karben

Slide1 : NITF News Industry Text Format An IPTC Standard International Press Telecommunications Council


Slide2 : Alan Karben Chairman, NITF President, XML Team Solutions alan@xmlteam.com www.xmlteam.com All-Stars at Sports Data Integration News Standards Summit – December 8, 2003 – Philadelphia


What is NITF? : What is NITF? An XML vocabulary for defining the content and structure of news articles. Because metadata is applied throughout, NITF documents are far more searchable and more useful than HTML pages in many critical aspects.


What's good about it? : What's good about it? Simple Stable Sanctioned Sound Semantic Structures


What's its history? : What's its history? Back in 1979: ANPA 1312 and IPTC 7901. Early 1990s: A successor is needed. Work on NITF began. "Back when XML was a four-letter word." 1998: NITF becomes XML-compliant Now, we're up to NITF 3.2


Who’s using it? : Who’s using it? A sampling:


What does it do? : What does it do? NITF supports the identification and description of a number of news characteristics. Documents are split between the global metadata section (up top), and the structured article (down below).


What does it do? (2) : What does it do? (2) Highlights of the metadata section include: Who owns the copyright to the item, who may republish it, and who it's about. What subjects, organizations, and events it covers. When it was reported, issued, and revised. Where it was written, where the action took place, and where it may be released. Why it is newsworthy, based on the editor's analysis of the metadata.


What does it do? (3) : What does it do? (3) The body of the document describes functionally the components of a news article. There's what we call "containers": headline byline paragraphs sub-headlines tables "media references" (generally photos, but also good for audio and video references)


What does it do? (4) : What does it do? (4) Inside "containers" go NITF's "enriched text", which includes references to: companies people titles locations events money hyperlink references


What doesn't NITF do? : What doesn't NITF do? NITF & NewsML NITF is designed to give structure to independent, generally text-based news articles. NewsML is for the structuring of multimedia news packages. There is no concept in NewsML of a paragraph or subheadline. There is no concept in NITF of a sidebar or alternative translations of the same document.


What's it look like? : What's it look like? Example towards end of presentation.


What's its future? : What's its future? A great and easy way for publishers large and small to keep their content flexible and searchable. We seek wider adoption beyond the traditional news industry. For Blogs: xHTML may very well stay the de-facto format for the full contents of "stuff you author in Moveable Type et. al." But NITF would be a good contender.


Slide14 : Specialized Content Featuring... SportsML: The Sports Markup Language IPTC Standards International Press Telecommunications Council


The S.C. Initiative : The S.C. Initiative The membership of the IPTC have been drawn together by their enthusiasm, nay, passion for news reporting. “Shouldn’t our interchange standards be as high as our journalistic standards?” anonymous former IPTC chairman However, they have to make money, too.


The S.C. Initiative (2) : The S.C. Initiative (2) The news business isn’t just about writing news stories. It’s also about data. Specialized data.


The S.C. Initiative (3) : The S.C. Initiative (3) We surveyed the field for topics that bring news companies revenue: Financial Data TV Listings Sports Weather Display Ads & Classifieds Election Results


Initiatives We Watch : Initiatives We Watch EML: The Election Markup Language (Oasis) Specifications for Publisher & Agency Communications Exchange XML (SPACE/ XML) AdsML: Electronic transfer of information in the advertising industry (NAA/Ifra) Financial Data Standards like MDDL and FixML


IPTC Initiatives : IPTC Initiatives ProgramGuideML Interchange of Radio/TV Program Listings Including: Title of show Times and durations Free v. Cable v. Premium v. Pay-Per-View Rerun? Genre/Episode Cast members And More!


IPTC Initiatives (2) : IPTC Initiatives (2) EventsML For listings that help find readers things to do Including: Event Name Event Backers Dates Places People (contacts and participants) Categories


SportsML : SportsML The world's first and only "group effort" to standardize how publishers refer to sports scores schedules standings statistics stories SportsML aims to be the global XML standard for the interchange of sports data. Open to participation by everyone


SportsML Simplifies : SportsML Simplifies SportsML makes: Parsing Databasing Querying Formatting all easier SportsML: Raises the bar on how content is digitally described Raises the level of “index-ability" of the content Lowers the costs for the receivers of syndicated content Lowers the amount of time it takes to “figure things out”


The SportsML Design : The SportsML Design One "core" vocabulary holds properties that are common to all sports "Plug-in" vocabularies for specific sports Core SportsML Vocabulary


Architecture : Architecture SportsML faced architectural challenges representative of any Specialized Vocabulary initiative. RDF-style: Not so many elements. A generic "" with "name" and "value" attributes (e.g., 60% of all Player X's Shots actually Scored) Advantages: Small DTD or Schema “Easily Extensible” (Fewer pesky updates needed to the DTD or Schema)


Architecture (2) : Architecture (2) SportsML-style: More elements and attributes Advantages: Everything is individually validatable against DTD/Schema Less room to be inconsistent


SportsML In Action : SportsML In Action An Illustration Of Cooperating Acronyms REST + RSS + SportsML + NITF  HTML Imagine you want to get yourself some coverage from a recent Philadelphia Eagles games to publish online....


SportsML In Action : SportsML In Action You could first build the query, using a form like this...


SportsML In Action : SportsML In Action This builds out a URL something like this: searchDocuments? sport-key=15003000 & league-key=l.nfl.com & team-key=l.nfl.com-t.20 & document-class=event-summary


SportsML In Action : SportsML In Action This gets you RSS something like this:


SportsML In Action : SportsML In Action Which leads you SportsML like something like this:


SportsML In Action : SportsML In Action And also SportsML+NITF like something like this:


SportsML In Action : SportsML In Action Which can give you HTML like something like this:


Slide33 : NITF & SportsML & Specialized Content IPTC Standards International Press Telecommunications Council www.nitf.org www.sportsml.com www.iptc.org