• Since 2005; offices in Washington DC, and Boston, MA
• Global leader in semantic technology & information integration
• Strong academic partnerships in US, UK, EU, Latin America
• Extensive experience building semantic technology based solutions

• 10 years building semantic technology based solutions
• US Gov't, banking, financial, energy, health/bio, retail
• 75+ years aggregate experience in the field
• 60% of employees: advanced degrees in semtech, AI, etc.
• Application development and consulting
  
• We've built lots of applications for and with our customers
• We're good at this because we love doing it
• Lots of semtech R&D: publicly and privately funded
  
• "Research as a Service"
• Contributed code to Sesame, Jena, OWL-API
• Committer status with Sesame
• Maintain several open-source semtech Java libraries

• Comprehensive training suite
• Based on a decade of customer engagements
• Covers the full spectrum of semtech and related technologies:
• Introduction to Semantic Technologies all the way to Advanced OWL Modeling 
• Training for all products in our stack
• Consulting & Support
• For our products, naturally
• But also for open source as well
• Protege
• Sesame, Jena

• Software
• Stardog
• Pelorus
• Stardog Web Framework
• HotPlanner
• Solutions
• SPANNER - Enterprise Semantic CMS
• XACML-DL - Policy Management
• SLDC - Linked Data Portfolio Management
• POPS - Expertise Location

• NASA is a big, interesting organization
• 100k employees spread over 12 centers across US
• Have a universe of data
• But one simple, terrestrial problem; finding experts
• COTS solutions were not working
• Data required to solve problem was already in-house
• Enter POPS
• We build a model over all relavent data sources
• Created a simple web interface & some social analytics to let NASA use their own data
• Saved them $38M a year!
• Now an official W3C case study on how to use semtech

• Experience ... We've been building solutions for Enterprise Semantics for nearly a decade
• Leadership ... We've led from academia, from industry, from standardization
• Technical Excellence ... We provide the leading OWL DL reasoner, the leading RDF database, and provide unique capabilities available no where else
• Focus on Customer Success ... We're uniquely focused on helping our customers solve the hardest information problems in the world

• The leading RDF database
• Currently at 2.1.1 (2014-02-13)
  
• 54 releases, or, 1 every 3 weeks
• 5000x increase in scale, 1000x increase in performance
• Pure Java
• Community & Enterprise Editions
• Rich feature set
• Reasoning, ICV, full-text search, ACID, High Availability ...
• Focus on developer experience

• Query
• Loading lots of data not useful if you cannot query it
• Query 100M triples with a throughput of 3M+ queries/hour, 1B at 500k queries/hour and 10B at 20k queries/hour.
• BSBM, 64 concurrent clients
• Fastest SP2B benchmark results at 5M, only known completion at 25M, close to completing at 100M
• Scale
  
• Up to 50B triples/quads on modest hardware
• Load rates up to 500k triples/second
  
• That's 100M triples in 3 minutes, 1B in 30m, and 20B in 20 hrs

• First and foremost, we are developers too
• We intend for the best out-of-box experience possible
• Excellent documentation
• Easy Installation, just unzip
• Command Line
• Modeled after the Git CLI, includes autocomplete support
• Intended for devs & admins alike
• Like Java?
• Jena, Sesame, Spring, SNARL API
• Prefer something else?
• Javascript, Ruby, Python, .Net, Groovy
• Access
• HTTP via W3C standards
• SNARL

• SPARQL 1.1
• Query, Update, Graph Protocol
• Highly optimized query planner
• Optimized for analytic/BI queries
• And queries from the reasoner
• But does not sacrifice performance at low scales
• Scalable query answering
• Intermediate results get big, and fast
• Runtime analysis of heap status; flow intermediate results into direct memory, or onto disk
• Query Management

• JMX Server Monitoring
• High Availability
• Hot backup & restore
• Access & Audit logging
• Web Console build on Stardog Web Framework
• PROV support

• Transactions
• ACID
• Guarded (optionally) by ICV
• 2 Phase Commit over all database components
• RDF Index, Lucene, KB, etc.
• Automatically managed by database
• Security
• RBAC model
• Based on Apache Shiro
• R/W ACLs for access to individual databases
• Administrative controls for actions against the DBMS
• Online/offline a database, modify security settings, etc.

