IPTC NewsCodes - Controlled Vocabularies for the News Media (EBU MDN Workshop 2021)

IPTC NewsCodes: Controlled
Vocabularies for the News Media
Brendan Quinn, Managing Director, IPTC
Jennifer Parrucci, Senior Taxonomist, The New York Times
Tuesday 25 May 2021

Overview
© 2021 IPTC (www.iptc.org) All rights reserved 2
What is IPTC? What do we do?
What are IPTC NewsCodes?
How do we maintain the IPTC NewsCodes?
How can IPTC NewsCodes be used?
The future of IPTC NewsCodes

What is IPTC?
• The global technical
standards body of the news
media
• We are an open, non-
partisan, non-political
organisation dedicated to
promoting interoperable
standards and best practices
in the news and media
industries
• We are a registered non-
profit organisation funded by

History of IPTC
1965
• IPTC founded to
safeguard the
telecommunications
needs of the press
• Accredited by
International
Telecommunications
Union in 1967
1979
• Creation of IPTC 7901
format for binary
transmission of news
content
1990
• Information
Interchange
Model (IIM), the
first multimedia
news exchange
format, created
with NAA
1995
•Adobe adopts IIM in
Photoshop, which led
to the development
of the IPTC Photo
Metadata Standard in
2004
2001
•NewsML, the first
XML news exchange
standard, was
released
•First set of
NewsCodes
controlled
vocabularies released
2008
•NewsML-G2
family of
standards
released,
including
EventsML-
G2 and
SportsML-
G2
2010s
•rNews
•ninjs
•RightsML
•Video Metadata
Hub
2020s
• Trust &
Credibility
additions to
•IPTC Photo
Metadata in
Google Search
results

Our voting members

Our associate members
• Agencia EFE Spain
• ATC - Athens Technology Center Greece
• BVPA Germany
• CBC / Radio Canada, Canada
• CEPIC Europe
• DeFodi Images Germany
• DPP Digital Production Partnership United Kingdom
• EBU - European Broadcasting Union Europe
• epa - European Pressphoto Agency Europe
• ETX Studio France
• Fingerpost UK
• FotoWare Norway
• iMatrics Sweden
• JIS - Jozef Stefan Institute Slovenia
• Keystone SDA Switzerland
• KNA - Kath. Nachrichten-Agentur Germany
• Lusa Portugal
• Mainstream Data USA
• MENA Egypt
• Microstocksolutions LLC USA
• Naviga USA
• NTB - Norsk Telegrambyrå Norway
• PR Newswire UK
• Profium Finland
• PUBLISH, Inc South Korea
• STA - Slovenian Press Agency Slovenia
• Sourcefabric Czech Republic
• TASS Russia
• XML Team Solutions USA
• Yuanben China
• … plus 13 Individual and Honorary Members

Our startup members
Special membership category open to small companies before they
have become profitable. Launched in late 2020
IMATAG Metadata Authoring Systems Scribely

IPTC’s technical standards
IPTC Photo
Metadata
Video
Metadata
Hub
NewsCodesM
ediaTopicsSu
bject Codes
NewsML-
G2 and
NewsML-1
RightsML
ninjs
rNews SportsML-
G2
EventsML-
G2
Historical
standards:
•IPTC 7901
•IIM format

IPTC Photo Metadata:
The global standard used to describe pictures
• IPTC metadata has been used since 1990 to store
descriptive, administrative and rights metadata embedded
in image files
• Since Adobe adopted it in Photoshop in 1995 it has been
used in millions of image files
• Originally based on IPTC Information Interchange Model
(IIM) standard, now based on XMP embedded metadata
framework (introduced by Adobe, now an ISO standard).
– Most software supports both formats but we encourage users to use
XMP as it has more fields, unlimited field lengths and multi-language
support
• IPTC Photo Metadata is used by:
– All major camera manufacturers: Canon, Nikon, Sony etc
– Photo software vendors: Adobe PhotoShop and Lightroom,
PhotoMechanic, GIMP etc
– Digital Asset Management systems: FotoWare, Extensis and many more
– Google image search results © 2021 IPTC (www.iptc.org) All rights reserved 9

IPTC Photo Metadata and Google Images
• In 2018, IPTC, CEPIC and
DMLA worked with Google
to display embedded IPTC
Credit, Copyright and
Creator fields in Google
Images search results

