Image Metadata for Artists

I’ve written quite a lot on this blog about metadata in your HTML—particularly RDFa and Schema.org—so I thought I’d turn my attention to another sort of metadata that is useful for artists: image metadata.

What is image metadata and why it is important for artists?

When an image is viewed as part of your web page then this information is usually obvious: it’s typically in the HTML alongside the image. But if someone downloads the image from your web page then link between that textual information and the image itself is lost - it’s just another anonymous image.

Image metadata is information embedded in the image file itself that details the copyright holder and date, the artist, the title of the artwork, etc.

If an image is downloaded from your website then the image metadata remains as part of the image file. Despite being removed from the web page your image will still contains your name, its title, copyright statement, and many other pieces of metadata.

Image metadata can be stripped out, sometimes inadvertently (more on this towards the end of this post), but it does provide one important layer of identifying information that can help you as an artist.

Exif, IPTC, XMP - which metadata standards and fields are important?

When I first starting looking at image metadata I was confused by the various different metadata systems—Exif, IPTC, and XMP—and the enormous range of possible fields in each different system. Here are some quick definitions:

After finding out all of that I was still confused, so I decided to ask an expert—David Riecks, photographer, writer, photo metadata consultant, and project leader at http://PhotoMetadata.org— to advise me on which to use:

Well I didn’t completely stick to David’s recommendations (I can see how keywords would be important for a stock photo gallery, but not so much for me as a non-stock artist), but his recommendations certainly formed the basis of which metadata fields I have targeted so far. I’m still refining my choices, but this combination seems to give good coverage to ensure that the images of my artwork are clearly labelled with my name and copyright detail, with a link to my website for good measure. All this should hopefully prevent one of my images from being considered an “orphaned work”.

So here's how I'm currently adding image metadata:

Exif fields IPTC fields XMP fields Content Example Data
Name Headline Description, Headline {THE TITLE OF THE ARTWORK} Badb Catha Drawing (03)
Copyright, Description, Caption Copyright, Caption Copyright {ARTIST NAME}, {YEAR} Copyright Paul Watson, 2014
Artist Credit Credit {ARTIST NAME} Paul Watson
Source Source {URL OF WEBSITE} https://www.lazaruscorporation.co.uk
Rights Copyright © {ARTIST NAME}, {YEAR} Copyright © Paul Watson, 2014

My process for adding metadata

There are some tutorials on how to embed image metadata at http://photometadata.org/META-Tutorials.

I use Linux (Kubuntu to be specific) so Photoshop isn’t an option for me, and so my process for adding metadata is to use digiKam, a free, open-source application that runs on Linux, FreeBSD, Mac OS X, or Windows.

The process is as simple as opening the image and choosing Image->Metadata->Edit All Metadata and adding your data to the relevant fields. It’s as simple as that.

Test one of my images using an online service

You can use a free online service to see the metadata that you have embedded in your image. One such tool is http://regex.info/exif.cgi - and here’s a link to see one of my images examined using it (opens in new window).

It’s not pretty and it has a lot of the original Exif metadata from the camera I used to take the photograph of that drawing. But it does clearly have my name, the title of the artwork, the copyright statements, and the URL of my website in the image metadata.

Losing Metadata - social media etc.

Unfortunately Metadata that you add to an image can just as easily be removed from it.

Firstly, someone could maliciously remove it for whatever reason - for example, copyright infringement.

Secondly—and probably far more frequently—someone could unknowingly remove it. Some image manipulation software can’t cope with metadata and simply strips it out. Web applications built using image processing libraries such as PHP:GD do this, and many social media websites either use such libraries (or deliberately strip out image metadata in order to slightly reduce file sizes), so be aware that your metadata can be removed if you upload an image to a social media website.

There is some more extensive research on how the various social media sites deal with image metadata at http://www.embeddedmetadata.org/social-media-test-results.php.

There is no way to stop someone from deliberately or accidentally removing your image metadata, but that shouldn’t stop you from adding it to your artwork. It provides another layer of attribution control that means you are identified as the artist.

I have a fairly laissez-faire attitude to non-commercial sharing of photographs of my artwork so long as they are attributed to me (preferably with a link back to my site) because, to quote Tim O’Reilly:

Obscurity is a far greater threat to authors and creative artists than piracy.

In my opinion the more people who see my artwork the better, and so if individuals want to share clearly attributed images of my art on their blogs or social media sites to say how great it is then I’m happy. And also more links back to my website help increase it’s position in search engines.

Because I want my artwork shared widely by fans, image metadata gives me at least some reassurance (with the exceptions mentioned above) that it can be traced back to me if the attribution in the accompanying text is somehow “lost”.

Further Reading

This post only really scratches the surface of image metadata. For further information start with the links below: