What is metadata and where is it used?

In general, metadata can be used to describe other data. Metadata is used when you want to sort more complex data  and put it into a more understandable form. More specifically, there are different types and applications of metadata. In this article we will look at the following questions and topics:

What types of metadata exist?

What do you need metadata for?

Who uses metadata?

Metadata in photos and PDF files

What about data privacy?

A distinction is made between structural, descriptive, administrative and technical metadata.

Structural metadata indicate how data are related to each other. They describe the relationship to other objects, events or information. They illustrate the scope and relationship of the individual data within a data set. Structural metadata provides information about a specific object or resource. This is often digital media.

Practical examples:

In a film on DVD, each section corresponds to a fixed length of the total running time. The structural metadata provides information on the order of the individual chapters on the DVD disc.

The table of contents of a book shows you how many chapters it contains and in which order they should be read.

Descriptive metadata are used to describe and identify data. They provide basic information on certain aspects. They also include keywords for simplified searches. A descriptive metadata record is, for example, the specification of the running time of a film.

Regarding the book, this includes the name of the author, the book title and the short overview (teaser) on the back of the book.

Descriptive metadata only becomes more complex or sophisticated when it is used to identify unique objects such as projects or websites with codes.

Administrative Metadata provide information about the rights, policies, restrictions and instructions regarding the data. Who has access to the data and may edit it? Depending on a user's qualifications, an administrator can restrict access to a file. Administrative metadata is used to make complex data clearer.

The example with the book is about the copyrights and the indication of the publisher.

Technical metadata nclude e.g. information regarding file format, size, version, etc.

In a book, this would be the page numbers and the indication of the edition.

Metadata is mainly needed to find data quickly. Imagine you are a photo agency and you are looking for a photo with a certain motif. Without metadata, it would be like in the past when you dumped out your shoebox with old photos to find a specific one in it. But a photo agency now has not just one, but thousands of shoeboxes and without labelling (marking) these boxes, the search for the one photo would be as successful as the famous search for the "needle in the haystack".

The principle of metadata has been used by libraries and administrations for hundreds of years. Metadata replicates this form of cataloguing in the digital age.

Especially in combination with a professional Datenbank metadata is an optimal means for a faster workflow.

As already mentioned in point 2, it is used by companies and sole traders to make their data easier to find and process.

But other organisations also have a great interest in metadata. They often reveal more about us than we think. For many big players like Amazon, Facebook, WhatsApp, Google, etc., it's all about user behaviour. By collecting metadata, they can draw conclusions about our consumption behaviour and use algorithms to make predictions about future actions. This then allows them to create personalised advertising and ads and thus higher profits.

Even authorities like the NSA use metadata to evaluate connection logs to ensure national security.

Again and again, experts warn against this influence, as political influence also becomes possible. Ethically, too, the question of whether we are becoming "transparent people" keeps coming up.

However, there will be no answer to this in this blog article, because the focus here is primarily on the benefits of metadata. A personal evaluation of the use of metadata will explicitly not be made.

With photos, a distinction is made between 3 different types of metadata: EXIF, IPTC and XMP.

EXIF data are usually collected automatically by the photo camera. This includes the time, date, location of the shot as well as the shutter speed and aperture used. Depending on the camera model and manufacturer, further data is possible.

In contrast, IPTC data usually not collected automatically. They can be added afterwards. These include titles, keywords and copyrights.

XMP data are a special metadata format developed by Adobe. It is control data for the use of Adobe and other image editing programmes. This can be, for example, change information or filters used.

This data is also used for simplified searching and structuring. In good Image databases the IPTC data can be added quickly and easily with the help of a clear input mask. Copyrights are also kept in view for you and you are "warned" in good time if they should expire.

In documents, the metadata is stored in XML format. However, as already mentioned in point 4, Adobe has its own XMP format. The XML code can simply be imported and converted into an XMP format. This means that the metadata can be "taken along" to all Adobe applications and in part also to other programmes without risking their loss. This should facilitate the publishing process.

More detailed information and application aids are available from Adobe.

As you can see, all kinds of information is collected and stored in the metadata. In order to ensure that this data is not misused, since 25.05.2018 the EU Data Protection Regulation also known as the GDPR came into force.

The main issue here is the protection of personal data.

So before you publish a photo or similar, you should check whether this is done in compliance with the GDPR.

Many professional databases point this out to you before publication or already check GDPR conformity when uploading the data.

Metadata is a great thing for structuring data to increase its discoverability and thus improve efficiency and workflow in a company. However, they should also always be protected from misuse. No one should benefit from it but you!