• Integrity Constraint Validation inside Stardog's transaction system
• Prevents data modifications that violate your integrity constraints
• Can be used in 'guard mode'
• Constraint violations abort transactions
• These violations can be explained
• Also supports 'oracle mode' aka 'middleware mode'
• Check whether a given graph & constraints are valid
• Executed outside a transaction
• Constraints expressed in SPARQL, OWL, SWRL, or Stardog Rules
• High-level declarative languages make it easy to author simple constraints, possible to author complex

Supervisor subClassOf supervises some Employee 

• If you are using ICV
• You may not understand why a constraint was violated
• Or you might want to show the reason to a user
• Explanations
• Tells you why the violation occurred
• Gives you the proof used to derive the violation
• Shows the exact data which violated the constraint

Supervisor subClassOf supervises some Employee
Alice a Supervisor 

VIOLATED Supervisor subClassOf (supervises some Employee)
   ASSERTED     Alice a Supervisor
   NOT_INFERRED x a Employee
                Alice supervises x 

• Embeds Lucene
• Automatically managed by the database as if another RDF index
• Enables full-text search over your RDF
• Literal values indexed by Lucene
• Use the Lucene query language to search
• Seamless integration with SPARQL
• Results from searching Lucene as SPARQL results
• Also available directly via SNARL API

• Make implicit information explicit
• Implicit in the schema, or the data, or both
• Represent domain knowledge in a formal, declarative model
• Called an ontology
• Like UML, but with formal semantics
• W3C specification called OWL, the Web Ontology Language
• Reasoners consume ontologies to derive new information
• Answer queries, find inconsistencies

• Complex, but manageable
• OWL divided into profiles with less expressivity but better computational properties

• Unmatched OWL Support
• All OWL2 profiles (QL, RL, EL, DL) and Stardog profile (SL)
• Caveats no equality reasoning, no datatype reasoning, DL over schema only
• Query time reasoning
• No write performance penalty
• Pay for what you use
• Reasoning services
• Consistency checking, Satisfiability
• Explanations
• Like ICV, you can get explanations for inferences

• Stardog supports SWRL a W3C standard
• Part of the SL profile
• You can use SWRL/RDF syntax, but it's terrible
• Much easier to use Stardog Rules
• If-Then style based on SPARQL Syntax

PREFIX :
PREFIX math: 
IF {
     ?c a :Circle ;
          :radius ?r
     BIND (math:pi() * math:pow(?r, 2) AS ?area)
}
THEN {
    ?c :area ?area
}

• Complex organizations face challenging IT problems
• Enterprise Semantics: model-driven information integration
• Provide solutions for decision support, analysis, BI
  
• Use all facets of semantic technology, AI, knowledge representation, etc.
• Smart data, not big data
• Scale is not a condition for utility
• Not all problems solved by adding more data

• Coming in 2.2
• Take advantage of the fact that RDF is a graph
• Graph measures: in-degree, out-degree, PageRank, betweenness centrality
• Clustering: weakly/strongly connected components, Bron-Kerbosch clique finding
• Path finding: BFS and shortest path
• Seamless SPARQL integration and SNARL API 
  

• Focus on Web part of Semantic Web
• Organizations don't always have semtech experts at hand
• Provide a framework that abstracts away those details
• Stick to well-known web technologies
• HTML, CSS, Javascript, JSON
• backbone.js as the model layer, SPARQLRoutes as middleware
• Goal is to provide good out of the box capabilities
• Faceted-browsing, semantic search, REST, CRUD, etc.
• Minimal configuration or programming required, just add data
• Provide basis for quickly building web apps based on semtech
• Aimed at data discovery/exploration use-cases
• Stardog Web Console built on this technology

• Built on Backbone.js
• Using a Yeoman generator to bootstrap the application
• SPARQL Routes
• RESTful middleware for SPARQL without all the query 
• Data exchange is JSON-LD
• You don't need to know RDF or fumble with RDF serializations
• Templates based on Handlebars
• Visualizations in D3

• For example, enforcing security (ACLs)
• Can Bob access Resource1?

Bob is-a Admin OR Bob created Resource1 OR (Bob hasRole ?r AND ?r canAccess Resource1) OR ... 

• Hard to maintain, encoded domain knowledge into the query
• Can leverage reasoning to simplify

Bob canAccess Resource1 

• More concise and maintainable
• Reasoner handles the implementing logic transparently