WG2 Interim meeting summary

Denise Warzel

WG2 Convenor

32N2022

WG2 Interim Meeting in London

• Attendees

• 6 Countries participated • Canada, China, Japan, Korea, UK, USA• Liaison from SC 7 Peter Linington ISO/IEC 10746 RM-ODP

• Reviewed Liaison assignments, program of work

• Established a task force on Identification and Registration

• Discussed future meeting schedule – trying to locate near WG3 and WG4 Nov 15-19 in San Jose, USA area

• Documents: SC32 Website London: 32N1935 Resolutions, 32N1936 Meeting Report

ISO/IEC 19763 Update

• Discussed 19763 (MFI) reorganization and refactoring of Part 2 and Part 4 to combine Core model and Model Mapping

• Discussed aligning 19763 Part-5 (Process), Part-7 (Service), Part-8 (Role&Goal) with 10746 RM-ODP and 19793 UML4ODP, and reviewing proposed MFI Part-6 Registration with MDR Part-6 Registration

• ISO/IEC 19763-3 MFI for Ontology Registration

• 32N1961A-Abstract_19763-3_2010.doc 28-Jan-2010 32N1961G-G11_Explanitory_Report-FDIS 19763-3.doc 24-Jan-2010 32N1961T-text_for_ballot-FDIS_19763-3.pdf 24-Jan-2010 32N1962-disp_of_comments-FCD_19763-3.pdf 24-Jan-2010 32N1962-disp_of_comments-FCD_19763-3.xls 22-Jan-2010

ISO/IE 24706 Update

• ISO/IEC 24706-1 Disapproved • This part of ISO/IEC 24706 specifies a set of data elements needed to describe

and register technical standards or other specification documents. The set of data elements and relationships is known as a Standards Registry.

• 32N1970-SummVoting-N1918-FCD_24706-1.doc 08-Feb-2010

• 32N1970-Summ_Voting-N1918-FCD_24706-1.pdf 08-Feb-2010

ISO/IEC 14567 Update

• ISO/IEC 14567 Information Technology – Notation of Format for Data Elements

• 32N1931G-G11_Explanitory_Report-FDIS 14957.doc 11-Nov-2009 32N1931T-text_for_ballot-FDIS_14957.pdf 11-Nov-2009 32N1932-disp_of_comments-N1678-FCD_14957.pdf11-Nov-2009

ISO/IEC 11179 Update

• 11179 (MDR) possible resolutions for project splits to create new Eds of all Parts to align with 11179-3 Ed 3

• ISO/IEC 11179-3 ED 3 MDR out for ballot comments due by

• 32N1983B-ballot_for_FCD_11179-3.doc 30-Mar-2010 32N1983Ta-Text-for-ballot-FCD_11179-3.pdf 07-Apr-2010 32N1984-disp_of_comments-CD_11179-3.pdf 30-Mar-2010

Metadata and Distributed Computing: An Opportunity and a Challenge

May 24, 2010

Denise Warzel

Associate Director

National Cancer Institute Center for Biomedical Informatics and Information Technology

WG2 Convenor

“If You Don't Know Where You're Going, Any Road Will Get You There.”

-Lewis Carroll: Alice’s Adventures in Wonderland, 1865

Outline

• This talk will highlight use cases for metadata in services oriented computing: SOA cloud computing and web services and discuss the opportunities and challenges for metamodels and metadata registries to support interoperability

This talk is intended to:

• Review ISO MDR and MFI Role/Purpose

• Depict a few use cases of Metadata and Data

• Set the stage for working group meetings and potential collaboration

This talk is NOT:

• A technical overview of WG2 program of work or particular ISO standards

MOTIVATION AND BACKGROUNDNCI Perspective

Motivation and Background: • Q: “Why do we have to worry about interoperability? Why can't we just build stuff that works?

