why your next product should be semantic by dr. david wood

@prototypodavid@3RoundStones.com

David Wood, PhD

Why Your Next Product Should Be

Semantic

Critical Business Benefits

• Better speed to market

• Lower running costs

• Increased market opportunities

M A N N I N G

David WoodMarsha ZaidmanLuke RuthWITH Michael Hausenblas

FOREWORD BY Tim Berners-Lee

Structured data on the Web

2 Door P

rizes

YouTube HDTV

watch videoswatch Better

videos

Publish videos

Share videos

Rate videos

Discuss videos

Linked Data RDBMS

Use data Use data

Publish data

Share data

Rate data

Discuss data

Data is easier and faster to change than code

“(Software development) cost estimation literally boils down to

"how long and how many people?"” -- Philippe Kruchten

0

10

20

30

40

50

60

70

Retail

Bars

Building

Food

Service

Accou

nting

Software

(trad

)

Legal/

Tax

Health

care

Educati

on

Childca

re

Software

(remote

)

SHRM 2008US Census 2007 Survey 2014

Payr

oll,

% o

f Tot

al

0

10

20

30

40

50

60

70

Retail

Bars

Building

Food

Service

Accou

nting

Software

(trad

)

Legal/

Tax

Health

care

Educati

on

Childca

re

Software

(remote

)

SHRM 2008US Census 2007 Survey 2014

Payr

oll,

% o

f Tot

al

“Computer Science is the only discipline in which we view adding a

new wing to a building as being maintenance.”

-- Jim Horning

class Language < ActiveRecord::Base ! has_many :examinations has_many :project_languages has_many :projects, :through => :project_languages ! attr_accessible :source, :target, :price ! validates :source, :target, :price, :presence => true ! def self.source_languages Language.find_by_sql("select * from languages group by source").map{|lang| [lang.source, lang.source] } end ! def self.target_languages Language.select(:target).uniq.map{|lang| [lang.target, lang.target]} end !end

SQL embedded in code

class Language < ActiveRecord::Base ! has_many :examinations has_many :project_languages has_many :projects, :through => :project_languages ! attr_accessible :source, :target, :price ! validates :source, :target, :price, :presence => true ! def self.source_languages Language.find_by_sql("select * from languages group by source").map{|lang| [lang.source, lang.source] } end ! def self.target_languages Language.select(:target).uniq.map{|lang| [lang.target, lang.target]} end !end

SQL embedded in code

SQL

SQL generators

client = Client.find(1, 10)

equivalent to: SELECT * FROM clients WHERE (clients.id IN (1,10))

SQL generators

client = Client.find(1, 10)

equivalent to: SELECT * FROM clients WHERE (clients.id IN (1,10))

SQL!generator

SQL generators

client = Client.find(1, 10)

equivalent to: SELECT * FROM clients WHERE (clients.id IN (1,10))

SQL!generator

That’s good, right?

SQL generators

class Amenity < ActiveRecord::Base has_many :amenity_lists ! named_scope :available, lambda { |start_date, end_date| { :select => 'amenities.*', :order => 'id', :conditions => [" amenities.id NOT IN (! SELECT aa.id from amenities aa, amenity_lists al WHERE! amenities.id = al.amenity_id AND! (! (? BETWEEN al.checkin AND al.checkout) OR! (? BETWEEN al.checkin AND al.checkout)! )! )", start_date, end_date ] } } end

Why?

Why?

• It is the tight coupling of DB+schema that causes tight code integration.

Back to the future

q = """SELECT ?x ?z ?y! WHERE {! course:%s course:termName ?x;! ?s ?t.! ?s ?d ?z.! ?t course:termName ?y.! FILTER (regex(?z,'[^a-z0-9]','i') && isLiteral(?z) )! }ORDER BY ASC(?s) """%var_value

schema

ontologyinference

schema

SPARQL

“Perfection is achieved, not when there is nothing left to add, but when

there is nothing left to remove.” -- Antoine de Saint-Exupéry

Next

schema

ontologyApplication relationships

Data structure

Next

schema

ontology

Next

inference

schema

ontology

Next

inference

schema

ontology

SPARQL!generator

Next

generic code

inference

schema

ontology

SPARQL!generator

SQL

SQL

1960s

1970s

1980s

1990s

2000s

2010s

SQL!generator

schema

ontologyinference

schema

SPARQL

SPARQL!generator

schema

ontology

inference

generic code

inference

schema

ontology

Bonobos!

Bonobos!

Animal Pants

Label Label

Comment Comment

Depiction Depiction

Order page

Color

Gender pronoun

Address

Label

Comment

Depiction

Depiction

Label

Comment

Gender pronoun Address

Color Order page

Lack of Context

Required Context

carrelet@fr

plaice@en_gb sand dab@en_us

lemon dab@en_us

lemon sole@en_us

carrelet@fr

plaice@en_gb sand dab@en_us

lemon dab@en_us

lemon sole@en_us

limande@fr

calimande@fr

plie@fr

Dr. Chris Welty

“Semantic Technology is not an end in itself, it is a means to an end, a set of tools and techniques for supporting

modern data analytics. It goes alongside other essential technologies

like machine learning, databases, information retrieval, etc.”

IBM Watson Team

M A N N I N G

David WoodMarsha ZaidmanLuke RuthWITH Michael Hausenblas

FOREWORD BY Tim Berners-Lee

Structured data on the Web

2 Door P

rizes

@prototypodavid@3RoundStones.com

David Wood, PhD

Why Your Next Product Should Be

Semantic

This work is Copyright © 2011 3 Round Stones Inc. It is licensed under the Creative Commons Attribution 3.0 Unported LicenseFull details at: http://creativecommons.org/licenses/by/3.0/

You are free:

to Share — to copy, distribute and transmit the work

to Remix — to adapt the work

Under the following conditions:Attribution. You must attribute the work in the manner specified by the author or licensor (but not in any way that suggests that they endorse you or your use of the work).

Share Alike. If you alter, transform, or build upon this work, you may distribute the resulting work only under the same or similar license to this one.