video, the actual usage behavior in intensity, form and regularity (A-3), and the organizational support by the operator of the interactive video (e.g. training, A-4).
After introducing the interactive video to the clinic processes, we validated its effect for several weeks by interviews, surveys performing a two-ways comparison (similar to [26]) at the end and six months after stationary treatment in the clinic, and automatically logging of usage behavior for one year. The results were coherent and showed great promise in E-Health scenarios and physiotherapy in specific and for instruction of motoric procedures in general. The latter measurement instrument, i.e. logging of usage behavior, was realized by the asynchronously and automatically communicating player and server component. With the builtin analysis tools we could see different phases in the usage of our interactive video and in numerous cases we witnessed long-term usage of the content in the desired way: the background information (anatomy, exercise management) were mostly consumed in the early stages of the treatment and stopped after a couple of sittings only to be resumed occasionally. The usage of the training programs was more consistent over time and only stopped at later stages in the recovery process. As reasons for discontinuing the training with the support of the interactive video we consider treatment success (i.e. regaining bladder control), individual patients neglecting the pelvic floor exercises as a whole, or boredom with the content of the video. Only the latter can be accounted as critical for the application scenario, yet was found to be natural considering the long term viewership despite the limited content in this early stage. To avoid boredom in such scenarios, the we could update the content from time to time or we can add a function
Individual Benefit (B): The dimensions of tangible and intangible benefit of using interactive videos are to be found in the last main category. From the first application scenario we derived learning success (B-1), performance improvements (i.e. here faster recovering patients, B-2) and satisfaction (B-3). As intended, the patients perceived the intended message of the interactive video, i.e. why to perform the exercises (background knowledge), how to cope with the training and how to execute concrete exercises. With these insights from the first application scenarios, we gained a solid understanding of how individual benefits, emerge and can be realized in this unique application scenario.
1
Further details in method used for analysis and further details on the results can be found in upcoming publications that are currently still in review process.
Therefore, the results remain preliminary and still mark a necessary step and can be built upon. B. Interacive videos in inter- and intracompany E-Learning of an industrial company In an ongoing partnership, we integrate interactive videos into an existing learning environment of a sensor producing company. In this learning system, new and experienced, internal and external sales personnel is being trained in the specifics of new and existing product lines as well as the physical and technical background to them. In an initial phase we focus on explaining the user interface of new sensor series that accurately measures distances as the interface itself is a prime innovation to this specific sensor series which the sales personnel has to be familiarized with. Within the user interface, the user calibrates the sensor and can also inspect the measured data from the sensor. This way the interactive video can serve a dual purpose. Firstly, it explains all the settings available in the user interfaces. Secondly, it helps finding problems with the sensor by explaining error patterns in graphs provided by the user interface and generated from the feedback data of the sensor. In this scenario, we also wish to exploit the collaboration functionality of our prototype as we seek to establish a platform where all internal and external personnel and maybe even customers can share experiences, discuss problems and questions and contribute to the comment by adding additional information independent of the desired file format. Especially the latter might help to decentralize and accelerate the production and editing process and further more improve the content quality by including more informational sources, especially field experience. The potential of this feature will be of special interest in the upcoming evaluations. In comparison to our first application scenarios, this shows how much the intention of interactive videos can change from a high quality audiovisual preparation of content to collaboration without neglecting the other. As of now, we conducted initial interviews and surveys to scout for further application of interactive videos within this firm and also gained direct feedback from the target audience. Over all, especially the external field workers that sell sensors of our partner company as an agency see high potential in this medium to learn, share and discuss their daily business with their peers and experts for the products. VII. SUMMARY AND RESEARCH OUTLOOK Albeit all we know about interactive videos, there is so much more to discover. In accordance with our interdisciplinary research project, we would encourage at least the following four disciplines to investigate interactive videos from their perspective. In our research project, media scientists have primarily investigated cognitive effects of design measures regarding all three modules of the interactive video environment. In order to probe didactic possibilities and boundaries, various forms of interaction, navigation and presentation are being compared for the described application scenarios. Yet, given the limitations of the application scenario in physiotherapy in particular, we
