July 11, 2017

Imagine: Interoperable images!

Stack of photos isolated from the background

It’s old news that we have become image-crazy in the past few years. Now that so many of us carry excellent cameras in our pockets—not to mention video recorders—more and more communication is done through images and videos. Remember when Twitter was just 140 text characters, back when the dinosaurs roamed the earth?


Why would you spend time trying to describe something when you can just show it? But how can we actually do things with images, rather than just showing them—with the flexibility and dynamics of what we can do with text?


Much of that, of course, is done in apps, and is done in casual communication. When it comes to publishing, images are mainly illustration or decoration. Even images that convey information are almost all static. The real work in most publications is still overwhelmingly done by text.

That’s rapidly changing, because user expectations are changing. Why would you spend time trying to describe something when you can just show it? (Accessibility is one reason; another of my blogs addresses that.)

Making images dynamic

But how can we actually do things with images, rather than just showing them—with the flexibility and dynamics of what we can do with text? What about something as simple as highlighting the part of an image you’re talking about?

And by the way, needing multiple formats and resolutions and sizes of an image—a thumbnail to click on, a decent size for your phone that won’t take too long to load, high resolution so you can zoom in on details (separate files, are you kidding?)—is a pain.

I have good news for you. There has actually been a lot of work done on this, by an organization that hardly anybody in publishing has heard of so far: the International Image Interoperability Framework, known as IIIF (pronounced “triple-I-F”).


Now we can exchange these supercharged images, with all their enhancements, with others using standard web technologies.


This comes out of the cultural heritage sector: mainly libraries, museums, universities, and similar institutions—many of the leading such institutions in the world. These folks have enormous repositories of content that have increasingly been digitized. But “digitized” often just means “scanned”: pictures of the objects. Even publications like newspapers are often available only as a collection of scanned pages. Think of all the different things on a single page of a newspaper! A user isn’t interested in the page; she’s interested in something on the page: an article, an ad, an image.

I first learned about IIIF in a presentation by Rob Sanderson to the IPTC, an organization I belong to that comes out of the news sector. I immediately saw how valuable their work would be to all types of publishing, so I recruited Karen Estlund, Associate Dean for Technology & Digital Strategies at the Penn State University Libraries and a member of the IIIF Coordinating Committee, to present at a session I organized for the SSP Annual Meeting in June. (You can see her presentation here.)

Breaking images out of their “format cages”

I loved how Karen articulated the problem statement: most images are “format cages” (think of those newspaper pages), often in PDF and often requiring a plug-in or other special software to be viewed. IIIF’s mission was to form a community of interest (now hundreds of organizations and individuals) to develop APIs and open source software and services to make images dynamic and interoperable.

The Image API, for example, enables sizing, zooming, panning, mirroring, and rotation in a standard, interoperable way so that the same image file can be used for that thumbnail, on the phone, or in the full resolution a scholar needs.

In addition to accommodating a wealth of structural, descriptive, and rights information about the object the image represents, the Presentation API enables transcriptions, translations, commentaries, etc. to be associated with the image or specific regions of the image—even irregular shapes.

And the Search API enables searching all those things: text within an image as well as the transcriptions, translations, and commentaries associated with it.

IIIF APIs enable supercharged, interoperable images

Now we can actually get to the things we want in that newspaper page—or that manuscript, or artwork, or diagram—and do things with them.

Best of all, we can exchange these supercharged images, with all their enhancements, with others using standard web technologies.

The work is ongoing, of course. In addition to getting ever more granular and to accommodating ever more types of images, IIIF is working on 3D images and time-based media like video.

All standards-based. All open. All shareable and interoperable.

Images doing real work! Imagine!