Autumn Meeting 2017
6 – 8 November, Barcelona (Spain)
The meeting will be held at
TRYP Barcelona Apolo hotel
Avinguda del Parallel, 57-59, Barcelona
More about it in the web.
Monday, 6 November
Morning sessions: 09:00 to 12:30
- Opening
- IPTC NewsCodes Working Group session
- Discussion topic: The Art and Science of Practical Metadata
- IPTC Photo Metadata Working Group session
- Metadata for 360degree and VR images (photo and video)
Afternoon sessions: 14:00 to about 18:00
- IPTC Video Metadata Working Group session
- Discussion topic: Trust in the Media
- IPTC EXTRA project: update on round 1, preview on round 2
Tuesday, 7 November
Morning sessions: 09:00 to 12:30
- IPTC Public Relations Committee session
- IPTC Rights Expressions Working Group session
about the W3C ODRL and a RightsML update - IPTC Sports Content Working Group session
about the sportjs development
Afternoon sessions: 14:00 to about 18:00
- IPTC NewsML-G2 Working Group session
about version 2.26 and the web version of the Implementation Guidelines document - IPTC News-in-JSON Working Area
about plans to extend the properties of the ninjs standard - Discussion topic: what type(s) of JSON formats fit IPTC standards?
Wednesday, 8 November
Morning sessions: 09:00 to 12:30
- Annual General Meeting of IPTC as company
- Discussion topic: Automating News – Speed is nothing without Accuracy
Afternoon sessions: 14:00 to about 16:30
- Overspill slot of previous sessions
- IPTC Standards Committee session
More about the discussion topics
The Art and Science of Practical Metadata
Metadata makes content usable – enriching search, enabling navigation and letting news publishers and consumers slice up content into relevant subsets. But what types of metadata should be applied? What is the best way to apply that metadata to content?
A discussion of the practicalities of metadata for the news and media industry. Topics include:
- What types of metadata are useful for which use cases?
- What are the pro’s and con’s of the different techniques for applying metadata – by hand, rule-based, statistical, Machine Learning, Deep Learning…?
- Do different media types require different types of metadata? Text, photo, video, audio, AR, VR… How do harmonize across the different media types, e.g. for search?
- Exactly how much metadata do you need? How do you strike the right balance between completeness and efficiency? Should metadata be maintained and delivered together with the content or independently (e.g. as sidecar XML or via an API)?
- Keeping metadata fresh – how do you manage the metadata applied to content over time – as stories evolve during the news cycle and in your archive of published material?
- For an individual item, do you need metadata about the metadata – for example to track the changes to metadata, where it came from or to deal with contradictory metadata?
- What kinds of metadata are useful to drive content analytics?
Automating News – Speed is nothing without Accuracy
Reporters and editors all agree that Accuracy is fundamental to good journalism. But when it comes to emergency news in the 24 hour news cycle, they face a huge dilemma: can Accuracy be compromised for Speed, and if so, to what extent.
See http://reutersinstitute.politics.ox.ac.uk/publication/speed-vs-accuracy-time-crisis
As the information age moves to more automated news production, the challenge is further complicated by the need to balance Speed, Efficiency and Value – without the real time human control of Accuracy.
At the Autumn Meeting, various news providers will present insights into their news automation capabilities and challenges. Topics include:
- Motivation: Why automate? Competitive edge? Enabling breadth of coverage? Resource efficiencies?
- Candidates: What areas of news can/should be leveraged for automation? Market data? Full news stories? News update notification?
- Accuracy: How does Speed impact Accuracy, whilst delivering Efficiency and Value?
- Control: What can be truly automated? When is a human handbrake required?
- Responsibility: What is the impact of delivering potentially market-moving news? What are the implications of incorrect automation?
Trust in the Media
Surveys around the world indicate declines in confidence that mass media report the news fully, accurately and fairly.
http://www.gallup.com/poll/195542/americans-trust-mass-media-sinks-new-low.aspx
http://www.rcmediafreedom.eu/Publications/Reports/Trust-in-Media-2016
http://www.edelman.com/insights/intellectual-property/2016-edelman-trust-barometer/
At the Autumn Meeting these facets of trust in the media should be discussed:
- Is there a role for the IPTC and its membership to help address this problem? In particular, is there a role to play for technical standards?
- We are aware of schema.org’s https://schema.org/ClaimReview and the Credibility Schema Work being pursued by Misinfocom https://misinfocon.com/building-technical-standards-for-credibility-59ef9ee4ab73. Are these helpful in tackling the problem?
- Is technology really the most important factor?
- What else should news organisations and vendors be doing?