openwebworks

Wednesday, June 17, 2009

We seek to solve online media publishing, cataloging, accessibility, findability, and preservation in one simple interface. Open source, open standards, modular architecture...this thing would solve common problems faced by public, educational, and community media.

What follows is the current draft of a white paper making the rounds at the University of Illinois, authored by me and the other members of the Illinois Multimedia Steering Group.

Whitepaper V2
Authors: Drew MacGregor, Brian McNurlen, Colleen Cook, Jack Brighton, Professor Bill Hammack
TIME: The Illinois Media Engine
(Intended audience: Illinois administrators, faculty, IT professionals)

The Illinois Multimedia Steering Group (MSG) is a cross-disciplinary user community dedicated to enhancing rich media at Illinois. Since early 2008 we have been developing a vision for media at Illinois, recently partnering with Professor Bill Hammack. This is version 2.0 of our whitepaper, with further revisions to follow.

The state of rich media at Illinois is currently in great disarray. While pockets of excellence exist at unit levels, Illinois lacks a campus-wide vision or strategy, precluding us from being leaders in this rapidly growing field. The key to maintaining and extending the University's reputation for academic excellence lies in offering a global audience the core of the university: The minds and thoughts of our faculty and students. Increasingly, the global audience is reached through online rich media. To become effective we must move from "individual heroics" to coordinated planning on a large scale.

In all three facets of the University- research, teaching, engagement- most indicators would show us lagging behind our peers in digital media. We are not effectively promoting our leading scholarship and works of engagement with video. We have no central facility for producing instructional media, as many of our peers do. Our website, now one of our most important marketing tools, does little to connect viewers to rich media content. Using the benchmark peer analysis outlined in the strategic plan, we are clearly falling short of these goals. The University of Illinois aspires to be the preeminent land-grant institution in the country, yet our use of rich media compares poorly with our CIC peers.

Examples from our peer institutions abound:

•    MIT physics professor Walter Lewin provides lectures free online. A 17-year old from India wrote "Through your inspiring video lectures I have managed to see just how beautiful physics is, both astounding and simple."
•    The YouTube sensation "The Large Hadron Rap" accurately conveys knowledge related to particle physics and the supercollider in the form of a “physics rap” - and has gathered millions of views.
•    Harvard has launched bigthink.com, featuring their greatest minds talking about the problems of the day in short, well-produced video segments.
•    In September 2008, the Stanford School of Engineering launched "Stanford Engineering Everywhere" (SEE), a service providing introductory Engineering courses for free online. The intent is to form virtual communities around the Creative Commons license with Stanford education at the core. The SEE initiative touches on several areas addressed in the 2008 Horizon report: Grassroots video; collaborative webs, collective intelligence; and social operating systems.

To remain competitive the University of Illinois needs to establish a rich and deep multimedia website to reach global audiences. Such a presence is now essential to attract students, faculty and the support of the public. The MSG believes that an Illinois vision of online education can extend and surpass existing models by leveraging our areas of excellence in a new digital media initiative. Illinois is renown as the institution that developed sound-on-film, plasma display technology, the transistor, public broadcasting, the LED, the first graphical web browser (Mosaic), and PayPal and YouTube through our alumni. But until now, we have abdicated our leadership position in rich media to peer and lesser institutions.

While it must be acknowledged that we are behind, the MSG believes Illinois has an opportunity to surpass existing models of rich media in higher education. We have an opportunity to innovate and lead by developing a new kind of content-rich web multimedia presence - a Web 2.0 new media service built on open standards and best practices in engagement, accessibility, library science, and engineering. Each of these areas has long been strength at Illinois. They are all coming together online, at a moment when students, faculty, alumni, and a growing global audience for rich media are looking for leadership and innovation. They are looking not only to connect with and consume media, but to interact with, create, and share content through emerging social media networks and practices.

Ronald Burt, in the "The Social Origin of Good Ideas", studied how good ideas grew in corporations; he observed that managers who "bridged", that is, knew what happened in other sections of their company excelled in useful idea creation. Here a social aspect applies: Those that observed the other sections through social contact could see the "holes" and thus bring forward powerful new ideas. Campus-wide Web 2.0 video sharing and other social networks provide the ability to bridge and see these holes. This is critical in our 21st century paradigm of cross-disciplinary research and teaching.

