Metadata

What is Metadata?

Essentially, metadata is data about data; it’s often (though not always) machine-readable information that determines the structure or organization of human-readable information or content. Metadata can appear on websites, databases, or elsewhere.

“Visible” data vs. metadata

Whereas the visible parts of a website—its words, images, videos, and other interactive elements—can be considered primary data, metadata is information about that visible content or elements. Webpage metadata can include things like word count, date published, summary for search engines, or the source of the article. Note that some of this metadata may also be visible data, for example when it appears as a webpage summary on a search results page.

What sort of data has metadata?

Almost all data will have some corresponding metadata. The amount and type of metadata depends on a number of factors such as what type of data it is, how it was created, where it’s stored, and for what purpose.

In the context of webpages, metadata can include:

• Word count, date of publication, summary, author, or source.
• Info about any photos or videos, such as the date the photo or video was taken/recorded; info about file size, format, and resolution; info about the device used to create the image/video; or even info about the location where a photo was taken or a video was recorded.
• On social media, the site or app you use may create metadata about your post, including things like date, location, or even handle, avatar, or identity. As time passes, the primary data (the message or image) accumulates more metadata, like how many times it was accessed or who accessed it.

Also note that there is no fundamental line between data and metadata. In most cases, what counts as data and what counts as metadata can be inverted simply by recategorizing your dataset. It’s all a matter of the use case for the dataset.

Does metadata pose a threat to privacy?

Some metadata is benign, and not inherently invasive to individual privacy. For example, much of the metadata about webpages—such as publication date—help ensure the most recent articles relevant to a search query appear toward the top of search engine results. On social media, likes can determine a post’s popularity, and ensure it appears near the top of a person’s newsfeed. (Though even this benign instance of social media metadata can expose personal info.)

However, some metadata can harm privacy. For example, when you shop online you’ll likely generate metadata by viewing an item for purchase. This creates a metadata element that can be tracked back to you. You see this in action later when your screen is flooded with ads for the item you viewed (a technique called retargeting). If the source of this metadata is ever hacked or otherwise compromised, your personal data could be exposed.

How can I reduce my metadata exposure?

It’s difficult to be truly anonymous on the Web. However, there are several measures you can take to reduce your metadata trail and improve your online anonymity.

• Don’t share your location automatically. Adjust your phone settings so location information is not included when you take a picture. Similarly, require all apps to ask you for permission to allow location sharing.
• Use a virtual private network (VPN) to mask your location. Some VPNs are better than others (some can even add more risk than they remove, so research before downloading/buying anything). Brave, for example, has a VPN built directly into the browser—no additional download required.
• Use a browser—like Brave—that has strong privacy protections. This can block or inhibit many of the techniques that websites use to collect metadata and track visitors, such as third-party cookies and fingerprinting.