IPTC Photo Metadata drives the
Google Images Licensable badge
• In 2020, Google started using IPTC Photo
Metadata to link “Licensable Images” to a
place where the image can be purchased
or licenced
• A “licensable” badge is displayed in the
search results grid
• Links from Google search results go directly
to a page where users can license the image
• Works even when images are used on
third-party sites
– The badge isn’t shown in thumbnail page, but “get this
image on” and “view licence details” in side panel will
work

Video Metadata Hub
• IPTC’s Video Metadata Hub is a universal
metadata schema developed with the key
goal of storing and exchanging metadata
in a safe and reliable way.
• The Video Metadata Hub is a single set of
video metadata properties based on the
properties in multiple technical formats,
joining metadata across QuickTime, XMP,
MPEG4, MPEG7, MXF, EBUCore, PBCore
and NewsML-G2 plus manufacturer-
specific metadata formats such as Sony
XDCAM and Panasonic/SMPTE P2.
• These properties can be used for
describing the visible and audible
content, rights and licensing information,
administrative details and technical
characteristics of a video in a system-
independent way.

Video Metadata Hub: What problem does it
solve?
• Metadata is stored in different ways in
existing video formats. For video editors it is
very hard to move video metadata between
systems and the semantics are not always
clear.
• For example, location information can be
described in all of the ways described in the
table
• IPTC Video Metadata Hub describes two
fields, Location Shot and Location Shown,
with very clearly defined semantics, and the
fields are mapped to fields of existing video
formats, so metadata can be moved
between systems from different vendors.
QuickTime location with role=0
MPEG7 creation/location
Sony XDCAM PlanningMetadata/Properties/Me
ta[@name="Location"]
schema.org locationCreated
SMPTE P2 ClipMetadata/Shoot/Location
EBUcore
coverage/spatial/location +
typeLink="ivqu:locationShot”
coverage/spatial/location +
typeLink="ivqu:locationShown"
OR subject +
typeLink="ivqu:locationShown"/s
ubjectCode + subjectDefinition

IPTC News Exchange formats - summary
• Binary news exchange format created in 1979! Still used by
many news agencies. No ongoing work.
IPTC 7901
• First-generation multimedia news exchange format created in
2001 when XML was new. Still used by some agencies and
publishers.
NewsML 1
• News Industry Text Format, a generic XML article markup that is
still used widely (although some agencies distribute articles in
XHTML)
NITF
• Re-thought version of NewsML incorporating EventsML-G2 and
SportsML-G2. New releases generally twice a year. Currently at
v2.28 (November 2018).
NewsML-G2
• Semantic news markup format, incorporated into schema.org
(led by Google)
rNews
• News in JSON. Simpler schema intended for APIs, JSON
databases (e.g. Elasticsearch, MongoDB) and lighter-weight use
cases
NinJS
Photo by Mr Cup / Fabien Barral on Unsplash

Proxy (MP4)
Metadata
Metadata Watch Services
2
2: Checking every 30
seconds for new
assets
3: Query asset metadata
via XDCAM air API
4: Pass asset metadata
to G2 XML creation
services
XML Creation Services
6
6: Send NewsML-G2 XML and
MP4 to AWS S3 bucket for
Ooyala
7: Get G2 XML
and MP4 by
OOYALA services
Planning
metadata
Planning
Metadata
5: Get MP4 from
XDCAM air S3
bucket
9
9: Get G2 XML
and MP4 by
Reuters
Connect
8: Send NewsML-
G2 XML and MP4
to AWS S3 bucket
for Reuters
8
1: Connect to XDCAM Air via API calls
DPP Metadata Exchange for News – using NewsML-
G2

RightsML: IPTC's Rights Expression Language
for the media industry
• In a modern media organisation, content comes from
many sources with many combinations of permissions and
restrictions
• The task of deciding which assets can be used on which
properties is complicated, error-prone and manually
intensive
• Mistakes can cost organisations millions in claimed
royalties and legal fees
• The solution: Machine Readable Rights – declaring in
markup a summary of what is usually contained in a legal
contract
• This ensures that both rights holders and clients can be
confident that the use of digital assets complies with the
appropriate rights and restrictions
• Allows publishers to quickly decide which assets can be
used in which contexts, or to flag assets for manual
checking © 2021 IPTC (www.iptc.org) All rights reserved 17
“Embargoed until XXX” “China out” “Web only”

