An Introduction to ccREL and
The Use of Semantic Technology at Creative Commons

Nathan R. Yergler, Creative Commons
Ben Adida, Creative Commons

Agenda

ccREL: the Creative Commons Rights Expression Language
Expressing ccREL with RDFa
Semantic Infrastructure @ CC

share, reuse, and remix — legally


<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
         xmlns="http://www.w3.org/1999/xhtml/vocab#">
    <rdf:Description rdf:about="http://www.lessig.org/blog/">
        <license rdf:resource="http://creativecommons.org/licenses/by/3.0/" />
    </rdf:Description>
</rdf:RDF>

Why Machines?

Programs should be able to answer simple questions about licensed works.

What is the work's license?
Commercial use?
Derivative works?
How do we attribute the work?

Previous Recommendation

Pervasive Problems

Opaque to parsers
Opaque to humans
Easy to screw up
Overly verbose

CC Rights Expression Language

"Compared to the previous recommendation, ccREL is intended
to be both easier for content creators and publishers to provide,
and more convenient for user communities and tool builders
to consume, extend, and redistribute."

Work Properties

License Properties

Work Properties

title
work license
attribution name
attribution URL
source work
CC+ (more permissions)

License Properties

permits
requires
prohibits
attribution URL
jurisdiction
deprecation date
legal code URL

Principles for Publishing ccREL in HTML

independence + extensibility
Don't Repeat Yourself (DRY)
Visual Locality
Remix Friendliness

Chasm Between Human and Data Webs [Semweb]

2 published documents, 2 document types

So much data on the web.... [Social Network]

So much data on the web.... [Craigslist]

So much data on the web.... [Flickr]

So much data on the web.... [SemTech2008 Program]

The Chasm

A significant gap between what browsers see and what humans see

Bridging the Human and Data Webs [Correspondence]

Bridging the Human and Data Webs [A Single DRY Format]

Change the visual, change the machine-readable

distributed under a 
<a href="http://creativecommons.org/licenses/by/3.0/">CC License</a>

To:

distributed under a 
<a href="http://creativecommons.org/licenses/by-nc/3.0/">CC License</a>

The Machine Readers should Follow

Bridging the Human and Data Webs [A Single DRY Format]

Bridging the Human and Data Webs [Extensibility & Modularity]

The Flickr Photostream

Bridging the Human and Data Webs [Extensibility & Modularity]

Flickr used by a Newsreader

Bridging the Human and Data Webs [Extensibility & Modularity]

Flickr used by a Digital Photo Frame

Bridging the Human and Data Webs [Extensibility & Modularity]

Flickr used by an Asset Management Program

A Link with Flavor

distributed under a
<a href="http://creativecommons.org/licenses/by/3.0/">
    Creative Commons License
</a>

A Link with Flavor

distributed under a
<a rel="license" href="http://creativecommons.org/licenses/by/3.0/">
    Creative Commons License
</a>

Text with Flavor

<h2>The Trouble with Bob</h2>
<h3>Alice</h3>

Text with Flavor

<h2 property="dc:title">The Trouble with Bob</h2>
<h3 property="dc:creator">Alice</h3>

why dc:title, why not just title?
which meaning of "title"? Article title, job title, real estate title?
license is a reserved HTML keyword, but title is not.
we must "import" this concept from somewhere.
the Dublin Core vocabulary: http://purl.org/dc/elements/1.1/
concepts including: title, creator, copyright, etc.

<div xmlns:dc="http://purl.org/dc/elements/1.1/">
    ....
</div>

Text with Flavor

<div xmlns:dc="http://purl.org/dc/elements/1.1/">
    <h2 property="dc:title">The Trouble with Bob</h2>
    <h3 property="dc:creator">Alice</h3>
</div>

A Few Articles

<div ...>
    <h2 property="dc:title">The Trouble with Bob</h2>
    <h3 property="dc:creator">Alice</h3>
</div>

...

<div ...>
    <h2 property="dc:title">Jo's Barbecue</h2>
    <h3 property="dc:creator">Eve</h3>
</div>

A Few Articles

<div about="/alice/posts/trouble_with_bob"...>
    <h2 property="dc:title">The Trouble with Bob</h2>
    <h3 property="dc:creator">Alice</h3>
</div>

...