The Illinois Media engine
The "traditional" approach would call for media and IT at Illinois to determine new workflows and try to centralize servers as best we can. However, this strategy reinforces the flawed premise that solutions can be determined solely via IT modeling. This sort of incremental change fails to change the community mindset, nor does it set Illinois apart as an innovator. Simply using the web to deliver video instead of via television doesn't make for new media. The latter requires three essential elements: a) an architecture of participation, e.g. the ability for the public to sort and rate it; b) the ability of any person at the University of Illinois to create and upload content, and c) a "long tail" - essentially an infinite digital archive.

We call for a 2-3 year open-source software development project that will set the University of Illinois apart as an innovator in the area of media distribution and online collaboration, and position us to massively grow our use of rich media for education, research, and outreach. We envision TIME: The Illinois/Internet Media Engine. For lack of a better description, TIME is partly a "YouTube for Illinois", in that we have a common database of media assets and a central front end tied to a distributed network of servers. However, the media engine goes far beyond the capabilities of YouTube, allowing a greater suite of tools to audiences and content creators. Ideally this engine would contain video of every seminar given on campus, every conference held at the University, a three-minute video of every faculty member and thought creator, every art performance. The real power of this medium resides, however, in the users. The key to a successful long tail lies in uploading a huge amount of content fitted with social bookmarks and folksonomic tagging that allow users to rate, comment and forward video. A user of the site should be able to easily search for content, browse by subject or department, sort by rating, length, and so on, and to rate and comment on videos. Additionally the videos should be easily downloadable to an iPod or other handheld device.

Tools would be modular, added, and enhanced as development time exists. The tools might include:
•    Simple uploading of media content ala YouTube, but with much higher quality;
•    The ability to add library-grade cataloging information (metadata) based on open standards and best practices;
•    User tagging of media assets (folksonomy), tied to an individual account, allowing a student to create personal bookmarks of lecture content for later review;
•    Granular authentication schemes for faculty to protect content;
•    The ability to create a playlist or curriculum from a database of assets created by a user, or shared by other users, for example linking various micro-lectures to video blogs and podcasts;
•    Student-created content as part of a publicly-accessible workflow;
•    Commenting, annotating, and tagging specific points of a lecture to allow threaded discussions within a media asset;
•    Management tools that let the creator determine how long content remains "live", if it can be downloaded, or tagged, all from one interface;
•    Hooks for captioning and transcripts to allow greater searchability into the media assets, and accessibility for all users regardless of perceptual abilities or impairments;
•    A repository for all source media files, so they can be digitally archived and reformatted as technologies advance, preserving access to these vital resources for generations to come.

At this time we find no tools available to provide this level of control over assets, audience control, accessibility, and intellectual property. Several departments have looked at commercial products to serve micro-lectures, but they primarily offer only a video server. The content must still be embedded in local web pages on an ad hoc basis. We propose a project using the web content framework Drupal as the development component, allowing us to share the fruits of our efforts with the larger academic community.

We envision four main areas of concentration related to rich media and the concept of a common media framework: Content Creation; Accessibility and Standards; Distribution; and Archiving and Preservation.

Content Creation
It is difficult to briefly summarize content creation because it takes so many forms. It ranges from cell phone-created content, to screencasts on a desktop, to lecture recordings, to high-end video productions. Each level of creation has its merits and cannot replace another. For example, the quality of content for a short student project need not be the same as an interview with a distinguished researcher. High-end content production can take between 2 and 50 hours per one hour of recorded content.

While the methods of creation differ, they all represent the University of Illinois and deserve to be considered as such. Currently we lack the ability to do much high-end creation because of a lack of resources, studios, and professional staff. As a result we have a large amount of user-created content that goes unsupported.

