Mike Lowndes

1
Semantic Web Think Tank - interim report
Museums Computer Group Spring meeting 2007

• Mike Lowndes
• For the SWTT Core Group
• Interactive Media Manager,
• Natural History Museum, London

2
Background
3
Why bother?

• Is being joined up important?
• A citizens Entitlement to culture (c.f.
  Scotland)
• Diversity and inclusion
• Knowledge Webs (Collections, research data)
• inter-domain interoperability? Bed and Breakfast
  (see MCG e-list)!
• Is the SW a relevant technology to enable this?
• Or is this a case of the technological tail
  wagging the cultural dog?

4
Some Web Issues For Museums (2005)

• 1997-2000 METADATA standards? Bowled a Googley.
• Whats good
• People trust online museums content and their
  links
• Its a very rich resource
• Websites dont share information in standard or
  consistent ways
• Whats less good
• The UK community is not organised for
  interoperation
• CHIN, Australian Museums etc, and other sectors
  Estate Agents and Railways.
• Our users only have Google.
• Portals are not a sustainable solution (unless
  they essentially do SEO).
• A better web needs a new infrastructure.
• Will the emerging Semantic Web deliver this? What
  does it mean for Museums?

5
The Semantic Web
6
The Semantic Web

• Tim Berners-Lee Web visionary and Head of W3C.
• Formally set off in 1998 Goal of the Semantic
  Web was the solution to information overload and
  the personalisation of the web.
• Adding logic to the web, Turning the web into
  a global database
• Semantic web software should be able to find,
  sort, classify, interpret, and present relevant
  content in context.
• This would be achieved via global use of
  metadata associated with URIs, and ontologies
• leading to vastly improved, smart browsing,
• Finally, agents which may seem intelligent
  because they can process a web that describes
  itself.

7
W3C Definition

• Tim Berners-Lee
• The Semantic Web is an extension of the current
  Web in which information is given well-defined
  meaning, better enabling computers and people to
  work in cooperation.
• For the Web to become a truly machine-readable
  resource, the information it contains must be
  structured in a logical, comprehensible and
  transparent fashion.
• This is the primary work required to enable the
  semantic web.
• Museums have lots of metadata and semantically
  rich content. Can the SW help us and our users?

8
Ontology

• DigiCULT
• The most typical kind of ontology for the Web
  has a taxonomy and a set of inference rules.
• What does an ontology do?
• Describes relationships between data,
  relationships between taxonomies. Its a
  mapping exercise.

9
Boxes, arrows and Acronyms, no clouds!
Context
User profile
OWL
FOAF
Other ontologies
Maps to
User query, or query generated by user behaviour
RDF-S/ OWL (CIDOC-CRM, SKOS)
SPARQLRuleML
Semantic Web Agent
Maps to and is constrained by
Identified ontology

• Accurate,
• meaningful
• Answers
• Actions
• Views of information

RDF (DC, RSS)

• A digital object
• http//something.somewhere/unique

Associated metadata
See the notes
10
Utopian?

• TBL sees the Semantic Web as based upon a whole
  bunch of ontologies mapped together.
• Instead of asking machines to understand
  peoples language, ask people to make the extra
  effort
• It is acknowledged that this is a vast and
  difficult thing to do
• The tools are not yet there
• DigiCult
• The Semantic Web is a direction, it is like
  North. You go North but you never arrive and say
  here it is.
• the main goals are achievable?
• It will be a part of the future web, but never
  all of it
• Any movement towards semantics increases the
  signal to noise ratio of the web
• It should and will be done where it can be
• Better for formalised knowledge anyway, informal
  knowledge can associate loosely or closely as
  desired or required

11
Web2.0 and the Semantic Web

• Joshua Allen, 2001 (Making a Semantic Web)
• Until anyone can create metadata about any page
  and share it with everyone there will not be a
  semantic web
• Web 2.0
• Is the web as an application platform
• Is the democratisation of content creation
  (interpretation)
• Is the two-way, community based web TBL initially
  wished for as Web1.0!
• Web 2.0 is NOT a new infrastructure for the web.
• wont do the job of providing the global
  database
• takes some steps in the right direction, but
  moves away from explicit structure
• provides platforms for SW applications