The internet works. The internet is human-to-human interoperability, not machine-to-machine.”• A: Future semantic infrastructure for working interoperability must:

• improve human<->machine interactions providing the ability to “query over open, and dynamic collections of heterogeneous and distributed information sources”

• Help finding reusable stuff BETTER SEARCHES • Reduce Data/transforming Data into Information, Knowledge

• Make use of semantic web technologies• Humans will build systems

• Cost and accuracy of systems dependent on availability of metadata• Support agile programming techniques

Specifications: information model, data dictionary, service descriptions, etc.• Requires a formalism to enable human <–> machine representation and interpretation• Enable discovery of reusable and interoperable services over the web (sSOA and Cloud Computing)• Google search 4/24/2010: “Results 1 - 10 of about 20,800,000 for lung cancer” (10/2009 20,600,000 –

Increase of + 200,000 in 6 months)

More results ≠ better

1“UML as an Ontology Modelling Language”, Cranefield and Purvis

The Metadata and Data (Kevin to talk more about this)

Valid Values

…

Cyclooxygenase Inhibitor

Common Data Element(Agent Name CTEP Agent Names)

Object Class(Agent)

NCI ThesaurusConcept ID = C1708

Property(Name)

NCI ThesaurusConcept ID = C42614

Data ElementConcept

(Agent Name)

Concept IDs:C1708 C42614

Value Domain(CTEP Agent

Names)

Representation(Name)

NCI ThesaurusConcept ID = C42614

NCI ThesaurusConcept ID = C1323

Doxercalciferol

NCI ThesaurusConcept ID = C2645

Taxol

NCI ThesaurusConcept ID = C1411

Ursodiol

NCI ThesaurusConcept ID = C1818

Data Item

MetaMODELMetaMODEL

Metadata Components

Metadata Ontology

MFI-3

Data ModelWG3

Metadata ModelsMFI-2

Metadata Data Elements

MDR-3

ADOPTION OF SOANCI Perspective

Adoption of SOA

Why SOA?

Define, design, develop, and deploy software components as reusable, sharable business capabilities

• Rather than “monolithic” applications where all business activities are “under the covers” Design reusable that are “Fit for Purpose”

• Reduce redundancy and duplication of effort

• LEGO® analogy

The LEGO® Analogy

NCI Enterprise Services

Applications

Semantically-aware Services Oriented Architecture (sSOA)

• Semantically-aware Service Oriented Architecture (sSOA) supports the challenges of integrating diverse classes of information distributed across a distributed, heterogeneous communities

• In addition to data integration, sSOA will enable the coordination of functionality between the various information systems

• Services will be implemented in a largely standalone fashion to allow for the rapid creation of composite applications via service marshalling or integrated with existing applications

sSOA needs to ensure working interoperability between differing systems that need to exchange specific classes of information

and/or coordinate cross-application behaviors

Challenges of Implementing sSOA

NCI Perspective

Common sSOA Requirements

• Federated Discovery Services

• Security Considerations for access/editing data

• Ability to make new semantic assertions about data

• Logging/Auditing/Provenance

• Subscriptions/Notifications

• ISO 11179 Ed 3• Ability to have referenceable items (via URI?)

• Combining Domain Specific Business logic with Generic Discover, Search and Retrieve, Compare, Assert

• E.g. Structured Eligibility Criteria “find all the patients eligible for the trial” would be an example of a composed service that is dependent on Semantic model discovery services (Generic) and Domain specific ‘find patients’ services (Business logic)

• Identification/Handling Duplication

• Lifecycle support for semantic content (statuses, versioning, etc)

• Linked Data - exposing, sharing, and connecting data via dereferenceable URIs on the Web.

A little more on “Linked Data”

Tim Berners-Lee outlined four principles of Linked Data in his Design Issues: Linked Data note, paraphrased along the following lines:

• Use URIs to identify things.

• Use HTTP URIs so that these things can be referred to and looked up ("dereference") by people and user agents.

• Provide useful information (i.e., a structured description — metadata) about the thing when its URI is dereferenced, using standard formats such as RDF-XML.