In addition to TIME, we propose a Virtual Center of Excellence for rich media creation, centered on the IT media community. This proposal would require all members of the virtual center to give 4 hours per week to media projects related to the campus. These hours might be spent giving workshops, creating tutorials, or assisting other media professionals as needed. Further, we posit the idea of formal subject-matter experts who are on call to units to act as a sounding board in their field. ACES has great expertise with field shoots, ATLAS with creating learning objects, WILL with broadcast-style productions and documentaries, and Computer Science with classroom capture.

These are steps that the MSG will take on our own without University involvement. However, if the growth of new media is important to the University, we must begin to treat it as such. Somehow we must find a way to subsidize media development to a greater degree. Cost-recovery models do not scale for rich media. Students tell us that podcasts and lecture capture are extremely important, thus we should fund it through student fees. Lastly, we must recognize that rich media is more than a recording of an event, it is a tool that can lead to learning outcomes.

If we begin to look at a distinguished lecture less as an event we record because others do it, and instead design it with specific learning objectives and audiences in mind we will begin to see new opportunities for learning communities. If we can educate professors that a 3-minute interview on their research can lead to greater media exposure, our faculty may embrace this medium with more enthusiasm. If we stop selling the idea of lecture-capture as a way to replace lectures, and instead focus on the ways students learn, our faculty and students might interact better.

A long-term goal might be to standardize on a single lecture capture system, but that requires capital and political buy-in. Instead if we can develop a physical center of excellence for new media, a media sandbox, we can spend time with faculty teaching them how to create better content on their own. This concept allows for greater scaling of content, and allows more users to create content for the media engine we propose.

Content creation can be easy; anyone with a camera can do it. Content creation is also hard when we talk about large-scale and high-end content creation. We want to make it easy for users to create and publish their content. We want to make it hard for our peers to copy our vision for rich media.

Accessibility and Standards
In 2008 the Illinois Information Technology Accessibility Act (IITAA) went into effect in the state of Illinois. The act requires captioning, transcripts, and alternate accessibility options on digital media to those with learning disabilities and audiences with visual and hearing impairments. Beyond compliance with the law, adherence to the specifications of the act creates a better learning experience for all users by making the content more searchable and accessible for all concerned.

Research into a viable transcription solution by our community and CITES has led to packages that do not scale in our large environment. With the reality that an investment in development funds may be less costly than an IITAA lawsuit and accompanying bad press, we encourage the University of Illinois system of campuses to fund a program the develop a scalable solution useful to all State organizations that produce rich media. The funding of such a project would result in reduced costs to other educational and government institutions that are struggling with the same accessibility challenges.

The University of Illinois features some of the top accessibility experts in the world. By creating an environment where they can develop a software solution or partner with a software development firm to create a scalable solution, the University will maintain leadership in accessibility, better serve all users of our content, and benefit the State of Illinois directly.
Ideally the transcription process would be open source and independent of format in order to create a modular solution to fit the many models of media creation. The output of the process could then be plugged into the larger media distribution model we propose. This is a significant problem, and one that major software firms have yet to solve. One short-term option might be to encourage greater service/learning time by giving students service credit for contributing to transcripts of content. Long-term we need to realize we have an opportunity that requires capital.

Distribution
Just as the 2008 Horizon report highlighted emerging technology related to new media, the 2009 report presents opportunities that play to the strengths of Illinois. In the area of Cloud Computing, Illinois researchers are at the forefront of networked systems.

Currently the distribution of media content at Illinois is compartmentalized, with units deploying media servers and multiple front-end publishing points. Finding multimedia content on the Illinois web is an exercise in frustration, even for advanced users. In 2007 the MSG deployed a product called Ensemble to create a common publishing point, and to allow audiences to locate media in an easy, searchable manner. This solution works within our current model, but isn't an effective solution for growth, nor does it address fundamental issues related to accessibility. Consolidating servers is an IT approach to the problem. We must address the systemic issues related to how our audiences find, use, annotate, and share content.

