Zach Beauvais

Data as metaphor

Written by Zach Beauvais

Dec 18, 2008

I have talked a lot about metaphor, both here and, perhaps sadly, to my friends and family. Metaphor and the abstract are true passions of mine, and I can’t help but see them everywhere. I suppose, it’s the nature of metaphor to be everywhere, really.

The essence of metaphor is understanding and experiencing one thing in terms of another.

Lakoff and Johnson (1980)

So, seeing (or, “experiencing”, since “seeing” is really a metaphor) one idea or concept in terms of another is a kind of abstraction. You’re essentially changing your perspective on something by bringing in another concept. Metaphor, generally, is about comparison and noting the similarity, but I suppose there can be an element of the dissimilarity which makes them work. So, if I use a literary metaphor (comparing two things without the use of a similating word like “like” or “as”) and say: “this computer is rubbish”; I’m fundamentally making a comparison between the two notions—”this computer” and “rubbish”. It is the similarity which I am stressing; and, on the surface, using “rubbish” as a sort of modifier of the computer.

However, there is a whole plethora of meaning in this statement, if you pull yourself back from it a bit. What’s rubbish? Rubbish is stuff we throw away; it can smell bad; it’s collected from our houses and fills holes in the ground; we don’t want rubbish; we don’t like rubbish; it’s a generic term for things we don’t like or are unhappy with. With this simple statement, I’m ever-so-casually bashing together large quantities of information and notion, and letting the meanings fall where they will. Inside this somewhere is the idea of “propositionality”, meaning that I’m letting the hearer of this statement draw their own conclusions to what I’m saying (he’s not happy with his computer, his computer may not be very good, he wouldn’t recommend it, he’s having a bad day…) some of which is intended, some of it not (at, least, not consciously). There are also cultural considerations in that there is a sort of social consensus that this metaphor “works” and that we must not literally interpret this statement as an intention to physically dispose of an object (which is good when you consider any time you’ve ever heard a person “understood or experienced in terms of” rubbish 😉 This leads me to think that there are also elements of disassociation between the two concepts, so that some of the meaning is actually in the difference between “rubbish” and “this computer”. I’m probably not going to throw it away (at least, not immediately); It’s probably something I’ve bought and have no intention of burying in the ground; I expect to be happy or satisfied with it (whereas, you wouldn’t about a used tea-bag). So, the two concepts modify each other, they’re like points in a perspective, making it possible to glean added meaning from the situation which is greater than just the two ideas themselves.

(if you’re still reading by here, email me, and we’ll have a pint!)

I mentioned in my previous post that data are used in abstraction. What I mean by this, is that a bit of data is “used” in a process when it’s a point of reference for something. This number + that number = another number, the two numbers are reference points for the sum. When I say: “I’m busy on the 2nd” it means that I’ve referred to a bit of data (a number on a calendar application, an email, or whatever) that I’ve used as a point of comparison. I’ve essentially understood the projected state of my schedule in terms of what i’ve already planned to occur. And, these bits of data are more and more powerful when the perspective you gain from them is more accurate.

When we get more reference points, and more interactions, our perspective becomes more flexible. We can abstract ourselves right out, and look at a very broad picture. Google’s Pagerank does this by mining clickstream information from a very, very large dataset using very simple reference points: the number of links to an item increases its position in the rank. Conversely, we can focus right in on a single notion or dataset using as many different references as possible to understand a limited set of transactions. Amazon’s book results page is full of this kind of perspective with user-ratings, purchase histories, browsing behaviour, and mathematical algorithms to give a very full picture (and options to accomplish a task) of a single notion.

So, I think that data are very similar to metaphor in that they are used to understand one thing (or set of things) in terms of another (or others).

The upshot of this is that we can refer to this abstract concept of “data” in terms that help us to understand both their significance and their utility to us. When I say: “I want my data to do this”, it’s not that helpful unless you understand that I’m trying to get all my reference points to produce a perspective to help me accomplish something. Which leads me to my main point: the whole point of metaphor is to help—or possibly enable—our understanding. Data should do the same. Collecting all the bits and pieces of information you incur by being a person and doing things should bring you some form of understanding leading to a benefit.

Related Articles


Is everyone a content creator?

A couple weeks back, I spoke at a client conference for Zengenti (where I work). We split the day into two broad streams – one focused on developers and sysadmins, and the other for "content." It's been interesting working for a CMS vendor, and I like the fact that...

read more