Samvera and IIIF: Opportunities and Challenges presented by Jon Dunn (Indiana), Simeon Warner (Cornell), Hannah Frost (Stanford), Adam Wead (Penn. State), and Trey Pendragon (Princeton). The document discusses the opportunities and challenges of integrating Samvera with the International Image Interoperability Framework (IIIF). It provides an overview of IIIF APIs and specifications. It also highlights several institutions' experiences with and plans for IIIF, including using it for audiovisual content, experiments with the Presentation API, and consuming and creating IIIF manifests in digital collections. Key challenges mentioned include tool support for IIIF 3.0, authentication,

1. Samvera and IIIF:
Opportunities and Challenges
Jon Dunn (Indiana), Simeon Warner (Cornell),
Hannah Frost (Stanford), Adam Wead (Penn. State),
Trey Pendragon (Princeton)
http://bit.ly/2018-samvera-iiif
Samvera Connect 2018, Salt Lake City, Utah

2. IIIF Community
Credits – There are no
original IIIF slide decks!
These IIIF introductory slides
include work from Mike
Appleby Jon Stroop, Rob
Sanderson, Tom Crane,
Sheila Rabun, Glen Robson,
Tom Cramer, and probably
others

3. IIIF Community Landscape
● Manuscripts
● Museums
● Newspapers
● Software
Developers
● Audio/Visual
● Discovery
● Text Granularity
● Libraries
● Museums
● Research Institutions
● Archives
● Galleries
● Aggregators● Cogapp
● Digirati
● Klokan Technologies
● LUNA Imaging
● OCLC - ContentDM
● Synaptica
● Zegami
● 4Science
● Etc.
● 51 institutional members
Open Source
Software
● Mirador
● Universal
Viewer
● OpenSeadrgon
● Loris
● IIPImage
Subject
Experts

4. Institutional Adoption (2017)
•http://iiif.io/community/
•IIIF Adopters Survey
Presi 2.x
Image 2.x
Presi 2.x
Image 2.x
Image
1.x
Image
1.x
Presi
1.x
Presi
1.x
Auth
Search
Authentication
Search
Image
2.x
Presi
2.x
Production Development Investigating

5. Global Community (2017)
● 100+ organizations
● 345+ million images

6. IIIF Community Groups
•http://iiif.io/community/groups
•3D
•Archives
•Manuscripts
•Museums
•Newspapers
•Outreach
•Software Developers

7. Technical Specification Groups
As needs arise within the community, new TSGs
are formed to work toward specificiations
•IIIF A/V
•IIIF Discovery
•IIIF Text Granularity
•New use cases considered as IIIF evolves:
•https://github.com/IIIF/iiif-stories

8. IIIF Consortium (IIIF-C)
•http://iiif.io/community/consortium
•51 institutional members (as of 2018-10-11)
•Sustainability and steering for IIIF
•Organizational structure:
• Executive Committee – core founding members, high level direction
• Coordinating Committee – active leaders for week to week activities
• Technical Review Committee – reviews API specifications, notes and new TSGs
• Editorial Group – writes API specifications

9. APIs
Current and coming...

10. IIIF Image API 2.1 (3.0 in draft)

11. Image API Request Syntax
1) Image Information
https://example.edu/{id}/info.json
2) Image Request
https://example.edu/{id}/{region}/{size}/{rotation}/{quality}.{fmt}

12. https://example.edu/{id}/info.json
{
"@context": "http://iiif.io/api/image/2/context.json",
"@id": "https://libimages1.princeton.edu/loris/ex.jp2",
"protocol": "http://iiif.io/api/image",
"height": 7200,
"width": 5204,
"profile": [
"http://iiif.io/api/image/2/level2.json",
{
"formats": [ "jpg", "png", "gif", "webp" ],
"qualities": [ "default", "bitonal", "gray", "color" ],
"supports": [ "canonicalLinkHeader", "profileLinkHeader", "mirroring", "rotationArbitrary" ]
}
],
"sizes": [
{ "height": 225, "width": 163 },
{ "height": 450, "width": 326 },
{ "height": 900, "width": 651 },
{ "height": 1800, "width": 1301 },
{ "height": 3600, "width": 2602 },
{ "height": 7200, "width": 5204 }
],
"tiles": [
{
"scaleFactors": [ 1, 2, 4, 8, 16, 32, 64, 128 ],
"width": 1024
}
]
}

13. https://example.edu/{id}/0,1024,1024,1024/1024,/0/default.jpg
Tiles:
the 95% use case (*)
tiles with powers of 2
scaling are what
pan-zoom viewers rely
upon
(*) wild guestimate

