Semantic Web Trajectory

WISS
28 Aug, 2014

Eric Prud'hommeaux, <eric@w3.org>
- Sanitation Engineer, W3C

W3C Mission

Lead the Web to its full potential.

Human-readable web exceeds expectations:
- architecture: URLs, HTTP/S, HTML, Unicode
- rendering: SVG, Canvas, (Flash), (VML), SMIL
- APIs: DOM, mouse, keyboard, touch
Machine-readable web parallel paths:
- XML, namspaces
- web services: SOAP, WSDL
- SemWeb stack

evocative picture of the Web

Web O' Data

What do we need from a Web of Data?

use cases
functional requirements

picture of machine-
readable web

?

Do the five stars capture that?

★	Available on the Web (whatever format) but with an open licence, to be Open Data
★★	Available as machine-readable structured data (e.g. excel instead of image scan of a table)
★★★	as (2) plus non-proprietary format (e.g. CSV instead of Excel)
★★★★	All the above plus, Use open standards from W3C (RDF and SPARQL) to identify things, so that people can point at your stuff
★★★★★	All the above, plus: Link your data to other people’s data to provide context

Originally developed by Tim Berners-Lee

motivational mug

Marketing challenges

privacy
- personal data
- strategic secrets

OPEN!

Marketing challenges

release engineering
return on investment
opportunity cost

COST!

Easing the investment

simple steps
incremental return

gentle ramp

Easing the investment

simple steps
incremental return
reduce initial investment

prohibitive investment prevents adoption

Past tech

RDF/XML - triples in pointy brackets
RDFa - triples in HTML
Turtle - triples in text
RDFS - open-world "schema" for triples
OWL - more open-world schema
SKOS - thesauri in triples
RIF - rules (not just about triples)
SPARQL - query triples
RDB2RDF - database as triples
LDP - triples over HTTP
Shapes - validation of triples

The Data Activity

TODO list

Stuff we're doing:

RDF validation
CSV
verticals: EGov, Health Care/Life Sciences

Stuff we're NOT doing:

Schema for JSON
JSON in XQuery
verticals: financial

CSV requirements

well-formed
canonical (schema-free) mapping
internationalization: multi-lingual, right-to-left
translation to triples, JSON, XML
table properties
microsyntax
syntax: delimiters, comments
embedded and external annotations
database: primary/foreign keys, (syntactic) type headers
external terminologies
CSV packaging/context
semantic type, mapping to URIs
default values
units (traditional datatypes)

putting off multiple headings, random access, normalization

How do we like

?

Silo image by Doc Searls, Shankill Peace Wall image by Jennifer Boyer. Both images from Flickr Some rights reserved

Project decisions

Data utility depends on whether it

Directs our actions
Affects our confidence
Costs less than the value it provides

Data can be measured in

Size
Comprehensiveness
Accuracy
Relevance

rigorous and methodical

Data investment

Unstructured

Creation cost: cheap
Sufficient for humans consumption
Expensive to process

Structured

Creation cost: moderate
Cheap to process
Expensive to merge

Pre-normalized

Creation cost: expensive
Cheap to process
Cheap to merge
Encourages viral growth?

Size/Structure (false) dichotomy

Organizational complexity

Directs our actions
Affects our confidence
Costs less than the value it provides

Data can be measured in

Size
Comprehensiveness
Accuracy
Relevance

rigorous and methodical

Clinical Informatics: information model

General-purpose graph structure for medical information:

activity times and states.
mood: considered, planned, accomplished:
Factors commonalities in e.g. Prescription vs. Injection.
acts: observations, substance administrations, procedures ...
entities: people, organizations, places.
roles: patient, nurse, surgeon.

Tells you where to write but not how to write:

terminology
data types
value sets

information model (RIM)

Clinical Informatics: terminology model

Taxonomies used for:

problem statements
tests performed
medications administered
reactions observed
diagnoses

terminology model

Clinical Informatics: connected models

information model
(Concepts and Relationships)
- Observations
- AdverseEvents
- Substance Administrations
on patient Isabella Jones
terminology model
(Terms and Relationships)
- Pneumonia a Pulmonary Disease
- Serum Sodium a lab test
- Penicillin an antibiotic
value sets
- Pneumonia, SLE, CHF
- mild, medium, severe
datatypes
- dates
- times
- intervals

Interoperability value proposition

Reduce redundancy
Commoditize systems
Effective quality metrics
Cheaper protocol development and distribution
Cheaper and more comprehensive drug reasearch

Interoperability state of the art

Short-term commercial interests oppose interop.
Lock-in provides stability.
Use cases percieved as sci-fi.

Technical barriers:

Standardization (precise communication) is expensive.
Multiple intersecting domains with intersecting authorities:
- drugs, diseases, tests, diagnoses (Terminology Model)
  pathology: SNOMED (IHTSDO), tests: LOINC, drugs: RxNorm, NDC, CPT (AMA), NPU (IUPAC), ICDⁿ (WHO), research results: MeDRA (US FDA), HCPCS, billing: NDC, efficacy: CQM (CMS) …
- graph of what happened to whom and why (Information model)
  Reference Information Model (HL7), SDTM (CDISC), MIMIC II database schema, I2B2, SMART, OpenEHR, all 700+ EMRs, closed (EPIC, Cerner) or open (CommCare)
- datatypes and units
  ISO21090, XML Schema, UCUM
Legacy systems (success of HL7 V2).
Insufficient or overly complex implementation guidelines.
Insufficient information model, inappropriate model, inconsistent data types.

The Three Star World

Screenshot of some data from City of Palo Alto - nicely visualised 3 star data

Case Studies

AT – Renewable Energy and Energy Efficiency Partnership (REEEP)
DE – German National Library (DNB)
EU – Europeana
EU – European Commission Directorate-General for Health and Consumers
EU – European Environment Agency (EEA)
EU – Publications Office of the European Union
IT – Agenzia per l'Italia Digitale
UK – BBC
UK – Companies House
UK – Department of Environment, Food and Rural Affairs (DEFRA)
UK – National Archives
UK – OpenCorporates
UK – Ordnance Survey
UN – Food and Agriculture Organization of the United Nations (FAO)

Data plumbing work

Empower groups

users, citizens, patients
geeks/developers
managers, politicians

Lead

teach
develop
proselytize

get dirty

Semantic Web Trajectory

WISS28 Aug, 2014

W3C Mission

Web O' Data

Do the five stars capture that?

Marketing challenges

Marketing challenges

Easing the investment

Easing the investment

Past tech

The Data Activity

TODO list

CSV requirements

How do we like

Project decisions

Data investment

Organizational complexity

Clinical Informatics: information model

Clinical Informatics: terminology model

Clinical Informatics: connected models

Interoperability value proposition

Interoperability state of the art

The Three Star World

Case Studies

Data plumbing work

WISS
28 Aug, 2014