have yet to investigate the producer and server module, as the clinic neglected to use those. Furthermore, established theories concerning creation and conception of media should be validated for interactive videos and might have to be adjusted accordingly. With the unique combination of characteristics that define interactive videos, interactive storytelling needs to be adapted for business purposes. From a computer science point of view, the described prototype and the other interactive video systems in existence are proof of the technical feasibility and leave nothing more to investigate at first glance. But new technologies, functional extension and new interfaces constantly provide new angles to further improve interactive video systems. Taking future research in our project as an example, we will partly integrate the producer module into a web platform. This way, users get the chance to link online content from various web sources. By this extension we hope to accelerate the production time which we found to be the biggest cost driver. Furthermore, the player module will be extended with a second screen concept to make use of smart TVs and similar web enabling technologies, e.g. Google Chromecast, to stress the entertainment potential of interactive videos by not only mobilizing them, but also port them to the living room. The player module will also be enhanced with a collaboration function for users among themselves and with operators. This is resembles our initial approach to foster collaborative content creation in our second application scenario. Legal sciences attend very contemporary topics that are also to be considered when dealing with interactive videos: copyrights and data security. Up until now, we gathered valuable practical experience in implementing interactive videos according to German law, even in the highly sensitive area of medical treatment. Beyond our first-hand experience, there is more ground to be covered when taking user generated content, open annotations and other feasible extensions of the interactive video concept into consideration. As for information systems research, the new forms of media like interactive videos same as wikis make full use of Web 2.0 concepts [3, 28] and distinguish themselves from traditional collaboration systems by being cheaper and mainly selforganized [29]. Companies apply various types of those new forms of media like social networks, wikis or weblogs [29] as instruments for internal and external marketing [30] and customer service purposes [3], for the organizational knowledge management [31] or to support inter and intra corporate collaboration and information exchange [30]. This means both opportunities and hurdles for the utilization of new media in companies, albeit there is plenty of promising application scenarios. The obstacles that are to be expected are a result of the state of isolation of social media from other enterprise systems and also of the shifting role of the consumer [31]. Due to the dissolution of the distinction between producer and consumer in social media, the user becomes a “prosumer”. And, in contrast to traditional collaboration or communication systems like CMS, participation in social media communities is hardly mandatory but voluntary. Hence the operating company of a social media system becomes reliant on its employee’s creativity and has to establish
interfaces to legacy systems where needed and provide incentives that foster participation, although the new set of rules demands to aim for the expectedly specific factors of motivation. These and further hurdles need to be investigated thoroughly for companies to make good use of the new media. Yet with these obstacles overcome, companies will also realize the many subtle or very tactile opportunities social media offers. First and foremost a company can benefit from the creativity of the users with relatively little management effort as soon as it reaches critical mass [31]. But the actual benefits and costs in an economic sense have yet to be made measureable and tangible. For this purpose, a measurement instrument that is as adaptable as our prototype must be designed and tested. To do so, information systems research with its long tradition of success and benefit measurement provides a rich body of literature and the necessary instruments to reach this goal. In contrast to existing models as artefacts from predominantly the American IS tradition, we seek also to compare beneficial aspects with monetary and non-monetary cost factors. In this article, we described concrete benefit dimensions and cost factors that we identified empirically in the first application scenario (instructional video in physical rehabilitation). In the current second application scenario (learning and discussion platform for a sensor producer) we will contrast these findings to generalize them. In the aforementioned first application scenario, the clinic refused to operate the server module as well as the producer module by itself. This signifies reservations towards using the entire software suite as a hurdle to be taken seriously. This is also a gap in our perspective on the benefits and costs of using our interactive video prototype entirely which includes actually producing and editing content with the company’s own resources. We will address this with the second and other upcoming application scenarios. Interpreted as an opportunity to complement the software environment with services, e.g. creating interactive videos as a service or operating a server, even more directions to investigate interactive videos become obvious. This is why we expect vast possibilities not only to apply interactive videos but also exploit economically. We will follow this idea by designing business models surrounding interactive videos.
[5]
A. M. Kaplan and M. Haenlein, “Users of the world, unite! The challenges and opportunities of Social Media,” Business Horizons, vol. 53, no. 1, pp. 59–68, 2010.