12
The Thinktank meetings
13
Thinktank structure and participants

• http//culturalsemanticweb.wordpress.com/
• Led by Ross Parry, Mike Lowndes, Nick Poole and
  Jon Pratty
• Other core group members Frances Lloyd-Baynes,
  Richard Light, Jeremy Ottevanger, Kostas
  Arvanitis, Areti Galani, Mia Ridge, Dan Zambonini
  
• A series of meetings comprising around 12-16
  people.
• Began in Leicester 7July 2006 Open Meeting,
  Ends with reporting in June 2007 at UKMW07
• Ross Parry - MC and keeping us all in order and
  on track
• Alex Whitfield (British Library) recording the
  sessions and transcribing for the blog.
• Special guests
• Jennifer Trant (Archimuse - steve)
• Brian Kelly (UKOLN)
• Paul Shabajee (HP Labs)
• Phill Purdey (MLA)

Jeremy Keith Mike Ellis (NMSI) Martin Bazley
(ICT4Learning) Andy Sawyer (Simulacra) Sebastian
Kruk (DERI) Robin Boast (Cambridge) Dylan Edgar
(SMC)
14
Format

• An informed but informal tone
• An exploratory conversation in the morning
• . that identifies one or two key points for the
  more focused afternoon discussion
• On-the-spot quick tutorials by some of our
  experts
• Summarising and recapitulating wherever we can
• Keeping an eye on
• our on-going main narrative (and themes) across
  all the meetings
• the final report we aspire to write

15
What have we explored?

• Leicester
• SW and sw
• Digital objects what are they?
• RDF/ Ontologies/ Topic Maps - introductions
• What do Museums do?
• Glasgow
• Participatory, interactive, retentive, trans-
  disciplinary, trans-cultural - Seamus Ross (part
  of DigiCULT)
• Co-contextuality objects have different meanings
  in different contexts. We can be context
  oriented as well as object oriented.
• Brighton
• Web 2.0 and semantics
• Tutored through tagging and folksonomies to
  microformats
• Folksonomies have emergent semantics how useful
  are they?
• Folksonomies and the formal SW interface can
  complement and improve each other
• Demonstrators instigated

16
What have we explored? 2

• Newcastle
• Learning Objects and objects for learning
• There are existing standards that may be
  significant
• Teachers want access to assets for reuse
• A semantic environment for teachers?
• London
• Collections Museum practice
• Multilingual issues! Not simple relationships
• SW seen as compatible with CM improvements and
  efficiencies can be understood. e.g.. Role in
  data cleansing
• A semantic web global loans system idea?
• More research needed outside the sector
• Cambridge
• Philosophical underpinning SW is just a tool
  not a magic bullet for knowledge representation
• Two strands of use Community of Practice (the
  Museums), Community of interest (the visitors)
• Practicality, Sector issues

17
Examples
18
Microformats - Jeremy Keith

• reuse of the tag attributes (mainly used for CSS)
  to wrap and semantically structure content in
  HTML pages.
• E.g.
• ltspan classvcardgt
• ltspan classfngtJeremy Keithlt/spangt
• lt/spangt
• loosely coupled semantics
• for humans and machines to read
• Use existing standards where possible e.g.vCard,
  iCal
• Accessed by humans via browser extensions, by
  machines via spiderssometime soon
• Coming soon, now the BBS has taken interest

19
JeromeDL - Sebastian Kruk

• A Semantic Digital Library
• Resource owners create metadata and local
  ontologies within the library
• Users can tag assets freely, tags can be mapped
  to the formal side
• Clever search
• Inference engine uses the ontologies created.
• Eg. Searching for references by the student of an
  academic.
• Metadata of a paper has no reference to the
  supervisor
• The SW profile (FOAF) of the author does
• Inference Engine uses this to infer the
  relationship
• Demonstration

