®
IBM Software Group
Linked Data as an Application Integration
Architecture
Martin Nally, IBM Fellow, Rational CTO
© IBM Corporation
IBM Software Group | Rational software
My story
1. A case study – how we use linked data as an integration architecture
2. The gaps and problems we found, with proposed solutions
3. A call to action
2
IBM Software Group | Rational software
Why do we need a new integration architecture?
 Case study - IBM Rational in 2004
 A leading provider of Application Lifecycle Management Tools
 Source Code Management (ClearCase, #1 market share)
 Defect/Change Management (ClearQuest, #1 market share)
 Requirements Management (ReqPro, #2 market share).
 …
 Mature products
 Designed and built in the ’90s
 Evolving to the web
3
IBM Software Group | Rational software
Times change – requirements change
 New Requirements
 Distribution of development teams around the world
 Want more tightly integrated end-to-end ALM solution
 Want broader solutions
 Need greater agility and lower cost of ownership
 Response
 Adapt existing products to the new environment
 Write glue code to integrate products
 Necessary, works, but has limitations
4
IBM Software Group | Rational software
Traditional Tool Integration. Ouch.

N2 possible point-to-point
connections
Limited coverage

Closed APIs
Vendor lock-in

Tight Coupling
Dependence on API details

Lockstep upgrades
Version incompatibilities

No place for cross tool functions
Global Query
Account Mgmt
Time for a new
architecture …
5
IBM Software Group | Rational software
What integration functions are really needed?
 Create a link between ‘artifacts’ in different tools
 E.g Test links to requirement. Development task links to requirement
 Within one tool, create an “artifact’ in another
 E.g. Within a Test tool, log a defect using test results
 Across tools, share common concepts
 People form a team to work on a project to produce the next release of a product
 Be able to query across information in multiple tools, e.g.
 Show me all the critical requirements that don’t have test cases
 Show me al the new test cases I should run on last night’s build
6
IBM Software Group | Rational software
Alternative Architecture - 1
Ent Arch
Requirements
Bus Proc
Model
Software &
Solution
Architecture
Development
Test
Repository
7
IBM Software Group | Rational software
Alternative Architecture - 1
 Standard Application Server Architecture – tried and proven
 IMS, CICS, JEE, Tuxedo, Encina, …
 Tried for tools many times before
 AD/Cycle, PCTE, …
 Pros:
 Effective when the function can be delivered and deployed in a highlycoordinated fashion
 Cons:
 Requires schema coordination across tools
 Does not provide a mechanism for integrating “existing ” or “foreign” tools
 Requires “central planning” by consumer (deploy 1 database) as well as
provider (integrate tools to a single schema)
Does not support an open ecosystem well
8
IBM Software Group | Rational software
Alternative Architecture - 2
Ent Arch
Bus Proc
Model
Development
“ESB – Enterprise Service Bus”
Requirements
Software &
Solution
Architecture
Test
9
IBM Software Group | Rational software
Alternative Architecture - 2
 Pros:
 Allows integration of existing applications
 Some of the process implemented in the “bus”
 Cons:
 Multiple copies of data in different tools
 Sensitive to changes in data model in any participating tool
 Sensitive to changes in the process
 Requires a lot of centralized implementation
Many of the disadvantages of traditional approach
Central point of control is bottleneck
10
IBM Software Group | Rational software
Problem
 We’ve been stuck with these problems and solutions for 20 years
 Is Martin going to claim he’s invented a new way to write tools that fixes this?
 Yes, and No
 We think there is another way
 We didn’t invent it …
…Al Gore did
11
IBM Software Group | Rational software
Linked Lifecycle Data
Data Integration for the 21st Century
 Inspired by Internet principles, implemented
with Internet technologies: simple interfaces for
exchange of resources
If the entire Web can connect like this, would the
same idea work for ALM?
 Loosely coupled: everything is a “resource”
linked together with URLs
 Technology neutral: treats all implementations
equally
 Minimalist: defines no more than necessary for
exchange of resources
 Incremental: deliver value now, add more value
over time
 Openly published standards: free to implement
and irrevocable
12
IBM Software Group | Rational software
Alternative Architecture - 3
 Pros:
 Allows integration of existing applications
 Supports open, federated data model
 Does not require data copying
 Supports independent evolution of products
 Cons:
Unproven (for our domain, at least)
Big paradigm shift (for us)
Lots of invention required
The rest of this talk focuses on our experiences
13
IBM Software Group | Rational software
OSLC and Open Community
http://open-services.org
Open Services for Lifecycle Collaboration
open community. open interfaces. open possibilities
 Minimalist/additive approach
 Not a “complete” definition for a given area
 Scenario driven scope
 Co-evolve spec and implementations
Identify
Scenarios
Call it a
spec
Iterate on
working
drafts
 Open participation around active core group
Gain technical
consensus,
collect nonassert
statements
14
IBM Software Group | Rational software
OSLC Community
 Eleven workgroups operating at openservices.net
Domain focused workgroups (e.g. CM,
QM, RM)
Common issues and patterns (Core)
Solution oriented workgroups (e.g.
PLM/ALM)
 Range of interests, expertise,
involvement
400+ registered community members (up
from 70 people at RSC 2009)
Individuals from 34+ different companies
have participated in OSLC workgroups
(up from 5 companies at RSC 2009)
Accenture
APG
Black Duck
Boeing
BSD Group
Citigroup
EADS
Emphasys Group
Empulsys
Fokus Fraunhofer
Galorath
General Motors
Health Care Services Corp
IBM
Institut TELECOM
Integrate Systems
 Worthy of note
PLM/ALM Workgroup