[6]
A. Stocker, A. Richter, P. Hoefler, and K. Tochtermann, “Exploring Appropriation of Enterprise Wikis,” Comput Supported Coop Work, vol. 21, no. 2-3, pp. 317–356, 2012.
[7]
D. Hutchison, T. Kanade, J. Kittler, J.M. Kleinberg, F. Mattern, J.C. Mitchell, M. Naor, O. Nierstrasz, C. Pandu Rangan, B. Steffen, M. Sudan, D. Terzopoulos, D. Tygar, M.Y. Vardi, G. Weikum, H. Koenitz, T.I. Sezen, G. Ferri, M. Haahr, D. Sezen, and G. C̨atak, Eds, Interactive Storytelling. Cham: Springer International Publishing, 2013.
[8]
O.-C. Park and R. Hopkins, “Instructional conditions for using dynamic visual displays: a review,” Instr Sci, vol. 21, no. 6, pp. 427–449, 1993.
[9]
T. Chambel, C. Zahn, and M. Finke, “Hypervideo and Cognition,” in Cognitively Informed Systems, E. Alkhalifa, Ed.: IGI Global, 2006, pp. 26–49.
[10]
T. N. Höffler and D. Leutner, “Instructional animation versus static pictures: A meta-analysis,” Learning and Instruction, vol. 17, no. 6, pp. 722–738, 2007.
[11]
R. G. Saadé, F. Nebebe, and W. Tan, “Viability of the "Technology Acceptance Model" in Multimedia Learning Environments - A Comparative Study,” Interdisciplinary Journal of Knowledge and Learning Objects, no. 3, pp. 175–184, 2007.
[12]
W. Müller, U. Spierling, and C. Stockhausen, “Production and Delivery of Interactive Narratives Based on Video Snippets,” in ICIDS '13: Proceedings of the Sixth International Conference on Interactive Digital Storytelling, Berlin Heidelberg: Springer, 2013, pp. 71–82.
[13]
C. Zahn, Wissenskommunikation mit Hypervideos: Untersuchungen zum Design nichtlinearer Informationsstrukturen für audiovisuelle Medien. Münster, New York, München, Berlin: Waxmann, 2003.
[14]
M. Ide and T. Winkler, “Hypervideo. Neue ästethische Prozesse im Web 2.0,” in Dieter-Baacke-Preis - Handbuch, vol. 7, Chancen digitaler Medien für Kinder und Jugendliche: Medienpädagogische Konzep-
REFERENCES [1]
the Success of Social Software in Corporate Environments,” in
[2] [3]
Medienprojekte, J. Lauffer, Ed, München: Kopaed, 2012, pp. 71–77. [15]
J. W. Harald Sack, “OSOTIS - Kollaborative inhaltsbasierte Video-
AMCIS '09: Fifteenth Americas' Conference on Information Systems
Suche,” in DeLFI '07: 5. E-Learning Fachtagung Informatik, Bonn:
2009, Red Hook, NY: Curran, 2009, pp. 27.11.2014.
Ges. für Informatik, 2007, pp. 281–292.
H. Hippner and T. Wilde, “Social Software,” Wirtsch. Inform, no. 47,
[16]
O. Aubert and Y. Prié, “Advene: Active Reading through Hyper-
pp. 441–444, 2005.
video,” in HT '05: Proceedings of the sixteenth ACM Conference on
C. Lattemann, “Social und Mobile Media in deutschsprachigen Unter-
Hypertext and Hypermedia, New York, N.Y.: ACM, 2005, pp. 235–
nehmen 2012,” HMD - Praxis der Wirtschaftsinformatik, no. 292,
244.
http://hmd.dpunkt.de/292/11.php, 2013. [4]
te und Perspektiven : Beiträge aus Forschung und Praxis : prämierte
P. Raeth, S. Smolnik, N. Urbach, and C. Zimmer, “Towards Assessing
[17]
C. Concolato, P. Schmitz, O. Aubert, Y. Prié, and D. Schmitt, “Ad-
A. Susarla, J.-H. Oh, and Y. Tan, “Social Networks and the Diffusion
vene as a tailorable hypervideo authoring tool,” in the 2012 ACM
of User-Generated Content: Evidence from YouTube,” Information
symposium, 2012, p. 79.
