EDUCAUSE Regional 2004 Presentation
Upcoming SlideShare
Loading in...5
×

Like this? Share it with your network

Share
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
579
On Slideshare
579
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
0
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide
  • 2000 – Summer – UMTV looking for video archive solution at NAB 2000 – Fall – Informix and Ascential 360 viewed as potential asset management tool 2000 – November UM team site visit to CNN – Eureka – the “indoor plumbing” of rch media asset management 2001 – Spring – Digital Asset Management defined as a possible infrastructure for the creation, experimentation and management of all asset types, but especially rich, time-based media (audio, video and 3D VR) 2001 – July – Provost and President’s office authorize VERY limited proof of concept using vendor donated software and development support 2001 – December 15: Proof of concept based on Virage, Ascential 360, Informix, demonstrated to J. Hilton, L. Rudgers, L. Monts, G. Krenze in Fleming 2001 – Winter-spring continuing demos until demo system moved to Media Union for storage. Informix purchased by IBM. Ascential fate uncertain 2002 – Spring-summer: Cross-campus team invited to develop RFP. Nine units participate. RFP issued July 2002. 12 vendors attend RFP info session; 4 respond 2002 – Fall: Five finalists narrowed to three vendor teams: Bearing-Point (KPMG Consulting)+Documentum; Sun+Artesia; IBM+Ancept. Live demos and hands-on Web test continue to March, 2003 2003 – March IBM+Ascential team selected for final negotiations unanimously by 8-unit selection team. 2003 – Spring Identification of “Affiliate” groups to be financial partners in Living Lab 2003 – July – Hardware and Software install begin in Living Lab
  • 2000 – Summer – UMTV looking for video archive solution at NAB 2000 – Fall – Informix and Ascential 360 viewed as potential asset management tool 2000 – November UM team site visit to CNN – Eureka – the “indoor plumbing” of rch media asset management 2001 – Spring – Digital Asset Management defined as a possible infrastructure for the creation, experimentation and management of all asset types, but especially rich, time-based media (audio, video and 3D VR) 2001 – July – Provost and President’s office authorize VERY limited proof of concept using vendor donated software and development support 2001 – December 15: Proof of concept based on Virage, Ascential 360, Informix, demonstrated to J. Hilton, L. Rudgers, L. Monts, G. Krenze in Fleming 2001 – Winter-spring continuing demos until demo system moved to Media Union for storage. Informix purchased by IBM. Ascential fate uncertain 2002 – Spring-summer: Cross-campus team invited to develop RFP. Nine units participate. RFP issued July 2002. 12 vendors attend RFP info session; 4 respond 2002 – Fall: Five finalists narrowed to three vendor teams: Bearing-Point (KPMG Consulting)+Documentum; Sun+Artesia; IBM+Ancept. Live demos and hands-on Web test continue to March, 2003 2003 – March IBM+Ascential team selected for final negotiations unanimously by 8-unit selection team. 2003 – Spring Identification of “Affiliate” groups to be financial partners in Living Lab 2003 – July – Hardware and Software install begin in Living Lab
  • Permissions and roles follow pre-determined lines emanating from corporate policy. Roles and permissions may be assigned by a large number of owners. Individuals may have many roles.
  • Permissions and roles follow pre-determined lines emanating from corporate policy. Roles and permissions may be assigned by a large number of owners. Individuals may have many roles.

Transcript

  • 1. Notification of Copyright © The Regents of the University of Michigan. Copyright 2004 © Louis E. King, 2004 © Lynn Johnson, 2004 This work is the intellectual property of the University of Michigan and the authors. Permission is granted for this material to be shared for non-commercial, educational purposes, provided that this copyright statement appears on the reproduced materials and notice is given that the copying is by permission of the copyright holders. To disseminate otherwise or to republish requires written permission from the copyright holders.
  • 2. Managing D igital Assets: A Coll a borative Ca m pus-Wide Exploration of Infrastructure and Application s University of Michigan Lynn Johnson, Ph.D., School of Dentistry Louis E. King, Digital Media Commons EDUCAUSE Midwest Regional Conference April 21, 2004
  • 3. Agenda
    • Vision and demonstration (King, 10 mins)
    • Content needs across campus (Johnson, 5 mins)
    • DAMS Partnership: Real World Applications of DAMS (Johnson, 10 mins)
    • DAMS architecture (King, 10 mins)
    • Success factors identified and issues under investigation ( King, 5 mins)
  • 4. What is the Vision for a Digital Asset Management System at the UM?
    • Ingest, manage, store and publish digital rich-media assets and their associated metadata.
    • Streamline the “workflow” required to create new works with digital rich-media assets.
    • Search, share, edited and repurpose assets in the academic model.