• Include links to other, related URIs in the exposed data to improve discovery of other related information on the Web.

Source: http://en.wikipedia.org/wiki/Linked_Data

A little more on “Linked Data” and standards

Tim Berners-Lee outlined four principles of Linked Data in his Design Issues: Linked Data note, paraphrased along the following lines:

• Use URIs to identify things.

• Use HTTP URIs so that these things can be referred to and looked up ("dereference") by people and user agents.

• Provide useful information (i.e., a structured description — metadata e.g. 11179) about the thing when its URI is dereferenced, using standard formats such as RDF-XML (representations of 11179 content).

• Include links to other, related URIs in the exposed data to improve discovery of other related information on the Web. (e.g. W3C SAWSDL? VOID?)

Other SC32 Standards?

Source: http://en.wikipedia.org/wiki/Linked_Data

22

What next? ISO/IEC 10746 RM-ODP Viewpoints—Support different concerns

““System”System”

Consider: These 3 represent the service Semantics

ISO 19763 Standards?

• Core “Model” Metamodel for describing domain specific ontology for Services

• Role of entity requesting the service

• Goal of entity requesting the service

• Process performed by the service• Business Activity, pre and post conditions

• Service (on or more business activites to complete a unit of work)

How will we implement/integrate these standards?

• Domain Specific Standards use to develop content, instances of the metamodels and metadata

• Generic ISO Standards used to represent the Domain Specific/Content standards so they are comparable across domains

SAP says: (Jake don’t choke)

• ISO 15000-5:2006 - CCTS (Core Component Technical Specification) Version 2.2 (draft version) based on 11179 Addresses:

• the lack of information interoperability between applications

• the development and representation of context driven approach for developing reusable business data building blocks in a syntax independent business semantic manner

• reuse regardless of domain or purpose of use (generic capabilities)

• tightening the rules for creation, naming, and management of the various core component constructs

SC32 Data Management and Interchange

SC32 WG2 ISO 11179 MDR describe the semantics of data (information models) held in databases

SC32 WG3/WG4 SQL describe tables that hold the data (instances of data)

SC7 WG2 ISO 19793 UML4-ODP describe the software systems that will create and use the data held in databases including business policies, contracts, information models, computational (behavioral framework)

SC32 WG2 ISO 19763 MFI describe the Business Capability Services (behavioral framework)

Abstraction

EXAMPLES OF DOMAIN SERVICES FOR SSOA

NCI Perspective

Kite and User Goal Level Use Cases for Semantic Metadata Registries

• Federated discovery, searching, and data aggregation

• User can query for all pre-cancerous biospecimens from caTissue instances like those at Washington University, Thomas Jefferson University, and Holden Comprehensive Cancer Center.

• User can query across all services for data that matches the search criteria even if the schema exposed by the service at each location has a slightly different syntax

• User can identify the sample obtained for Glioblastoma multiforme (GBM) and the corresponding CT image information. This query can be performed by querying across caTissue and NBIA.

• User can discover attributes that can be used to join tables in different services and perform a query to join them.

• User can find out if a sample used in an expression profiling experiment is available for a SNP analysis experiment. This query can be performed by querying across caTissue and caArray.

• Similar to above

(Continued Kite and User Goal Level Use Cases)

• Data element equivalence and discovery

• Find all malignant breast cancer tumors, return all tissues that have site "breast" or auxiliary site is a subtype of "breast" across different tissue banking systems, even if these have been coded differently in different systems

• User can discover matches to search criteria across different services even if coded differently as long as there is synonymy in the related metadata

• Find a standard data element that matches your local data element, assert that these are the same

• User can discover matches based on metadata structures (ISO 11179 data element) and make new assertions about equivalence

• Find all prostate cancer specimens, return all specimens with a clinical diagnosis of "prostate cancer" or related terms (query expansion based on ontology)

• Similar to above only this one makes reference to use of an ontology along with the metadata semantic description

(Continued Kite and User Goal Level Use Cases)

• Semantic Metadata Driven Workflow discovery and authoring

• When dragging services onto the authoring tool dashboard, these services should be automatically "piped" together where applicable (i.e. when output from 1 service maps to the input of another service).