Siemens leadership
Open source



Eclipse Lyo Project Proposal
Eclipse Mylyn project restructuring and
positioning of OSLC
Mantis, Forges
Customer and integrator involvement

GM, Northrop Grumman, Tieto, Integrate
Systems
15
Lender Processing Services
Northrop Grumman
Oracle
QSM
Rally Software
Ravenflow
Shell
Siemens
Sogeti
SourceGear/Teamprise
State Street
Tasktop (Eclipse Mylyn)
Thales
Tieto
TOPIC Embedded Systems
UrbanCode
WebLayers
15
IBM Software Group | Rational software
Problem 0 - culture
 People have been brainwashed for years
 XML (don’t get me started)
 Operations, parameters
 SPARQL Protocol for RDF is a typical example of the mess
 Relational schemas and Object-oriented class models
 Let’s invent new media types
 Cling to closed world assumptions like to a security blanket
 Desktop user interfaces
16
IBM Software Group | Rational software
Problem 1 – creating data on the web
 It seems obvious that you use POST to create, but what do you POST to?
 How do I find the things that already exist?
17
IBM Software Group | Rational software
Proposed Solution 1 – “Atom Publishing for LD”
 A resource to which you POST to create new things, GET to find existing
things
 Not really like APP, much simpler
 But APP saw the need, defined the use-cases, got traction
 We did not get this part right at http://open-services.net OSLC 
 Observation – the simplest way to define a collection in RDF is a set of
triples with common subject and common predicate
 That is all there is to it!
18
IBM Software Group | Rational software
Solution 1 - APP for LD example (Trig format)
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>.
@prefix bp: <open-serives.org/bp/>.
@prefix : <http://example.org/test/>.
:testCases = {
:testcases
bp:commonSubject :proj1TestCases;
bp:commonPredicate rdfs:member.
:proj1TestCases
rdfs:member :testCase1;
rdfs:member :testCase2.
}
POST to this resource to create a resource
-will also add a triple that points to the new resource
GET to find [triples that reference] existing resources
19
IBM Software Group | Rational software
Problem 2 - But it will get big; I need to paginate
 Yes, you need pagination
 APP has this
 Solution proposal:
 Define <URL>?firstPage as a resource
 Triples are a strict subset of those in URL (no change of subject)
 Define :nextPage as a predicate
20
IBM Software Group | Rational software
Solution 2 - Pagination example
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>.
@prefix bp: <open-serives.org/bp/>.
@prefix : <http://example.org/test/>.
:testCases?firstPage = {
:proj1Testcases
rdfs:member :testCase1;
rdfs:member :testCase2.
:testCases?firstPage bp:nextPage :testCases?page2 }.
:testCases?page2 = {
:proj1Testcases
rdfs:member :testCase3;
rdfs:member :testCase4.
:testCases?page2 bp:nextPage rdf:nil }.
21
IBM Software Group | Rational software
Problem 3 - I only want the metaData
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>.
@prefix bp: <open-serives.org/bp/>
@prefix : <http://example.org/test/>.
:testCases?metaData = {
:testcases
bp:commonSubject :proj1TestCases;
bp:commonPredicate rdfs:member.
}
22
IBM Software Group | Rational software
Basic Profile for LD
 Problem 2 - Now know where to put them and find them
- what does a resource look like?
1. Use URIs as names for things.
2. Use HTTP URIs so that people can look up those names.
3. When someone looks up a URI, provide useful information, using
the standards (RDF*, SPARQL).
4. Include links to other URIs so that people can discover more things.
This is a terrific foundation, but …
 In practice, people have difficult interpreting this