    • Prepare for future application of campus-wide rights and intellectual property management to existing assets.
    Create an infrastructure that will lower the barriers preventing us from using time-based media in a manner similar to our use of text and images today!
  • 5. What is the DAMS Living Lab?
    • Collection/File Management: History of Art Department, School of Music
    • Data Mining: Business School, School of Information
    • Exchange: School of Dentistry, College of Pharmacy, English Department, Department of Psychology
    • Notation: College of Education
    • Workflow: School of Information, School of Music
    A partnership between UM, IBM and Stellent (formerly Ancept) to provide an experimental DAMS infrastructure for proof-of-concept projects using rich media in higher education.
  • 6. Collection/File Management
    • History of Art slides
    • Dentistry procedures
    • Music Performances
    • ePortfolio
    • eReserves
    • Campus events
    A repository for gathering, tagging, managing and sharing rich media objects for teaching, learning, comparison, exhibition or hobby.
  • 7. Collection/File Management
  • 8.  
  • 9. 1 2
  • 10. 1
    • 2
  • 11. 1 2
  • 12. 1 2
  • 13. 1
  • 14. 1 3 4 2
  • 15. 1 2
  • 16. 1 2
  • 17. 1 2
  • 18. Data Mining
    • Visiting lecture series
    • Learning objects
    The use of text based search engines to expose the content of rich media by searching metadata created by video and audio analysis tools. (voice to text, voice & facial recognition, optical character recognition, etc.)
  • 19. Data Mining
    • Searchable Video Analysis Metadata
    • Keyframes, scene changes & clip playback
    • Speech to text
    • Facial recognition
    • OCR
    • Closed caption
  • 20. Media Exchange
    • Learning materials – lectures, presentations, and articles, etc.
    • Learning modules
    • Video clips
    • Video of performances
    • Sound bytes
    • Commercials
    • Foreign TV programming
    An organized center for trading digital media.
  • 21. Notation
    • Marking and coding “learning moments” in classroom footage
    • Marking and coding problem areas in video-based usability testing
    • Facial recognition in panel presentations
    • Number recognition in athletics
    The use of metadata to mark and code events in rich media for retrieval, comparison, and analysis.
  • 22. Workflow
    • Distance learning course production
    • Promotional CD production
    A virtual environment for guiding a collaborative team’s work process in rich media through a series of gated activities.
  • 23. Real World Applications for DAMS: A Partnership
    • Goal:
    • To help determine if a single DAMS can serve the needs of multiple content areas thereby serving as a proof-of-concept for a campus-wide implementation.
    • Members:
      • School of Education
      • School of Dentistry
      • College of Literature, Science and the Arts (Language Resource Center)
      • School of Nursing
      • College of Pharmacy
      • School of Social Work
      • Information Technology Central Services
      • University Libraries
      • CARAT/Rackham/Fellow—Yong-Mi Kim (SI)
  • 24. Why Do We Need DAMS?
  • 25. Why do we need a DAMS?
  • 26. Our Successes To Date Evaluation
    • User’s evaluating themselves
    • Please rate your perceptions of your knowledge, experience, and confidence of … metadata
      • Ex. I know why I want to use metadata.
    • Please rate your perceptions of your knowledge, experience, and confidence of … Digital Asset Management Systems
      • Ex. I know what I want the UM Digital Asset Management system to do.
    Baseline Participant Perception Indicators (PPI) were developed and administered by the College of Pharmacy
  • 27. Our Successes To Date Metadata
    • Metadata issues
    • Heterogeneous collections with domain-specific description needs, vocabularies
    • Metadata entry will be handled by non-specialists
    • Granularity of description vs. effort
    • Consistency vs. flexibility
      • Controlled vocabulary v. free text
    • Enable search
      • Across subject area
    Metadata is, “A cloud of collateral information around a data object”. Without metadata information objects cannot be found.
  • 28. Our Successes To Date Metadata
    • Descriptive Metadata
      • UM_Core
        • Elements of Dublin Core
        • Local elements
        • General, simple metadata standard intended for use by non-specialists
      • IMS
        • Metadata standard focusing on educational uses
    • System Metadata
      • Metadata automatically collected by Ancept system
        • File size, file name, data rate, etc.