What does IPTC do? Events
• We run two face-to-face members meetings each year where we update on
new standards and allow members to share best practice and collaborate on
projects
– Recent meetings:
• Athens, Greece and Toronto, Canada in 2018
• Lisbon, Portugal and Ljubljana, Slovenia in 2019
• Online meetings in 2020 and 2021
– Upcoming meetings:
• Online, October 18– 20 2021
• Hopefully Tallinn, Estonia and New York City in 2022
• We run the IPTC Photo Metadata Conference in association with CEPIC, the
photo industry association
– Berlin in 2018, Paris in 2019, online in 2020 (with over 200 registrants!)
– 2021 meeting: online in Autumn
Diego Ceccarelli, Bloomberg at
IPTC Autumn Meeting 2018,
Toronto Canada

What does IPTC do? Software development
• We run development projects related
to our work
– The IPTC EXTRA open source rules-based
text classifier
– The GetPMD photo metadata checker
– The PhotoMetadata Explorer (available to
members only) to understand how
publishers use IPTC Photo Metadata
– Reference implementations of RightsML
parsers
– SportsML parser and SportsJS converter
• We work with open source software
developers to support their correct
implementation of IPTC standards
– Phil Harvey’s Exiftool
– SourceFabric’s SuperDesk newsroom
software
– ffmpeg, mediainfo © 2021 IPTC (www.iptc.org) All rights reserved 19

What does IPTC do? Partnerships and advocacy
• We provide advice to other groups and work with other associations on
standards
– W3C
• Joint work on Open Digital Rights Language (ODRL)
• RightsML is a “profile” of ODRL
– Schema.org
– JPEG / ISO
– Content Authenticity Initiative
– Trust Project, Journalism Trust Initiative
– CIPA (creators of EXIF)
• We also work with companies on implementing and
integrating our standards into their products
– Adobe
– Google
– Camera makers, newsroom software vendors, sports stats systems,
etc…

Overview

IPTC News Codes: concept taxonomies for news

IPTC subject classification systems: a quick
history
• IPTC originally created the Subject Reference System
(SRS) in 1995, in association with the Newspaper
Association of America (NAA)
– Defines many vocabularies such as Subject Qualifier, Genre,
Media Type, Media Type, and Gender, but main one is Subject
aka Subject Code
– All are available in an XML “TopicSet” format (part of NewsML
1)
– Subject contains 17 top-level Subjects, each with three-letter
alphabetic and 8-digit numeric references across three levels
– SubjectCodes are/were available in English, French, German,
Italian, Japanese, and Spanish
– Used within IPTC Photo Metadata Standard as “IIM 2:12
Subject Reference” (called “Subject Code” in the latest IPTC
Photo Metadata Standard) – and still used in Photoshop IPTC
metadata editor interface…
– Problems: too-strict hierarchy, unbalanced granularity, hard to
edit. Couldn’t accommodate requests from users.
An excerpt from the SRS documentation

This led to MediaTopics
• So in 2010, the Media Topics vocabulary was
born
• It contains around ~1,200 terms for
categorising news content
• Terms and definitions in 12
languages/variants (so far):
– Arabic, Simplified Chinese, Danish, English
(British and US), French, German,
Norwegian, Portuguese (for Portugal and
Brazil), Spanish, Swedish

Media Topics: IPTC’s main subject vocabulary
• We had many requests for new terms in the Subject Codes
vocabulary that we couldn’t accommodate due to its strict
hierarchy
• So in 2010, the Media Topics vocabulary was born
• It contains around ~1,200 terms for categorising news
content
• Terms and definitions in 12 languages/variants (so far):
– Arabic, Simplified Chinese, Danish, English (British and
US), French, German, Norwegian, Portuguese (for
Portugal and Brazil), Spanish, Swedish
• Also 200+ controlled vocabularies for news content
– Audio Codec, Colorspace, Contributor Role, News
Provider