 It doesn’t go far enough into detail for our purposes
Proposed Solution 2 – the “Basic Profile for LD”
 Mostly just some elaboration on the 4 rules
23
IBM Software Group | Rational software
Basic Profile – 12 rules (and counting)
1. BP Resources are HTTP resources that can be created,
modified, deleted and read using standard HTTP methods.
2. BP Resources use RDF to define their states.
1. POST any resource to a BP collection, not just a BP resource
2. Use the null relative URI for the subject when POSTing
3. You can request an RDF/XML representation of any BP
Resource.
1. Turtle, please. JSON, please.
4. BP clients use Optimistic Collision Detection during update.
5. BP Resources use standard media types.
24
IBM Software Group | Rational software
Basic Profile – 12 rules (and counting)
6. BP Resources use standard vocabularies.
Prefix
Namespace URI
dc
http://purl.org/dc/terms/
rdf
http://www.w3.org/1999/02/22-rdf-syntax-ns#
rdfs
http://www.w3.org/2000/01/rdf-schema#
bp
http://open-services.net/ns/basicProfile#
xsd
http://www.w3.org/2001/XMLSchema#
25
IBM Software Group | Rational software
Basic Profile – 12 rules (and counting)
7. BP Resources use a restricted set of data types.
 Boolean: A Boolean type as specified by XSD Boolean
http://www.w3.org/2001/XMLSchema#boolean, reference: XSD Datatypes.
 DateTime: A Date and Time type as specified by XSD dateTime
http://www.w3.org/2001/XMLSchema#dateTime, reference: XSD Datatypes
 Decimal: A decimal number type as specified by XSD Decimal
http://www.w3.org/2001/XMLSchema#decimal, reference: XSD Datatypes
 Double: A double floating point number type as specified by XSD Double
http://www.w3.org/2001/XMLSchema#double, reference: XSD Datatypes.
 Float: A floating point number type as specified by XSD Float
http://www.w3.org/2001/XMLSchema#float, reference: XSD Datatypes.
 Integer: An integer number type as specified by XSD Integer
http://www.w3.org/2001/XMLSchema#integer, reference: XSD Datatypes.
 String: A string type as specified by XSD String http://www.w3.org/2001/XMLSchema#string,
reference: XSD Datatypes.
 XMLLiteral: A literal XML value
http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral
26
IBM Software Group | Rational software
Basic Profile – 12 rules (and counting)
8. BP Resources set rdf:type explicitly.
9. BP clients expect to encounter unknown properties and
content.
10.BP clients do not assume the type of a resource at the end of a
link.
11.BP servers implement simple validations for Create and
Update.
12.BP Resources use simple RDF predicates to represent links.
1. “Anchors” in addition, not instead
1. only if adding properties to links
2. Use reified RDF statements for anchors
27
IBM Software Group | Rational software
Problem 4 - Shapes (Validation, constraints)
 RDFS and OWL do inferencing
 “If it walks like a duck, and quacks like a duck, it must be a duck”
 In software (as in real life) it is also useful to go the other way
 “If it wants to be a good duck, it must walk like one and quack like one”
 Everyone in software knows this model
 Database Schemas
 Object-oriented programming class definitions
 This model is actually very important for writing practical software
 Need to support more than one perspective on a duck
 It’s a food-source, a source of insulating materials, a hazard to aircraft, a bird
Problem - this is a gap in RDF technologies
28
IBM Software Group | Rational software
Proposed Solution 4 - Shapes
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>.
@prefix bp: <open-serives.org/bp/>.
@prefix : <http://example.org/test/>.
:teatCaseShape = {
:testCaseShape
a bp:Shape;
bp:describedClass :TestCase;