14. https://example.edu/{id}/square/200,/0/default.png
==
https://example.edu/{id}/0,998,5204,5204/200,/0/default.png
Thumbnails:
the 4% use case (*)
● preferred sizes
● special square
support
(*) wild guestimate

15. https://example.edu/{id}/3930,550,980,2630/full/0/default.jpg
X
Y W
H

16. https://example.edu/{id}/3930,550,980,2630/250,/90/default.jpg

17. https://example.edu/{id}/3930,550,980,2630/250,/45/default.jpg

18. https://example.edu/{id}/3930,550,980,2630/250,/90/gray.jpg

20. Content
Canvas
Sequence
Manifest
Collection IIIF Presentation API 2.1
“The objective of the IIIF Presentation API is to provide
the information necessary to allow a rich, online
viewing environment for primarily image-based
objects to be presented to a human user [...]. This is
the sole purpose of the API and therefore the
descriptive information is given in a way that is intended
for humans to read, but not semantically available to
machines. [... It] explicitly does not aim to provide
metadata that would drive discovery of the digitized
objects.”
— http://iiif.io/api/presentation/2.1/#objectives-and-scope

21. Shared Canvas Data Model
Canvas
A digital surrogate
for a physical page
which should be
rendered to the
user (from Shared
Canvas)
May be x,y, x,y,t or
t in Presentation 3
The canvas is an
empty space, in
order to present
something we need to
paint resources onto it

22. Shared Canvas Data Model & Annotation
Image resource
painted – via
annotation with
motivation
sc:painting --
onto Canvas

24. http://demos.biblissima-condorcet.fr/chateauroux/demo/

25. Shared Canvas Data Model & More Annotation
Transcription
(sc:painting)
Commentary
(oa:commenting)

26. Content
Canvas
Sequence
Manifest
Collection

27. Content
Canvas
Sequence
Manifest
Collection

28. Content
Canvas
Sequence
Manifest
Collection
← change in
Presentation 3.0

29. {
label: "The institution of civil government"
metadata: [
{ label: "Author", value: ["Benjamin Hoadly"] }
]
…
}
Content
Canvas
Sequence
Manifest
Collection

30. {
label: "The institution of civil government"
metadata: [
{ label: "Author", value: ["Benjamin Hoadly"] }
]
…
}
{
label: "The XYZ Collection"
manifests: [
…
]
}
Content
Canvas
Sequence
Manifest
Collection

31. Other Properties
● Descriptive Properties
● Rights and Licensing Properties
● Technical Properties
● Linking Properties
● Paging Properties

32. Content
Canvas
Sequence
Manifest
Collection
Annotatio
n
Annotatio
nList
Layer
Range
Presentation
2.1 → 3.0

33. IIIF Authentication API
● http://iiif.io/api/auth/1.0/
● v1.0 released January 19, 2017
● (18 months from first public draft)
Doesn’t do authentication per se but provides an
interaction pattern allowing existing authentication
infrastructure (CAS, OAuth, etc.) to be used to control
access to IIIF resources

34. IIIF Authentication API
Specification describes how to
● From within a viewer, initiate an interaction with an access
control system so that a user can acquire the credentials
they need to view restricted content
● Give the client just enough knowledge of the user’s state
with respect to the content provider to ensure a good user
experience (including providing alternate images)

35. Authentication patterns
● Login
○ The user will be required to log in using a separate window with a UI
provided by an external authentication system.
● Click through
○ The user will be required to click a button within the client using content
provided in the service description.
● Kiosk
○ The user will not be required to interact with an authentication system,
the client is expected to use the access cookie service automatically.
● External
○ The user is expected to have already acquired the appropriate cookie,
and the access cookie service will not be used at all.

37. Authentication Example: Click-through

38. IIIF Content Search API
● http://iiif.io/api/search/1.0
● Specification for searching within annotations in a
single IIIF resource -- implements ^F like functionality

39. Annotation motivation terms

45. But what about discovery?
● Search API provides only search over annotations within IIIF resources
● Discovery TSG working on ways to support discovery of IIIF resources
○ http://iiif.io/community/groups/discovery/charter/
● 4 areas of work
○ 1. Crawling and Harvesting → Change Discovery API 0.1
○ 2. Content Indexing
○ 3. Change Notification
○ 4. Import to Viewers

46. IIIF API
Roadmap
2018: Presentation 3.0
Image 3.0
Change Discovery 0.2
2019: Authentication 2.0
Change Discovery 1.0
Notification 1.0
Import 0.9 (Beta)
2020: Search 2.0
Import 1.0