<div about="/alice/posts/jos_barbecue"...>
    <h2 property="dc:title">Jo's Barbecue</h2>
    <h3 property="dc:creator">Eve</h3>
</div>

Image within an Article

<div about="/alice/posts/trouble_with_bob"...>
    <h2 property="dc:title">The Trouble with Bob</h2>
    <h3 property="dc:creator">Alice</h3>
</div>

Image within an Article

<div about="/alice/posts/trouble_with_bob"...>
    <h2 property="dc:title">The Trouble with Bob</h2>
    <h3 property="dc:creator">Alice</h3>

    <p>The trouble with Bob is that he takes
    much better photos than I do:</p>

    <div>
        <img src="/bob/photos/sunset.jpg" />
        <span>Beautiful Sunset</span>
        by <a href="/bob/">Bob</a>
    </div>
</div>

Image within an Article

<div about="/alice/posts/trouble_with_bob"...>
    <h2 property="dc:title">The Trouble with Bob</h2>
    <h3 property="dc:creator">Alice</h3>
    
    <p>The trouble with Bob is that he takes
    much better photos than I do:</p>
    
    <div about="/bob/photos/sunset.jpg">
        <img src="/bob/photos/sunset.jpg" />
        <span property="dc:title">Beautiful Sunset</span>
        by <a rel="dc:creator" href="/bob/">Bob</a>
    </div>
</div>

Image within an Article

The power of `about`

Adding Fields: Digg

<div about="/articles/cheneys_glasses">
    <a rel="dc:source" href=".." property="dc:title">
      Enhancement of Reflection in Cheney's Glasses
    </a>
    <h3 property="dc:creator">Rick</h3>
    
    1957 diggs
    <a href="up?article_id=42">up</a>
    <a href="down?article_id=42">down</a>
    <a href="bury?article_id=42">bury</a>
</div>

Adding Fields: Digg

<div about="/articles/cheneys_glasses"
     xmlns:digg="http://digg.com/vocab/">
    <a rel="dc:source" href=".." property="dc:title">
      Enhancement of Reflection in Cheney's Glasses
    </a>
    <h3 property="dc:creator">Rick</h3>
    
    1957 diggs
    <a href="up?article_id=42">up</a>
    <a href="down?article_id=42">down</a>
    <a href="bury?article_id=42">bury</a>
</div>

Adding Fields: Digg

<div about="/articles/cheneys_glasses"
     xmlns:digg="http://digg.com/vocab/">
    <a rel="dc:source" href=".." property="dc:title">
      Enhancement of Reflection in Cheney's Glasses
    </a>
    <h3 property="dc:creator">Rick</h3>
    
    <span property="digg:score">1957</span> diggs
    <a href="up?article_id=42">up</a>
    <a href="down?article_id=42">down</a>
    <a href="bury?article_id=42">bury</a>
</div>

Adding Fields: Digg

<div about="/articles/cheneys_glasses"
     xmlns:digg="http://digg.com/vocab/">
    <a rel="dc:source" href=".." property="dc:title">
      Enhancement of Reflection in Cheney's Glasses
    </a>
    <h3 property="dc:creator">Rick</h3>
    
    <span property="digg:score">1957</span> diggs
    <a rel="digg:up" href="up?article_id=42">up</a>
    <a href="down?article_id=42">down</a>
    <a href="bury?article_id=42">bury</a>
</div>

Adding Fields: Digg

<div about="/articles/cheneys_glasses"
     xmlns:digg="http://digg.com/vocab/">
    <a rel="dc:source" href=".." property="dc:title">
      Enhancement of Reflection in Cheney's Glasses
    </a>
    <h3 property="dc:creator">Rick</h3>
    
    <span property="digg:score">1957</span> diggs
    <a rel="digg:up" href="up?article_id=42">up</a>
    <a rel="digg:down" href="down?article_id=42">down</a>
    <a rel="digg:bury" href="bury?article_id=42">bury</a>
</div>

Adding Fields: Flickr

<div about="/photos/timbl/12345">
    <img src="/photos/timbl/12345_medium.jpg" />
    <span property="dc:title">
      Me and Vint Cert, partying Beijing style
    </span>
    <h3 property="dc:creator">DanC</h3>
    
    Other Sizes:
    <a href="12345_small.jpg">small, 300x200</a>
    <a href="12345_big.jpg">big, 800x600</a>
    
    Taken with:
    <a href="...">Canon Rebel XTi</a>
</div>

Adding Fields: Flickr

<div about="/photos/timbl/12345" xmlns:photo="...">
    <img src="/photos/timbl/12345_medium.jpg" />
    <span property="dc:title">
      Me and Vint Cert, partying Beijing style
    </span>
    <h3 property="dc:creator">DanC</h3>
    
    Other Sizes:
    <a href="12345_small.jpg">small, 300x200</a>
    <a href="12345_big.jpg">big, 800x600</a>
    
    Taken with:
    <a rel="photo:camera" href="...">Canon Rebel XTi</a>
</div>

Adding Fields: Flickr

<div about="/photos/timbl/12345" xmlns:photo="...">
    <img src="/photos/timbl/12345_medium.jpg" />
    <span property="dc:title">
      Me and Vint Cert, partying Beijing style
    </span>
    <h3 property="dc:creator">DanC</h3>
    
    Other Sizes:
    <a rel="photo:alternate" href="12345_small.jpg">
      small, 300x200
    </a>

...