  • 29. Our Successes To Date UM_Core
    • DC_Title
    • DC_Creator
    • DC_Subject
    • UM_SecondarySubject
    • DC_Description
    • DC_Publisher
    • DC_Contributor
    • DC_Date
    • DC_Type
    • DC_Format
    • DC_Identifier
    • DC_Source
    • DC_Language
    • DC_Relation
    • DC_Coverage
    • DC_Rights
    • UM_Publisher
    • UM_AlternatePublisher
  • 30. Our Successes To Date System Metadata
  • 31. Our Challenges to Date Metadata
    • Content domains have poor legacy metadata
      • SNODENT
      • Need controlled vocabularies within education
    • Next Step:
    • Ingest significant amount of information and “test” the metadata
      • UM_Core
      • Dentistry
      • Education
  • 32. Our Successes to Date -- Privacy Rights & Permissions for Access Control Lists (ACLs)
    • Research Conducted
      • Gathered samples documents from four campus units
      • Reviewed for common elements
      • Legal review for healthcare consents
    • Next Steps
      • “Wizard-like” guide for writing consents (Web)
      • Summary of case studies (Web)
      • Publish manuscript referring to Web site
  • 33. Future Directions
    • Focus Groups
      • Identify potential uses of DAMS by students and faculty
    • Ingest, ingest, ingest, ingest
      • Dentistry—15 most commonly used videos
      • Dentistry—300 diagnostic quality oral pathology images
      • Education—student produced videos for e-portfolios
      • Education—classroom videos for research and professional development
  • 34. Future Directions
    • Validate
      • U_M Core Metadata
      • Domain specific metadata
      • Usefulness of voice to test
  • 35. DAMS Architecture
  • 36.
    • IBM and our identified partners will deliver an end-to-end digital asset management system to the “Living Lab” as a working demonstration environment.
    • IBM and the University of Michigan will identify areas for collaborative research projects around subjects such as digital rights, open standards, and learning technologies.
    • IBM and the University of Michigan will use the Lab to support the delivery of pilot projects that will be used to explore a variety of existing and emerging digital asset management technologies.
    • IBM and the University of Michigan will co-create a marketing and communications program to promote the Lab’s efforts across campus, to the Big Ten CIC, to other higher education organizations and the Merlot consortium.
    IBM & Stellent
  • 37. What space does Dams occupy? Production, Publications, Broadcast Content Collaborative Research Archived Collections Casual Learning & Exploration Course Materials Digital Libraries Departmental Storage Team Workspace/Storage Content Mgmt. Systems Personal Share-folders Production Systems Institutional Repositories Collaborative Learning Types of Collaboration Ad-hoc Sharing ePortfolios Course Mgmt. Systems Individual Content Owners Institution Individual Browsing Research Portal Development & Content
  • 38. What is the place of DAMS in the campus infrastructure? Applications, Course Management Systems, Production Systems DAMS Institutional and Individual Assets Network Storage Publishing: Teaching, Collaboration, Production, Distribution, Broadcast Authentication & Authorization
  • 39. DAMS Component Services Ingest Publish Manage Enterprise Data Producers Collaborators Audience Encode Transcode Metatag Proxies Encrypt Store Traffic File Serve Streaming Broadcast Web Pub. Printing CD/DVD View Metadata Access Workflow Version Check in/out DRM Course Mngmnt Campus Services Campus Broadcast Print Publishing Secure Web Public Web Store Unit Unit Unit Unit Unit Unit Near-line Near-line Near-line Near-line Offline Offline Offline Authoring Stations Media Appliances Remote Users Campus Users Studios
  • 40. DAMS Living Lab Configuration
    • Local source:
    • Tape Deck
    • Live Media Stream
    • Scanner
    • Existing Digital File
    • Remote Source:
    • Telestream ClipMail Pro
    • FTP upload of existing digital file
    Library Server Resource Manager Ancept Media Server Metadata creation Version control Check-in/out Workflow XML Websphere Tivoli Storage Management Asset Processing Streaming Servers IBM VideoCharger Apple QuickTime 1 TB storage Telestream Flipfactory Transcoding Metadata Extraction Proxy Creation Virage Encoding & Logging Metadata Extraction Speech-to-text Voice, face recognition DB2 SMART Self-Management And Resource Tuning IBM Content Manager Metadata Mngmnt. Resource Management Security Cosign single sign-on Remote iSCSI Storage 1TB
  • 41. DAMS Living Lab Configuration
    • Local source:
    • Tape Deck
    • Live Media Stream
    • Scanner
    • Existing Digital File
    • Remote Source:
    • Telestream ClipMail Pro
    • FTP upload of existing digital file