47. Indiana University

48. Indiana University: Context
● Not an early adopter of/participant in IIIF
● Believer in the mission/goals - founding member of IIIF Consortium
● Main motivation/participation: A/V
● 20+ year history in A/V repositories, teaching and learning, research,
annotation tools
● Large amount (10 PB+) of digitized A/V content
● Holy grail: Interoperability between tools and content

49. Why IIIF AV?
•Can we do for time-based media what IIIF already does for images
and things made up of images?
•Describe complex AV works in an interoperable way, via a manifest
• Access, viewing
• Annotation
• Search…
•Use IIIF to present time-based media consistently with image-based
objects
•Benefit from same IIIF mechanisms for linking, metadata, content
search, access control

50. IIIF Presentation API 3.0:
Canvases can now have duration(0,0)
(3110, 2102)
🕐

51. •This canvas has a
mixture of image, video
and text annotations,
targeting different
regions of the x,y space
and different extents of
the canvas duration.
•The user interface gives
the user control of the
canvas time

53. IIIF A/V Work at Indiana University
• Avalon/Hyrax:
• Hyrax support for Presentation 3.0 manifests, including for AV
• IIIF AV player based on MediaElement.js
• Representation of playlists as IIIF manifests
• IIIF Timeliner
• Musical form diagramming tool consuming/creating IIIF manifests
• AMP: Audiovisual Metadata Platform
• Machine learning + human processing to create and augment timecoded
A/V metadata
• Output as IIIF manifest?

55. Indiana University:
Other IIIF Opportunities and Challenges
• Implementing IIIF for images
• Extending IIIF beyond the Libraries, e.g. Art Museum
• Proprietary DAM systems

56. Cornell

57. Cornell Engagement with IIIF
● Involved in early discussions leading to first prototypes
● Editorial contributions since Image API 1.0 in 2012
● Early testing and experiments with static file solutions
● Use of Artstor IIIF support in Samvera application
Belief that ubiquitous IIIF will benefit scholars and the public. It will:
● create a rich environment for use and reuse of image/audio/video content
● encourage open access (without insisting upon it)
● support an environment of shared annotation
● make efficient use of library resources (reusable/shared not bespoke)

58. https://digital.library.cornell.edu/catalog/ss:20987268
ff
"Digital Collections Portal"
● Central place for
image collections
● Samvera application
● Content on Artstor
● Uses Artstor IIIF
Image API and
OpenSeadragon
viewer
● Just single images so
far, no support for IIIF
Presentation API yet

59. Experiments with IIIF
Presentation API
● Exploring
presentation and
sustainability
questions
● Use of static tiles
and manifests
convenient for
demonstrations
and mockup
Temporary location: http://resync.library.cornell.edu/iiif/hms_lion

60. Opportunities and Challenges
● Using IIIF more at Cornell!
● Broad and complete support of Presentation 3.0 in tooling, then stability
○ Significant changes 2.1→3.0 including adding time
○ Some upgrade fatigue already
● Sustainable growth of community and scope of work
○ Museum and archive involvement around the world
○ Images → Audio and Video → 3D?
● Scholar/user annotations
○ Consistent and effective UIs
○ Storage / sharing / re-use in ways that meet user needs and understanding
● Effective, efficient and sufficiently flexible tooling in Ruby/Samvera
○ Good handling of separate image/media servers
○ Workflows, modeling and generation of presentation information, manifest creation…
● Persisting URIs

61. Stanford

62. Opportunities
● Breaks down constrained, UI-based access
● Enables researcher-oriented tooling in repository
context (e.g., Mirador)
● Aligned with “collections as data”
● Encourages tech sharing more broadly,
e.g., museums
● Makes Samvera solutions highly competitive

63. Challenges
● Adoption by Google, other web content giants
● Broad uptake of tools by researchers
● Making it understandable to technically-inclined
peers

64. Penn. State

65. Princeton

66. Samvera for
Manifest
Creation

67. Figgy: Creating & Serving Manifests

68. IIIF in our Catalog

69. Spotlight: Consuming IIIF Manifests

70. Digital Cicognara: Bringing Manifests
Together
https://cicognara.org

71. Opportunities and Challenges
1. IIIF 3.0 A/V Support
2. Authentication implementation for our private resources.
3. Align to IIIF Discovery Work for our endpoints.
4. Implementing IIIF Search & Annotation.
5. Participate in more initiatives where we can share our
manifests.

72. Discussion
https://iiif.io/
https://groups.google.com/forum/#!f
orum/iiif-discuss
https://iiif.io/community/