pb:propertyConstraint :testsRequirementConstraint }.
:testsRequirementConstraint = {
:testsRequirementConstraint
a bp:propertyConstraint;
bp:describedProperty :testsRequirement;
bp:rangeShape :requirementShape }.
29
IBM Software Group | Rational software
Problem 4 - Updating data on the web
 You use PUT, right?
 Observation – SQL has no equivalent of PUT
 Problem 4 – PUT is good for documents, not a good idea for data
 Read-only properties
 Independent perspectives
 Proposed solution 4
 “Put” more energy into PATCH
30
IBM Software Group | Rational software
Cross-product query
 Query server independent of storage, broader scope
 Schema-less query
 Exploit RDF
 XML and XQuery were tried, did not work as well
31
IBM Software Group | Rational software
Implementing query on data not in a repository
HTTP GET/PUT/POST/DELETE
SPARQL
RDF
represent’n
Jena
RDF
represent’n
32
IBM Software Group | Rational software
UI Integration
 Problem 5:
 A test product wants to allow testers to open defects in another product
 Defects are customized – don’t know in advance what fields they have
 Defects have context (projects, components) and “meta-data” (defect states,
defect types, …)
 Option 1 – teach test tool about defects, their types, constraints,
contexts
 Option 2 – let the test tool use a UI component from the defect tool to
create the defect
 allows much simpler interface, looser coupling
 Relies on human third-party
 Problem 6 – “hover help” on links
 Solution – UI preview protocol
33
IBM Software Group | Rational software
Security
 Requirements
 Allow each product to have its own security administration and
authentication
 Provide optional central place to configure users, integrate with LDAP etc
 Allow each product to manage it’s own access control lists, pre-conditions
and post-actions
 Provide optional central federation of ACL administration
 Solutions
 For products that implement their own authentication, federate with OAuth
 For products that want to manage ACLs, provide federation
 Provide optional building blocks for new-construction that take over admin,
authentication and ACLs
34
IBM Software Group | Rational software
How do you write an Ajax UI?
 Attempt 1 – AJAX clients handle multiple-resources
 Pros:
 Responsive UI – no page refresh between resources
 Cons:
 Complex clients (1990s client-server)
 Closed system (can’t link to unexpected resources)
 Future attempt:
 Page-per-resource
 Ajax inside pages
35
IBM Software Group | Rational software
Miscellaneous architecture/design
 Tags, not folders (still some controversy)
 Relationships/predicates, not collections (still some controversy)
 Exploit RDF, eschew WebDAV
 Programmable resources (but stay RESTful)
 Back links for inverse navigation?
 Discovery
 What is userID?
 Transactions
36
IBM Software Group | Rational software
Need more help to shift to a web-centric design
Old
New
Services
REST
Files
Resources
Repositories
Linked open data
J2EE apps
Heterogeneous apps
Tightly-coupled apps
Loosely-coupled apps
Relational/XML data/Objects
RDF
SQL/XQuery
SPARQL
authentication
OAuth
Folders
Tags
WebDAV
RDF
Collections
RDF Predicates
37
IBM Software Group | Rational software
Conclusion
 Linked Data could be huge in enterprise computing
a very significant new application integration model
 Some significant gaps to close if we are to succeed
Better definition and documentation of best practices, e.g.
 12 practices of Basic Profile
 UI guidelines (resource-oriented UI)
A few important specification gaps, e.g.
 POST-able collections
 Shapes/constraints/validation
 Security
 Let the W3C take up the challenge!
38