Adding Fields: Flickr

<div about="/photos/timbl/12345" xmlns:photo="...">
    <img src="/photos/timbl/12345_medium.jpg" />
    <span property="dc:title">
      Me and Vint Cert, partying Beijing style
    </span>
    <h3 property="dc:creator">DanC</h3>
    
    Other Sizes:
    <a rel="photo:alternate" href="12345_small.jpg">
      small, <span property="photo:dimensions">300x200</span>
    </a>

...

Adding Fields: Flickr

(RDFa) Conclusion

Growing the Data Web from the Existing Human Web
Keep your metadata close
Distributed Extensibility: find vocabularies, reuse them, extend them, evolve them.
Learn More: rdfa.info, rdfa.info/wiki.

Semantic Infrastructure @ CC

Hypocrisy?

CC recommends Semantic Web technologies for marking works
Our own systems used anything but

The Present (Q1 2008)

Redundant License Metadata

XML to drive API
RDF to serve to other consumers
Comphrensive list of licenses: which one?

Multiple Locations for "Engine" Information

XSLT rules
XML to drive "questions"
Jurisdiction information — both a question and a license property

Custom Formats, not so great

API clients become CC-specific
How do I format my "answers"?
How can I verify that I'm getting the correct answer?

Solutions

RDF as the canonical source of information
A single wrapper library for our services
SPARQL queries that can be used to query for specific types of information

CC in OOo

OpenOffice.org add-in to support CC licensing documents
Initial test-bed for pure-RDF/SPARQL approach

cc.license

Python library which provides a standard API for license information
All license metadata comes from RDF
Exploring replacing our "questions" definition with more RDF
Goal: Drop in ccREL metadata for any license (CC or not) and go

The Information Stack

Two wikis: public and internal
Lots of information in both
Search and re-purposing sucked

[Semantic] Content Directories

[Semantic] Content Directories

[Semantic] Content Directories

Additional Tools

Semantic Forms
Semantic Layers
Semantic Calendar
Header Tabs
MediaWiki Templates

Using Templates

Benefits

Content about each subject can be stored in the article (as opposed to spread across different subject area articles)
Facts and summaries can be queried and included in many different locations
We can repurpose information via RDF, XML and CSV export
SPARQL endpoint available

An Introduction to ccREL and The Use of Semantic Technology at Creative Commons

Agenda

Why Machines?

Previous Recommendation

Pervasive Problems

CC Rights Expression Language

Work Properties

License Properties

Work Properties

License Properties

Easier to Publish

Principles for Publishing ccREL in HTML

RDFa

Chasm Between Human and Data Webs [Semweb]

2 published documents, 2 document types

So much data on the web.... [Social Network]

So much data on the web.... [Craigslist]

So much data on the web.... [Flickr]

So much data on the web.... [SemTech2008 Program]

The Chasm

A significant gap between what browsers see and what humans see

Bridging the Human and Data Webs [Correspondence]

Bridging the Human and Data Webs [Correspondence]

Bridging the Human and Data Webs [A Single DRY Format]

Change the visual, change the machine-readable

The Machine Readers should Follow

Bridging the Human and Data Webs [A Single DRY Format]

Bridging the Human and Data Webs [A Single DRY Format]

Bridging the Human and Data Webs [Extensibility & Modularity]

The Flickr Photostream

Bridging the Human and Data Webs [Extensibility & Modularity]

Flickr used by a Newsreader

Bridging the Human and Data Webs [Extensibility & Modularity]

Flickr used by a Digital Photo Frame

Bridging the Human and Data Webs [Extensibility & Modularity]

Flickr used by an Asset Management Program

A Link with Flavor

A Link with Flavor

Text with Flavor

Text with Flavor

Text with Flavor

A Few Articles

A Few Articles

A Few Articles

Image within an Article

Image within an Article

Image within an Article

Image within an Article

The power of about

Adding Fields: Digg

Adding Fields: Digg

Adding Fields: Digg

Adding Fields: Digg

Adding Fields: Digg

Adding Fields: Digg

Adding Fields: Flickr

Adding Fields: Flickr

Adding Fields: Flickr

Adding Fields: Flickr

Adding Fields: Flickr

(RDFa) Conclusion

Semantic Infrastructure @ CC

Hypocrisy?

The Present (Q1 2008)

The Problems

Redundant License Metadata

Multiple Locations for "Engine" Information

Custom Formats, not so great

Solutions

CC in OOo

cc.license

The Glorious Future

The Information Stack

Content Directories

[Semantic] Content Directories

[Semantic] Content Directories

[Semantic] Content Directories

Additional Tools

Using Templates

An Introduction to ccREL and
The Use of Semantic Technology at Creative Commons

The power of `about`