Evidence and Antelopes: Buckland’s “Information as Thing”

Periodically, we’re taking a classic paper in information science and breaking it down into its argument, some clear examples, other perspectives from the field on the same issue, and what it can mean for an information architecture practice today. This week, we’re looking at Michael Buckland’s “Information as Thing.” [PDF]

Most subjects can be taught in at least two ways: starting from what’s easy to learn or starting from the fundamental building blocks of what makes it true. In math, you can start with addition or you can start with what numbers are. In information architecture, you can start with how to make navigation a little bit better or you can start with what information is and why that question matters. Seeing as we’re all stuck inside for the moment, we’re going to indulge in a little theory.

In “Information as Thing,” Buckland argues that there are three types of information:

1. Information-as-process: When somebody is informed of something, what they know has changed. That change is information-as-process: I read a news article to find out what happened in the American presidential primary election. I have gone through a change in what I know. This change is the process of becoming informed.
2. Information-as-knowledge: As somebody becomes informed, they acquire a new intangible fact. This intangible fact is information-as-knowledge: From reading that article, I now know that New Hampshire had record turnout for the Democratic primary. That is a new intangible fact that I have. This intangible fact is information.
3. Information-as-thing: Things are used as evidence, to create knowledge and facilitate a process: I went through this change in what I knew and acquired an intangible fact by interacting with a tangible object: a news article. This article is also information.

Like in the example I’ve used above, you can see how this works neatly with traditional documents like books, articles, and other media. We also get information, however, from other objects that are not primarily intended as information. For instance:

1. Last night, I also looked at an arrangement of bottles in my refrigerator, becoming informed about what was there. (Information-as-process).
2. Based on looking at the bottles in my refrigerator, I gained the knowledge that we had ketchup. (Information-as-knowledge.)
3. I went through this change in what I knew and acquired an intangible fact by interacting with a tangible object: A bottle of ketchup. (Information-as-thing.)

In this example, that bottle of ketchup is information just like a news article is, because it helped me gain knowledge and therefore become informed.

Research is often the process of extracting information-as-knowledge from objects in the real world, or using them as evidence for a fact or an idea. Therefore, any object that can be used as evidence for a fact or idea is information-as-thing.

This means that whether or not an object is information is contextual. An object that was always intended to be informative, like a news article, might always be information, but other objects can change depending on their context. To build off of the example Buckland gives in his paper, an antelope existing in the wild is just an object, but an antelope that has been observed by researchers and used as evidence for the average size of an antelope is now information.

What kind of information an object is also contextual, even for traditionally informative objects. To a news reader, a news article is information about what happened in New Hampshire last night. To a media researcher, the same news article is information about what narratives are at play in the 2020 election. When it comes to other kinds of objects, information also varies. To a zoologist, an antelope might be information about the size of antelopes. To a geneticist, an antelope might be information about the heritability of coat color. To an artist, an antelope might be information about light and shadow in space. These are all equally correct, and part of the value of information-as-thing is its ability to have different meaning in different contexts.

As an aside, objects can be both traditionally and non-traditionally informative, especially as time passes. I remember an anecdote from one of my grad school readings from a researcher who spent a summer sitting in a dusty archive in Europe, reading handwritten merchants’ accounts from hundreds of years ago. After many weeks of doing this, another researcher joined her in the archive, but didn’t read any of the accounts, he just deeply smelled each one and occasionally made notations in his own notebook. Completely aghast, she finally asked the newcomer what he was doing, and he explained that he was studying the periodic plagues that would strike the city. Each time it did, the merchants would douse their papers in vinegar to try to stop the illness from spreading. They didn’t write down that plague was in the city, but they didn’t have to, because their accounts from that time still smelled of vinegar. Those objects were doing double-duty as information.

The difficulty with this definition of information, as Buckland admits, is that it means we’re unable to confidently say that anything couldn’t be information:

“If anything is, or might be, informative, then everything is, or might well be, information. In which case calling something “information” does little or nothing to define it. If everything is information, then being information is nothing special.”

Unsurprisingly, there are many competing definitions of information which get more specific about what precisely is and is not information. Claude Shannon famously provided a very precise and mathematical definition of information that’s essential for certain areas of engineering [PDF]. Karl Popper and others working in epistemology have defined an ontology that makes “information” and “knowledge” synonyms and limit it to the world of books, documents, and scientific theories.

These days, if you’ve come across an IA-focused definition of information, it’s probably Abby Covert’s, from How to Make Sense of Any Mess. She refines some of these ideas to be much more accessible and allow for an important new distinction. Covert divides information-as-thing into content and data and defines information solely as information-as-knowledge, acquired through interpretation, or information-as-process:

“The most important thing I can teach you about information is that it isn’t a thing. It’s subjective, not objective. It’s whatever a user interprets from the arrangement or sequence of things they encounter.”

Where Buckland is largely concerned with knowledge that might be gained from a single object, Covert connects this more directly to the knowledge that might be gained from several objects and the relationships between them.

To return to my previous silly example, a ketchup bottle in my fridge is evidence that I have ketchup. When seen in sequence with other bottles, it might be evidence that we have too much ketchup and need to stop buying it or that we have plenty of condiments for hot dogs.

Covert also adds an extremely important refinement to this idea:

“The difference between information, data, and content is tricky, but the important point is that the absence of content or data can be just as informing as the presence.”

Not only can an arrangement of bottles in my refrigerator inform me that we do have ketchup, but it can also inform me that we’re out of soy sauce. Extrapolating from there, we can easily imagine how I can interpret other kinds of information from sequences or arrangements of things: Something is the most important, something is here that doesn’t belong here, these are the kinds of things that this place has.

If everything can be information, and information is actually everything, and knowledge, and a process, what use is any of this? It can be overwhelming, but I think it helps in three ways.

Firstly, it helps us remember that people actually mean lots of different things when they talk about information, and we can be clear about what we mean in each instance. When a product designer says that a user is on a page to “get information,” we know that means that they’re expecting the user’s knowledge about the world to change in some specific way while they are on that page, and we can figure out how to help.

Secondly, it means that we can be specific and realistic about what it means to architect information. As an information architect, I have no access to the intangible ideas in people’s heads or the changes they’re undergoing as they interact with a product. I can, however, create sequences and arrangements of things. I can even evaluate those sequences and arrangements of things with users to see whether information-as-process and information-as-knowledge are happening in the way we expect them to.

Lastly, Buckland’s definitions help remind me that information architecture is not a digital phenomenon. If ketchup bottles and antelopes are also information, it means we can and should architect information in the real world, too. Many products and experiences we work on exist both in digital and real-world contexts, and we need to ensure we are arranging things so they’re conveying the same information, regardless of their context.

As esoteric as some of this might seem, it’s essential to start with antelopes and ketchup bottles because information architecture is older, wider, and deeper than the internet. Too often, I hear “information architecture” used as a synonym for “sitemap.” Most of us don’t build sitemaps (honestly, as a full time information architect, even I don’t), but taking this wider perspective, it becomes clear that almost all of us are doing information architecture all the time. By thinking about what information really is, by engaging with the disagreements about it, we can start seeing what we’re doing and do it better.