20
The SWTT

• A Semantic workstation for teachers.
• diagram

Haystack (MIT) an RDF-PIM
21
Other spheres Martin Doerr Digital Libraries

• Not from the SWTT itself, we did want Martin to
  take part, and here are some of his conclusions
• It is feasible to create effective, sustainable,
  large-scale networks of knowledge
• CIDOC-CRM as a possible framework
• Thesis
• Once there is a global model, we must invest in
  managing and preserving co-reference. Else no
  large-scale networks of knowledge will ever
  emerge.
• Co-reference clusters can be distributed and are
  scalable.
• Folksonomies (and Web 2.0)
• Large number of individual tagging actions result
  in the emergence of the semantics of tags
• (these can be) lightweight, dynamic ontologies

22
Where is this journey taking us?
23
Lines of Enquiry

• Understanding the technology and its limitations
• The semantic web for the community of interest
• Questioning the user experience and expectations
  - sw
• The semantic web for the community of practice
• Policy, practice and delivery in the sector -
  SW

24
Wider issues

• What our discussions on the burgeoning SW have
  done is to focus us on a lot of broader,
  fundamental issues that the sector has to solve
  from National strategy to web publishing
  practice.
• Danger of being seen to be another lobbying group
  whingeing on about lack of resources?
• Must not lose track of the core questions
• How do we get (our content) joined-up?
• is the SW a solution to this?

25
Museums in their environment

• SW could have a very large influence on not just
  how museums view their collections, but on how
  the world views museums in the decades ahead.
• Requires
• Policy
• Skills
• Funding
• Standards
• implicit in the resourcing of any large scale
  infrastructure
• Policy
• Systemic issues
• Business models
• Practice

26
More questions than answers?

• Some evidence that Online Collection Systems are
  only used for finding not learning about objects.
  Interpretation and Access are not the same!
• Co-contexuality sharing of contexts not
  always object oriented context -oriented
  objects?.
• Semantic interoperability a goal waiting on a
  business case? NOF-Digi!? EDL 3.0?
• Getting our metadata in order across the sector
  is a good start, is it enough?
• Is the SW / sw something that your Museum
  perceives it can gain from?
• Its happening elsewhere. We need to be organised,
  ready for it, or well fall further behind.
• Case for a shift in Museum thinking about its
  content from Outreach to Inreach
• Museums as a resource, not the sole owners of
  interpretation?
• User experience Better than Google? If you can't
  beat them, join them. Or at least talk to them.

27
Potential practical outcomes

• Short term
• A small but significant change to SPECTRUM
  documentation standard, at the point of knowledge
  capture, introducing a semantic element.
• Demonstrator proof of principle SW, sw and
  Web2 applications commissioned.
• The brute force test (Google co-operative search)
  Mike Ellis
• Topic Maps from Museum data - Richard Light
• Microformats for Museum, Object, learning
  resource? Jeremy Ottevanger (MoL)
• RDF based search tools Mike Ellis and Dan
  Zamobinini (Box UK)
• Preliminary specification for a semantic
  environment for learning objects.
• Advocate the need for URIs for museum objects.
• Medium term
• Museum environment
• Digitisation still project based, not
  infrastructural. Needs to change.
• Web usage whats meaningful?
• 42 key performance indicators (DCMS), 24 of those
  related to access. Only 3 really involve
  electronic access to collections all 3 imply
  improvement in terms of larger numbers? In the
  current funding environment is this sensible?

28
Potential strategic outcomes

• Awareness raising and / or Advocacy
• Write Netful of Jewels 2.0 Recasting the Net
• a vision and a roadmap?
• Diversity of uses and users
• Openness not access
• Dont do it yourself
• Services and infrastructure requirements scoping
• Evidence-based policy recommendations
• Investigate new business models/flexible models
  in a changing environment.
• Future of the ThinkTank
• A User Advocacy group?
• A policy group?
• More than a talking shop!

29

• Thank you