I’m reading an interesting book by Janet Murray called Inventing the Medium: Principles of Interaction Design as a Cultural Practice. She is articulating things that I’ve thought for a long time but is also surprising me a lot, making me think about things in new ways. The book is about the digital medium and how objects that we use in this medium influence the way we think about the world. She argues that technological change is happening so quickly that our design for the medium hasn’t kept up. Designers use the conventions that work well in one environment in a different environment without really thinking about whether those conventions make sense in that second environment. As a result we get user interfaces (which is a term she doesn’t like but which I’ll use because most people interested in these things have a pretty good idea of what we mean by the term) that are far too complex and difficult to understand.
One idea that strikes me as particularly important and useful is Murray’s argument that designers create problems when they separate “content” from the technology on which the “content” is viewed. Like McLuhan, Murray believes that “the medium is the message,” by which she means “there is no such thing as content without form.” She goes on to explain, “When the technical layer changes, the possibilities for meaning making change as well.” In other words, if you change the device through which you deliver the content, the tools needed to help consumers understand that content should probably also change. My favorite personal example of the failure of this idea is the Kindle, Amazon‘s e-reader. I’ve owned my Kindle for about three years and I mostly love it. One thing that feels problematic to me, however, is the reporting of where you are in the book that you’re reading. Printed books are divided into chapters and pages and it is easy to see how much further the reader has to go to the end of the book. Readers trying to read the same book might have difficulty if they are using different editions because page numbers won’t match up but the divisions into chapters should still be the same. If a page of text in a physical book corresponds to a screenful of text on an e-reader, page numbers don’t really make sense in e-books, mainly because the reader can change the size of the font so that less or more text is able to be shown on the screen at a given time. This means that the Kindle doesn’t have page numbers. But readers probably want to be able to jump around e-books just as they do in physical books. And they want to know how much progress they’ve made in an e-book just as they do in a physical book. So Amazon introduced the idea of a “location” in their e-books. The problem with a “location,” however, is that I have no idea what it corresponds to in terms of the length of the book so using locations doesn’t give me a sense of where I am in the book. For that purpose, the Kindle will tell me the percentage of the book that I’ve currently read. I think the problem with these solutions is that the designers of the Kindle have pretty much used the idea of pages, changed it only slightly and unimaginatively, and it isn’t as informative in the digital medium as it is with a physical book. I don’t know what the solution is but Murray suggests that the e-reader designers should think about the difference between “content” and “information” in their design.
Murray distinguishes between “content” and “information” and thinks that device designers have problematically tried to separate content from the technology on which this content will be viewed. So the designers of the Kindle see the text of the book as the content, something they don’t have to really think about in designing their device. Instead, Murray suggests that they focus on information design, where the content, which in this case is the text, and the device, in this case the Kindle, cannot be separated. The designers should think about the affordances provided by the device in helping to design the information, which is meaningful content, with which the reader will interact.
Another example appeared in my Facebook timeline last week, posted there by one of my friends pointing out the fact that the Mitt Romney campaign is insensitive at best and hostile at worst to women. The post is a video of Romney’s senior campaign advisor Eric Fehrnstrom, appearing on This Week with George Stephanopolous a week ago, calling women’s concerns “shiny objects of distraction.” Watching it, I was annoyed and horrified by what I was supposed to annoyed and horrified by. But I also noticed the ticker tape Twitter feed at the bottom of the video. The headline-type feeds at the bottom of the screen on television news have become commonplace, despite the fact that they don’t work particularly well (in my opinion). I’ve always felt that the news producers must know that the news they are presenting is boring if they feel they have to give us headlines in addition to the discussion of the news anchors. But in the video of Romney’s aide, the rolling text at the bottom of the screen is not news headlines but a Twitter feed. So the producers of This Week have decided that while the “conversation” of the show is going on, they want to present the “conversation” that is simultaneously happening on Twitter about the show. There are several problems with this idea, not least of which is that most of the tweets that are shown in the video are not very interesting. In addition, the tweets refer to parts of the program that have already gone by. And finally, the biggest problem is that the Twitter feed recycles. In other words, it’s not a live feed. They show the same few comments several times. Someone must have thought that it would be cool to show the Twitter conversation at the same time as the show’s conversation but they didn’t bother to think carefully about the design of that information or even which information might be useful to viewers. Instead, they simply used the conventions from other environments and contexts in a not very useful or interesting way.
Another of Murray’s ideas that strikes me as useful is the idea of focusing on designing transparent interfaces rather than intuitive interfaces. Intuition requires the user to already understand the metaphor being used. In other words, the user has to understand how an object in the real world relates whatever is happening on the computer screen. This is not particularly “intuitive,” especially for people who don’t use computers. I’ve been thinking about the idea of intuitive interfaces since I started teaching computing skills to senior citizens. For them, it is not “intuitive” that the first screen you see on a Windows computer is your desktop. And once they know that, it isn’t “intuitive” to them what they should do next because it’s all new to them and so they don’t have a sense of what they CAN do. For example, they can put a piece of paper down on a real desktop. Metaphorically, you can put a piece of paper (a file) down on the Windows desktop but the manner in which you do that is not “intuitive.” The first question I always get when I talk about this is: How do I create a piece of paper to be put on the desktop? Of course, that’s not the way they ask the question. They say, “How do I create a letter?” That’s a reasonable question, right? But the answer depends on lots of things, including the software that’s installed on the computer you’re using. So the metaphor only goes so far. And the limitations of the metaphor make the use of the device not particularly intuitive.
Murray argues that focusing on “intuition” is not what designers should do. Instead, designers should focus on “transparency,” which is the idea that when the user does something to the interface, the change should be immediately apparent and clear to the user. This allows the user to develop what we have typically called “intuition” as she uses the interface. In fact, lack of transparency is what makes many software programs feel complex and difficult to use. Moodle, the class management system that my University uses, is a perfect example of non-transparent software. When I create the gradebook, for example, there are many, many options available for how to aggregate and calculate grades. Moodle’s help modules are not actually very helpful but if the software was transparent, that wouldn’t matter. I would be able to make a choice and immediately see how it changed what I was trying to do. That makes perfect sense to me as a way to design software.
This book is full of illuminating observations and has already helped me to think more clearly about the technology that I encounter.
As someone who designs and create software help systems I can totally relate to this post. Providing the correct level of help for complex design software is not only challenging, but delivering it in a way that is consumable presents other challenges. Breaking help systems out of the “book” paradigm has been a big challenge as that’s what users are familiar with, but the same metaphors do not work well for digital content. Today’s users expect to be able to access the same information in any format delivery mechanism they choose (mobile, desktop, tablet, etc…) and yet the content they’re consuming isn’t easily adaptable for all of those devices. Another interesting observation that you touched on is that expectations change based on age and generation. Some users feel that they’re not getting all of the same help content that they used to because they’re no longer getting a printed manual, yet all of the exact same text exists in a digital format. Other users want easily searchable text with lots of video and the ability to turn that content into a custom help system. The gap between the two experiences (printed book and interactive digital content) can be quite large.
Thanks for getting the gears turning for me this morning.