Systems Research, vol. 23, no. 1, pp. 23–41, 2012.
[18]
N. Braun and M. Finke, “Interaction of Video on Demand Systems
ing effectiveness,” Information & Management, vol. 43, no. 1, pp. 15–
with Human-Like Avatars and Hypermedia,” in IDMS '00 Proceed-
27, 2006.
ings of the 7th International Workshop on Interactive Distributed Mul-
[26]
timedia Systems and Telecommunication Services, 2000, pp. 172–186. [19]
Toward a Theory of Communication Visibility,” Information Systems
C. Zahn, U. Oestermeier, and M. Finke, “Designs für audiovisuelle Hypermedien – Kognitive und kollaborative Perspektiven,” in Know-
Research, vol. 25, no. 4, pp. 796–816, 2014. [27]
ledge Media Design, M. Eibl, H. Reiterer, P. F. Stephan, and F. Thissen, Eds, München: Oldenbourg Wissenschaftsverlag GmbH, 2006,
[21]
[22]
[24]
T. Reisberger and S. Smolnik, “Modell zur Erfolgsmessung von Social-Software-Systemen,” in Multikonferenz Wirtschaftsinformatik
F. Shipman, A. Girgensohn, and L. Wilcox, “Authoring, viewing, and
2008, Bichler, M, Hess, T, Krcmar, H, Lechner, U, Matthes, F, Picot,
generating hypervideo,” ACM Trans. Multimedia Comput. Commun.
A, Speitkamp, B, Wolf, P, Ed, Berlin: GITO-Verlag, 2008, pp. 565–
Appl, vol. 5, no. 2, pp. 1–19, 2008.
577.
F. Lehner, B. Siegel, C. Müller, and A. Stephan, “Interaktive Videos
[30]
A. Richter, J. Heidemann, M. Klier, and S. Behrendt, “Success Meas-
und Hypervideos – Entwicklung, Technologien und Konzeption eines
urement of Enterprise Social Networks,” in WI '13: Proceedings of the
Authoring-Tools,” Wirtschaftsinformatik, Universität Passau, Passau,
11th International Conference on Wirtschaftsinformatik, 2013. [31]
S. Smolnik and G. Riempp, “Nutzenpotenziale, Erfolgsfaktoren und
Diskussionsbeitrag W-28-08.
Leistungsindikatoren von Social Software für das organisationale
F. Lehner and B. Siegel, “E-Learning mit interaktiven Videos - Proto-
Wissensmanagement,” HMD - Praxis der Wirtschaftsinformatik, no.
typisches Autorensystem und Bewertung von Anwendungsszenarien,”
252, https://www.wiso-
in GI-Edition / Proceedings, Vol. 153, Lernen im digitalen Zeitalter,
net.de:443/document/HMD__HMD200612111723302935142325242
A. Schwill, Ed, Bonn: Ges. für Informatik, 2009, pp. 43–54.
91, 2006.
F. Lehner, “Interaktive Videos als neues Medium für das eLearning,” HMD - Praxis der Wirtschaftsinformatik, no. 277, 2011.
[25]
no. 65, pp. 17–37, 1997. [29]
and Hypermedia, New York, N.Y.: ACM, 2005, p. 217.
Passauer Diskussionspapiere - Schiftenreihe Wirtschaftsinformatik [23]
T. O'Reilly, “What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software,” Communications & Strategies,
F. Shipman, A. Girgensohn, and L. Wilcox, “Hypervideo expression,” in HT '05: Proceedings of the sixteenth ACM Conference on Hypertext
P. Mayring, Qualitative Inhaltsanalyse: Grundlagen und Techniken, 11th ed. Weinheim: Beltz, 2010.
[28]
pp. 357–371. [20]
P. M. Leonardi, “Social Media, Knowledge Sharing, and Innovation:
D. Zhang, L. Zhou, R. O. Briggs, and J. F. Nunamaker, “Instructional video in e-learning: Assessing the impact of interactive video on learn-