Development of a media distribution model for Illinois should follow two tracks. First, we must develop a 2-3 year plan to centralize media servers to either Centers of Excellence or within Central IT. These servers would be standards-based, not relying on proprietary systems and expensive cost models. Further we must stop limiting our servers to "instruction" or "administration" and open these servers to address all of our areas of service and engagement. No matter the eventual model for IT at Illinois, greater coordination of servers tied to a common file service is critical to growth and stability. Maintaining the current paradigm of many small servers will preclude growth of media use to smaller units, thus driving faculty to post their content to external sites where they will be forced to forfeit some of their intellectual rights. Such an ad hoc approach is both wasteful of university resources, and ultimately unsustainable. We must build a solution that will gracefully scale, and provide for the preservation of valuable media assets.

Second, we must develop a front-end for all media at Illinois. Whether units buy into a central server model or create proprietary content, audiences need a single access point to search for media at Illinois. Ensemble works for this now, but licensing prevents large-scale use for departmental use, let alone student-created content. Our concept for TIME would permit all content to be found within one common front end. The front-end development would create the following opportunities that do not currently exist:

•    Departments could create micro-lectures of 5 minutes and manage them within the common media framework. Rather than have to rely on a server admin to move data and create links, the content manager front-end would provide faculty members control over which learning objects were presented to students at any given time. Further, a professor could also share knowledge on a subject to colleagues in other departments or other Universities, because we would be able to provide them with control of their own academic content. This is currently an area of interest for the College of Business.
•    Channels could be developed to target audiences such as alumni groups. Playlists could be created using a common interface for all users, with multiple link points managed by a central database.
•    Researchers at institutes such as the Information Trust Institute or the Advanced Digital Sciences center could pool their media content with other departments and groups to create large-scale online learning centers. Imagine a channel unique to an Illinois media database where lectures from the Siebel Center could be combined with a researcher's video blog, a live conference from Singapore, and a guest lecture at the Beckman institute. Right now we have to email a lot of links around; in a new model we treat media assets as learning objects.
•    The concept doesn't preclude the use of YouTube, iTunes U or other networks, as they play an important role for many audiences. The idea is that all Illinois content could be shared, branded, and hosted here while maintaining intellectual control. It could also be uploaded and shared with other hosting systems with greater efficiency than by simply using their usual interfaces.
•    The common media engine would position Illinois with the ability to create multiple cross-disciplinary learning centers that would make the SEE program seem limited in scope.
•    A common media distribution system would permit granular access control, allowing specific content creators to post directly to a site, change access permissions from public to password protected, or allow for download to mobile devices.
•    We also need to be flexible in our planning to allow the next generation of online tools to blend with whatever we design now.
•    Well-considered planning could engage other media beyond audio and video, opening TIME to be extensible to images, blogs, and other user content…truly adding the "multi" to the media.

In both the networked server and media framework development tracks, the collaboration of multiple disciplines and units would be required. The scope of the collaboration is daunting and potentially massive, with researchers acting as focus groups and IT and media professionals interacting as subject experts. We do not minimize the scale of what we propose; rather we see it as an opportunity to engage in a common campus objective that leverages the strengths of our faculty, our departmental expertise, and both central and decentralized IT staff.

The 2008 and 2009 Horizon reports both reference Mash Ups and the Personal Web as emerging technologies and trends. The old model of posting a media asset and links to servers doesn't work for new web users. To be relevant and speak to our audiences we must be more user-centric and make the content mobile and interactive. New communication tools get socially interesting when they get technologically boring - for young people today the new social tools are passing beyond normal and heading to ubiquitous with invisible coming soon. The University must have a meaningful presence in Web 2.0 before invisible fully arrives, otherwise our use of the medium will appear ham-handed and graceless.

Archiving and Preservation
A critical component to the University and the future of digital media is the question of how we handle archiving and preservation. We have the top research library and Graduate School of Library & Information Sciences, but we have yet to develop a model of collaboration between media producers and our archives. Further, we lack a common cataloging framework to share information across units.