Human-readable views
https://cv.iptc.org/newscodes/mediatopic/

https://show.newscodes.org/

https://www.iptc.org/std/NewsCodes/treeview/

https://www.iptc.org/std/NewsCodes/IPTC-MediaTopic-NewsCodes.xlsx

Machine-readable views
• All NewsCodes are maintained as NewsML-G2 Knowledge
Items, mapped to W3C SKOS
• Available as:
– NewsML-G2 Knowledge Items (XML)
– SKOS RDF/XML
– SKOS RDF/Turtle
– SKOS RDF/JSON-LD
• Can be obtained from cv.iptc.org via URL query params or
HTTP content negotiation
• Examples:
– https://cv.iptc.org/newscodes/mediatopic/20000008
?lang=en-GB&format=json
– https://cv.iptc.org/newscodes/genre/?lang=x-all&format=rdfxml
– wget -O IPTCscene.ttl http://cv.iptc.org/newscodes/scene/
?format=rdfttl
• See guidelines at https://www.iptc.org/std/NewsCodes/guidelines/cv.iptc.org-
guidelines.html

More than just subjects/topics
• CVs for Audio Codec, Colorspace,
Contributor Role, Financial Instrument
Type, Genre, News Provider, Trust
Indicator, Video Scaling, etc…
• Most are used in NewsML-G2 files for
conveying information about
news/media content
• We keep references to third-party
codes such as ISO Country Codes,
BCP47 Language tags etc
• All NewsCodes are accessible from
https://cv.iptc.org/newscodes/

More than just subjects/topics (2)
• We have dozens of vocabularies
for sports, used by SportsML,
SportsJS and a possible
SportsRDF standard:
– Event outcome (win, loss, tie/draw,
undecided)
– Professional status
(amateur/professional)
– Core statistics (score, attempts on
goal, possession time, …)
– American Football Penalty types
(illegal pass, roughing, illegal
motion, …)
– Basketball Player Roles (shooter,
blocker, assist, …)
– ... And MANY more © 2021 IPTC (www.iptc.org) All rights reserved 32

Overview

The IPTC NewsCodes Working Group
• Group Objective: The maintenance of
IPTC metadata controlled vocabularies
for categorizing content – branded as
NewsCodes – and the creation of new
ones when required
• Main focus is usually on Media Topics
but the group also maintains the other
NewsCodes
• Work on Sports NewsCodes is done
jointly with the IPTC Sports Content
Working Group
• Also functions a collaboration/discussion
group for taxonomists from media
organisations
Recent work:
• Media Topics Additions, edits, definition reviews
• Genre Review
• Media Topics → Wikidata Mapping
• Adding and maintaining translations for
MediaTopics
• Entities discussions

Criteria for adding new terms
Balance with granularity in other branches of the Media Topics taxonomy
Will adding the proposed Media Topic(s) keep the level of granularity fairly even across the
vocabulary?
Threshold of coverage
Is the proposed Topic likely to be covered with reliably regular or seasonal frequency by
participating member media organisations?
Not covered by another term or combination of terms
Does the proposed Topic have distinct semantics and scope from all existing Media Topics, or
combination of Media Topics?
Supported and/or validated by other general taxonomies
Does the proposed Topic have representation in external widely available general subject
taxonomies, such as Wikidata, Library of Congress etc.?
Does not belong to any highly specialised taxonomies
The proposed Topic should not have representation in specialised domain taxonomies such
as Medical Subject Headings (MeSH).
https://iptc.org/std/NewsCodes/guidelines/#_when_should_a_term_be_added_to_media_topics

Examples of recent changes to MediaTopics
Recently retired terms:
• sports facilities – Use sport venue instead
• inline skating – Use roller sports instead
• accomplishment –Use award and prize or record and
achievement instead
• people – Use more specific terms instead
Recent hierarchy changes:
• dropped criminal investigation is now a child of
investigation (criminal)
• bodybuilding is now a child of competition discipline
• award and prize and record and achievement were
moved to the top level of human interest because we
retired the parent term accomplishment
• birthday, celebrity and high society were moved to
under the top level human interest term because we
retired the parent term people.
Recvently added terms:
• men
• women
• women’s rights
• animal abuse
• border disputes
• cheerleading
• child care
• disinformation and misinformation
Recent label chahges:
• online media -> online media outlet
• traffic crime -> reckless driving
• trade fair -> trade show or expo
• online media -> online media industry
• parliament -> legislative body
• synchronised swimming -> artistic swimming (as per IOC
changes for 2020 Olympics)