• Service Metadata in standard structures can be used to test whether outputs of one service match inputs to another

• In cases where services cannot be directly piped together, the tool should help identify shim/transformation services that can be used. If there do not exist shims to assist in piping services together, the authoring tool should help (automatically) generate shim services based on the semantic requirements registered in the metadata registry.

• Similar to above, supports discovery

• The ability to describe a published paper as a "metadata description" and "SOP"... and then use that metadata for repeating the experiment, or for search/discovery/authoring new workflows, capturing any new steps or features of the new workflow + the original SOP in the metadata registry.

How Does sSOA apply here? More on workflows

Hypothesis-Driven Workflows

(cont.)

Multiple Workflows Converted to Services

Internet

(cont.)

New Hypothesis and Workflow Using Existing Services

Internet

A service is an implementation of a well-defined functionality that can be discovered and reused in different applications or business processes

Generic Services and sSOA

Application 1

caGrid

Application 2 Application 3 Application 4

Service A

Internet

Domain Specific Composed Services: Correlation Organization Person PAtient core services (COPPA)

caAERS

caGrid

C

PSC C3PRIntegration Hub

O P PA …

NCI ClinicalTrialsReportingProgram

…

caBIG® Clinical Trials Suite

NCI Enterprise Services

Internet

EXAMPLES OF GENERIC SERVICES FOR SSOA

Metadata Services

User Goal/Sea level use cases for sSOA Distributed, federated metadata and model repositories and operations

Model Services• CRUD

• Record• New Assertions, Rules/Results

Queries/Assertions

• Update• Version• Retire

• Validate• Discover

• Federated Query• Find related metadata

• Compare/Merge/Assert• Harmonize

• Graphical/Visualize• Delete/Unregister• Subscribe/Notify

• History information

• Transform • Transformation

• Reuse• (vs Copy)

• Extend• Extend Class• Add Relationships• Extend Permissible Values

• Constrain• Sub-Set Classes• Sub-Set Permissible Values

• Find related services

• Find related documentation

User Goal/Sea level use cases for sSOA distributed, federated metadata and model repositories and operations

Metadata ServicesData Element

• “CRUD”• Validate • Discover Reusable Content

• Usage Information• Search/Query

• Federated• Compare/Merge/Assertion• Create New from Existing• Discover related Models• Discover related metadata items• Discover related Services• Discover related Rules• Discover related Forms

Value Domain• Same as Data Element +• Create Sub-set (Constrain)• Extend (create new from existing)• Semantic Transformations

• Explicitly based on mapping e.g. to the same Value Meaning

• Syntactic Transformations• Source representation Target

representation

Data Element Concept• Same as Data Element +• Discover related DEs• Discover Value Sets• Create Data Element from existing

DEC and VD• Create new DEC from existing DEC

User Goal/Sea level Use cases for distributed, federated metadata repositories and model repositories and operations

Registry-Registry Services• Import Content

• Export Content

• Update Content

• Search

• Submit Content (Submitter)

• Register Content (Registrar)

• Update Registration• Registration Service (Kite Level)

• …

General Services• Annotate with semantic

concepts

• Subscribe to changes

• Reuse (Classify/Categorize)

How would you measure success?

• JTC1 SCs working together to address these challenges

• Availability of internet users to discover and integrate data from multiple services based only on the commonly deployed service metadata• Usable by an end user without extra programming

• Commonly used Software Engineering and Implementation tools make adoption transparent

• Implementation of interoperable services by vendors

• Enterprises are able to achieve Working interoperability across business domains

• Creating applications quickly from existing component parts

Potential Benefits of SC32 Data Management and Interchange WGs Collaboration:

Toward the ultimate goal of:

• An information and software services infrastructure that supports the goals for communities that wish to interoperate over the internet