    Library Server Resource Manager Asset Processing Transcode IBM x345 dual 2.67GHz Xeon, 1.5 GB DDR Telestream FlipFactory 440 GB Ultra 320 (10K rpm) SCSI Raid 5i Windows 2000 Server Encode and Log IBM x345 dual 2.67GHz Xeon, 1.5GB DDR 146 GB Ultra 320 (10K rpm) Osprey 560 Video Capture Card Video Logger (Virage) Windows 2000 Server Video Streaming IBM x345 dual 2.67GHz Xeon, 1.5 GB DDR 146 GB Ultra 320 (10K rpm) QuickTime Server IBM Video Charger Windows 2000 Server Device attached storage 1 TB - IBM Ultra 160 (10K rpm) IBM Ultra 160 (10K rpm) Content Manager 8.0 (DB2, Tivoli, Websphere) Ancept Media Server IBM P645 2-way 1.2GHZ, PWR4+, 8GB 293 GB SCSI U3 (15K rpm) AIX 5.1 Remote iSCSI Storage 1TB
  • 42. DAMS Living Lab UMCE Integration
    • Local source:
    • Tape Deck
    • Live Media Stream
    • Scanner
    • Existing Digital File
    • Remote Source:
    • Telestream ClipMail Pro
    • FTP upload of existing digital file
    Library Server Resource Manager Asset Processing Transcode IBM x345 dual 2.67GHz Xeon, 1.5 GB DDR Telestream FlipFactory 440 GB Ultra 320 (10K rpm) SCSI Raid 5i Windows 2000 Server Encode and Log IBM x345 dual 2.67GHz Xeon, 1.5GB DDR 146 GB Ultra 320 (10K rpm) Osprey 560 Video Capture Card Video Logger (Virage) Windows 2000 Server Video Streaming IBM x345 dual 2.67GHz Xeon, 1.5 GB DDR 146 GB Ultra 320 (10K rpm) QuickTime Server IBM Video Charger Windows 2000 Server Device attached storage 1 TB - IBM Ultra 160 (10K rpm) IBM Ultra 160 (10K rpm) Content Manager 8.0 (DB2, Tivoli, Websphere) Ancept Media Server IBM P645 2-way 1.2GHZ, PWR4+, 8GB 293 GB SCSI U3 (15K rpm) AIX 5.1 SMB SAMBA Cosign ITComm Remote iSCSI Storage 1TB
  • 43. Lessons Learned - Managing Access Control Lists The commercial sector defines access and control based on the individual’s relationship to the company (role) Easily managed centrally through system defined ACLs Higher education defines access and control based on the individuals relationship to the media Requires distributed management through User Defined ACLs Access & Control Hierarchy Roles Access & Control Hierarchy Roles Decision Making + - . Board of Directors Executive Staff Administrators Customers - Level 1 Customers - Level 2 Customers - Special Rights Holders / Creators Licensees Administrators Collaborators Groups Open Access Board of Directors Executive Staff Administrators Customers - Level 1 Customers - Level 2 Customers - Special Regents Executive Staff Faculty Students Staff Friends/Affiliates
  • 44. Lessons Learned - More Access Control Lists Needed In the commercial sector, access to media is defined and controlled centrally Dozens of Access Control Lists In Higher Education, individuals and groups control access to media and define their own levels of access 100,000+ Access Control Lists Rights Holders / Creators Licensees Administrators Collaborators Groups Open Access Board of Directors Executive Staff Administrators Customers - Level 1 Customers - Level 2 Customers - Special
  • 45. Lessons Learned - Architecture Neighborhood Central Local Live Tape/CD/DVD Internet Appliance Satellite Video Logger Flip Factory (Optional) AMS 3.5 Content Manager DB2 Spinning Disk Spinning Disk Nearline/Offline Backup Tivoli Print Web CD/DVD Media Streaming Real/Win/QT Course Mgmt ePorfolios Personal Storage Capture Ingest Manage Store Publish X X X X
  • 46. Lessons Learned - Policy
    • Copyright issues must be addressed in a systemic way – UMCore metadata schema will likely require a rights declaration
    • Digital Rights Management modules may be more cost effective managing regulatory issues such as FERPA (Family Educational Rights and Education Act) and HIPAA (Health Insurance Portability and Accountability Act)
    • Education, training, and support on metadata and file quality will help distribute the work load and maximize the ability to share materials
  • 47. Long Term Issues
    • Ongoing interface design to meet project and user requirements. (Taking into consideration asset management’s inherently different approach of presenting multiple items, each of which may have a different set of user capabilities associated with it)
    • Integration with other academic tools or portal.
    • Relationship to Library, Institutional Repository and federated catalogue searching.
    • IP, Copyright, Use and Misuse policy.
    • Building a great user experience.
  • 48. UM DAMS Contacts University of Michigan DAMS Initiative http:// sitemaker . umich . edu /dams/ Louis E. King leking @ umich . edu Producer, Digital Asset Management System Lynn Johnson, Ph.D. lynjohns @ umich . edu School of Dentistry Alan McCord, Ph.D amccord @ umich . edu Vendor and Institutional Relationships John Merlin Williams jmerlinw @ umich . edu Director, Digital Media Tools Lab