How we work – GitHub, Zoom, groups.io
• New terms are suggested by IPTC members via
Slack, email, GitHub, groups.io list or external
users on our public discussion list iptc-
newscodes@groups.io
• We track, prioritise and classify suggestions using
GitHub issues
• We discuss suggestions on GitHub issues and at
our bi-weekly Zoom calls
• When agreed, we make changes to the XML files,
re-generate the public assets and upload them to
the site
• We also regularly re-visit existing branches to
update labels, definitions and hierarchy
• Suggestions can take from 2 weeks to a few
months to be adopted, depending on urgency
and time to next release

How we work – translations
• Currently we have a fairly manual workflow for
dealing with language translations
• We maintain terms in en-GB and en-US, and then
send updates to translation partners who submit
their work in Excel spreadsheets
• We can then merge changes directly from
spreadsheets into our XML files using a custom
script
Future translations
• We’re considering using a translation platform to
handle this workflow - https://crowdin.com/?
https://lokalise.com/?
• We could then integrate translation with our
GitHub repository

Staying up to date
• We generally release updates every
two to three months
• New changes are announced on:
– iptc.org/news as a blog post
– iptc-newscodes@groups.io
– Twitter @iptcupdates
– Internal Slack and email discussion
lists for IPTC members
• There’s no obligation to take
every update straight away, it’s
up to you

Overview

Classification of news content
• Some users of Media Topics:
– Swedish news agency TT and their customers
– Danish news agency Ritzau
– Agence France-Presse (AFP)
– dpa (Deutsche Press-Agentur)
– NTB Norway
– ABC Australia
• Many of these use MediaTopics internally or
in agency-to-publisher feeds, not available
to the public
• Content Management Systems
• Auto-tagging and language analysis tools
(e.g. iMatrics, expert.ai)
• Media Monitoring services (e.g.
http://www.pressmonitor.ca/)

News Credibility and Mis/disinformation
• We don’t want to make yet another
standard. We want to work with existing
groups to ensure that their work can be
applied to our standards, and will extend
ours where necessary
• We worked with industry groups (such as
the Trust Project and Journalism Trust
Initiative / Reporters Sans Frontiers) to help
them to:
– define journalistic credibility indicators
– publish indicators in a useful way
– map indicators to our existing news metadata
standards
• We help fact checking groups to best use
our standards and technologies to make
fact-checking easier and more scalable

NewsCodes to support Trust & Credibility
• We added a new IPTC
NewsCodes trustindicator
vocabulary
• New terms in IPTC NewsCodes
genre vocabulary
• The vocabularies can be used
in NewsML-G2 and ninjs
documents (or anywhere else)
• Guideline document:
“Expressing Trust and
Credibility Information in IPTC
Standards” (currently in late
draft form)

Trust indicators used to enhance news for users
• Agencies can distribute
news marked up with
Trust Indicators using
our CVs
• Publishers can pass on
the indicators in
schema.org markup
• Platforms can use the
schema.org markup to
flag credibility to users
• Users can use tools such
as NewsGuard interpret
and analyse credibility
markup

Extending MediaTopics for your own use
• We often receive requests from media organisations to add terms that would be too specific for
us to add to the general MediaTopics vocabulary…
• … but that shouldn’t stop you from adding terms yourself!
• For example our CV stops at the term “animal” (medtop:20000500). To add your own animal
terms, you can ”mint” your own IDs within their own namespace.
• Create a new term and use the skos:broader relationship to show that it’s a child of the
MediaTopic term
• You can then import your new terms into your database without any fear of falling out of sync
with future MediaTopics updates
medtop:08000000 human interest - Item that discusses individuals, groups, animals, plants or other
objects in an emotional way
 medtop:20000500 animal - Human interest and cultural stories related to animals, including pets
 mycv:12345 cat Small domesticated carnivorous mammal with soft fur
 mycv:23456 lion Species of big cat often found in Africa and Asia
 mycv:34567 penguin Large flightless seabird of the southern hemisphere
 medtop:20001237 anniversary - The celebration or commemoration of …

Media Topics → Wikidata Mapping
• In the past we had mapped the top 2 levels
of MediaTopics to Wikidata
• The project to map all the MediaTopics
arose from an IPTC member’s need to have
links from Wikidata back to MediaTopics
because they are building a classification
tool that will use Wikidata IDs.
• We now have mappings for 93% of
MediaTopics
• Some broader terms couldn’t be mapped,
so we plan to create those terms in
Wikidata