We propose to change this as well in this concept of TIME. First, we must move toward a standard metadata standard such as PBCore, a standard rapidly being adopted in the moving image archival profession. Second, we must engage the University Archives on methods for archiving content in the near and long term. We often hear that it's a shame we don't have recordings of renowned Classical Studies Professor Scanlon, but we ignore the current notable faculty we aren't capturing for future generations. Third, the Library must be a key stakeholder in the common media framework we propose. Rather than asking the archivists to work on a separate system, we must ensure that all assets and meta-information can be exported to the library system so as to eliminate the hurdles we face today. Further, we need the information sciences professionals to have a say in the user interface. In this way we will move toward a model where rich media/IT/and the library are all on the same database.

Right now we lack archiving of most of our department-created content. To get it into the archive is another process outside our workflows. Many of us don't even know what should be archived. An integrated media database and framework would enable the stakeholders to better share and preserve what will someday be priceless as our rich digital heritage. It is also an opportunity for Illinois to lead.

Emerging Opportunities
Two significant educational and research opportunities emerge from the concept of TIME. First, the development of the engine would create a "trusted learning engine" for K-12 educators in the State of Illinois. Educational models created by professors and top students would be made available to schools. Teachers would be able to create learning objects for their students by logging in and selecting media by defined criteria (highest ranking, "top creator status", subject categories, user tagging, etc). Currently a teacher must seek out and filter YouTube content or locate it on TeacherTube.com. TIME would provide them with a trusted resource based on Illinois research and teaching.

The creation of TIME would also create new research opportunities for Illinois faculty by providing data that is currently unavailable or segmented. The possibility to study how users interact with each other socially while at the same time gathering quantifiable statistics about the engine would benefit multiple fields. These research specialties might discover new interconnections through the collected data. Server logs and hard data would benefit these groups, among others:

•    Data mining researchers;
•    Server and multimedia Quality of Service research;
•    Networking and system design engineers;
•    Educators investigating the effectiveness of learning objects and/or new media on age groups;
•    Library and Information Science researchers investigating folksonomy, user tagging, and archiving on a large scale;
•    Sociologists and communications researchers interested in seeing how social networking interacts with education.

TIME can benefit Illinois by engaging students and educators, while opening new researching avenues made possible by the dataset it generates.

Summary
If we are to be the preeminent Land Grant University, we must take initiative and lead by example. We are already far behind our peers in terms of rich media systems; incremental change will not set us apart. Instead we propose our vision for a software development project to create the Illinois Media Engine.

Our proposed project has the potential to bring the community closer together with a common goal. It will require us to collaborate in new ways, while checking our individual agendas at the door. It will lead to growth stature, resources, and opportunities as we enable faculty and students to take ownership of their own media creations. It will set us apart as a visionary leader in the future of education. It is not an easy proposal, but only a bold call to action will move us beyond merely keeping up with our peers. To be blunt, we propose a "moon shot", a project to engage the campus to work together on something we cannot create individually.

This 2-3 years-long software project will require server resources, programmers, and a project manager. It will also require subject-matter experts to contribute their time to the project as advisors. It will require IT media staff and the Library to act as both consultants and clients.

In this time of shrinking budgets we know a proposal like this will be met with skepticism. We believe there are opportunities for funding from outside sources, alumni donors, and seed money. If we fail to work toward a large goal, our media creation will continue to be fragmented and differently branded. Change in the area of media at Illinois must happen, and if it is important that we speak to audiences in the new media world, we need to make it happen. Berkeley, MIT, Harvard, Michigan and others are reaching out to underserved audiences through their online media efforts. Viewers in the developing world can now view micro lectures from experts in a variety of disciplines. Why not Illinois? Rather than comparing ourselves to these institutions, what if Illinois were to be the standard by which others choose to measure themselves?

We hear a great deal about Cyberinfrastructure; what better way to stake our claim to the future than to share a media engine designed by top researchers with the rest of the world. Beyond dissemination of our knowledge through rich media, Illinois will position itself as the innovator of future knowledge distribution. Imagine the Illinois Media Engine providing all public schools in Illinois a secure space to upload and preserve student created content into the future. What if TIME enabled us to combine University of Illinois media with other state agencies to provide true public education in a trusted learning environment?

We invite the campus administration, faculty, and IT professionals to endorse this concept and assist us in creating a development team to implement the Illinois Media Engine.

 

(0) Comments/Permalink | Categories: media • technology