My name is Sergey Fomin; I’m the Сo-founder&CEO of Picvario. We created a system that provides a service that reads and puts your photos, videos, and audio files in order. I know firsthand the shortcomings of working with media materials — I have been running a photo management agency for 20 years. Media outlets use more than 5,000 photos from our archive every day.
Today, I will be telling you why you need metadata in this digital age and I will also be sharing a few insights into the digital future of working with content. As deepfakes, synthetic content, inclusivity in media, and NFT are our reality and rapidly evolving.
Content:
- Metadata: What is it, and why do we need it?
- A short introduction to IPTC/XMP.
- Systems for reading and populating metadata.
- The evolution of digital (un)reality.
Metadata: What is it, and why do we need it?
In the digital era, we generate a tremendous amount of content. Just to give you an idea — the average person produces 1.7 MB of content per second! And that’s taking into account that 35% of humanity still does not have access to the Internet.
To navigate through all the digital material produced, metadata is used — it specifies the basic information about a file. The information is written within the files themselves or in support files: XML, XMP, JSON, etc. Metadata allows the user to read why, when, and where the media file was created and whether it can be used.
There are different types of metadata, the most common being EXIF and IPTC/XMP.
EXIF metadata contains primarily technical information about the digital file — the camera settings when the photo was taken. The standard includes information on geolocation, camera model, flash mode, file format, date and time of creation, etc.
However, EXIF only includes dry facts about equipment settings, which is insufficient to understand the context of the media file. The juicy bits are hidden in the IPTC/XMP metadata. The standard contains more than 200 metadata fields, thereby revealing the essence of the file: content, name, description, purpose, date of creation, keywords — tags, and other information. Particular attention is given to the authorship — you can specify who the creator is and the terms of the license for the photo or video.
A short introduction to the IPTC/XMP
The IPTC/XMP standard was developed in the late ’60s by a news media group: the International Press & Telecommunications Council. Now, the organization has more than 60 media representatives, including giants such as Adobe, Agence France-Presse, Bloomberg, Reuters, and Getty Images.
I am a representative of the Council of European Media Agencies (CEPIC) for the IPTC organization and am involved in developing the standard. Hence, our software Picvario supports the IPTC standards. In this sense, one of the advantages of Picvario is that we are involved in developing metadata standards that affect the digital media file management industry.
When sending photos or videos, we often lose information about the media file: title, description, authorship of the content, etc. Today, our main goal is to develop and expand industry standards for the exchange of asset data.
By the way, you can check which metadata is attached to your photos and edit it, if necessary, through our free program MetaEditor.
Systems for reading and populating metadata
Most professional software can work with IPTC/XMP metadata, such as Adobe Photoshop, Bridge, ACDSee, and others. This applies to separately developed DAM systems — including Picvario. We originally developed a platform that supports existing standards.
Here’s a practical example of using media content management systems:
- The photographer sends photos and videos from the scene.
- The media editor must promptly broadcast these materials in lower resolution to the news correspondents for publication and place all the originals in an archive.
- It’s also vital for the publisher to ensure copywriting and image license information are included.
All the above-mentioned processes the content has to go through can be simplified by using DAM systems.
In addition to being used in the media, digital asset management systems have also been used in various other industries. Media content management tools are successfully used in educational institutions, marketing agencies, manufacturing plants, etc. As a result, cultural organizations: galleries, libraries, archives, and museums are starting to use DAM systems.
In 2018, there was a tragedy in the art sector due to the lack of a content storage system. The National Museum of Brazil had a fire in which more than 20 million pieces burned and were irretrievably lost. If the museum had the DAM platform, digital copies of the art objects could have been saved.
The evolution of digital (un)reality
IPTC/XMP members are actively lobbying and developing inclusivity within metadata standards. The latest IPTC/XMP release introduced the Alt-text field and the extended description field.
The innovations apply to more than 320 million visually impaired and blind people worldwide. With Alt-text, you can enter a description of the image. This information is read out loud by screen readers, and people with visual impairments can «visualize» the photo or other visual content from the description narrated to them. The good news is that the Alt-text field is integrated into Picvario.
Other topics are on the agenda. For example, this spring, IPTC/XMP representatives discussed the current challenges of working with media content. One hot topic is the role of metadata for NFT. For context, NFT is a digital asset that can be bought and sold. The object’s value determines its price: it can be an image, a picture, a video, the first tweet in history, etc. For NFT to be of value on the market, you need metadata describing the characteristics of the asset.
At the IPTC/XMP meeting, other ideas in the digital world were also brought up: for example, using artificial intelligence (AI) to create synthetic content. Today, we can imagine anything; AI can recreate any fantasy we have.
Another topic that only the laziest person hasn’t written about is the improvement of deepfake production. While it used to take a lot of time and video footage to create a realistic deepfake, today, you can create a deepfake live. For many politicians, celebrities, and other media personalities it’s no longer a laughing matter. To combat fakes and authenticate digital images, C2PA was created by a coalition of media giants — Microsoft, Adobe, BBC, Intel and others. It is expected that the content producer will assign a digital certificate to the asset, specifying information about the primary format, volume, owner, etc.
According to Hani Farid, a computer scientist at the University of California, it would take about five years to create a quality deepfake. However, the practice has shown that he was wrong; frighteningly realistic models began to be created in two years.
In any case, digital technology opens up tremendous opportunities for mankind; the main thing is to look with our eyes wide open and pay attention – just so we do not miss anything.