Overview

Wikidata → Media Topics Mapping
• We map from MediaTopics to Wikidata, but
not the other way around
• Wikidata has a property “IPTC NewsCode”
which maps Wikidata entities back to
MediaTopics
• This has been updated ad-hoc by Wikidata
users: already has over 700 mappings
• We plan to update these automatically with
a script and continuously maintain them as
we continue our work
• Coming later in 2021

Mapping to other CVs via wikidata
https://cv.iptc.org/newscodes/mediatopic/20000285
https://www.wikidata.org/wiki/Q253613
https://www.bbc.com/news/topics/c8nq32jw8vjt/personal-finance
https://id.loc.gov/authorities/subjects/sh85048263.html

Media Topics for contextual advertising?
• We have had discussions with several
publishers who are considering using Media
Topics for contextual advertising
• As more users opt out of behavioural
tracking cookies, contextual advertising (i.e.
advertising based on the content of the page
rather than the behaviour of the user) is a
promising alternative
• But to do this, advertisers need a
standardised vocabulary to target against
• IAB taxonomy is very advertiser- and
product-oriented, not really suitable for news
content
• Could MediaTopics provide a solution?
https://wan-ifra.org/2021/05/emotion-and-intent-based-contextual-advertising-emerging-as-promising-alternatives-to-third-party-cookies/

Future work: Named entities?
• Many members have asked whether we will
support named entities (people, places,
organisations, events)
• News by its nature needs new entities all the
time
• New names are typically added to Wikidata 1-2
months after they emerge (but this is improving)
• If news organisations all add a new entity as part
of their workflow, we could end up with dozens
of duplicates
• Going back to edit previously tagged stories to
fix duplicates is a hassle
• In discussions, we agreed that IPTC doesn’t have
the resources to build and maintain a news
entity database for the world
• But perhaps a “best practices guide” would be

Summary: NewsCodes and MediaTopics
Resources
• Info on NewsCodes
• Info on MediaTopics
• IPTC NewsCodes Guidelines
• Human-readable views of
MediaTopics:
– CV Server
– Excel
– HTML tree view
– Interactive tree view
• Guidelines for machine-readable CVs
• Public discussion group on groups.io
• For IPTC members:
– GitHub repository
– Members-only discussion group
– Slack channel

Thank you!
Brendan Quinn
Managing Director
IPTC
mdirector@iptc.org
@brendanquinn on Twitter
Jennifer Parrucci
Senior Taxonomist,
The New York Times
jennifer.parrucci@nytimes.co
m

Appendices

More info on Trust Indicators

Ideal “B2B2C” workflow for trust indicators

Mappings between Trust Project, JTI and IPTC
standards
Mappings are generally direct to NewsML-G2
tags
or “typed links” – a URL with a “qualifier”
showing what kind of link is being described.

IPTC Trust Indicator taxonomy - Hierarchy
• Editorial policy
– Corrections policy, coverage policy, editorial
diversity, ethics, feedback, no-bylines, unnamed-
sources
• Party Info
– Archive, award, knows about, knows language
– Journalist info
• Biography, journalist image
– Organisation info
• Diversity policy, diversity staffing report,
ownership and funding info, senior editorial
staff (masthead)
• Piece of work info
– Corrects resource / corrected by resource
– External / internal source materials, secondary
Browsable web view

Wikidata RDF and SPARQL interface
• Wikidata exposes interfaces in RDF and SPARQL
• Queries can inspect wikidata entities by type or
property:
#People that received both Academy Award and Nobel Prize
SELECT DISTINCT ?person ?personLabel WHERE {
?person wdt:P166/wdt:P31? wd:Q7191 .
?person wdt:P166/wdt:P31? wd:Q19020 .
SERVICE wikibase:label {
bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en"
.
}
}
• “wdt:P166” is “award received”
• “/wdt:P31?” means “this property or any child
instance of it”
• “wd:7191” is “Nobel Prize”
• “wd:Q19020” is “Academy Awards”

IPTC NewsCodes - Controlled Vocabularies for the News Media (EBU MDN Workshop 2021)

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (20)

Similaire à IPTC NewsCodes - Controlled Vocabularies for the News Media (EBU MDN Workshop 2021)

Similaire à IPTC NewsCodes - Controlled Vocabularies for the News Media (EBU MDN Workshop 2021) (20)

Dernier

Dernier (20)

IPTC NewsCodes - Controlled Vocabularies for the News Media (EBU MDN Workshop 2021)

Notes de l'éditeur