I had some opportunity to develop high concept Prezis with various clients in recent months. In doing so I learned a lot about how far you can push the technology and in what scenarios the software plays to its strengths.1
Mind you that Prezi works best as an interface for content you prepare with authoring tools outside of Prezi. If you have a means to create vector graphics in a flash file format, you can enjoy intricate layering and zooming effects to navigate your material.2 As this brief introduction may show you, there are ways to overcome Prezi’s technical limitations for both presenting, and—even more relevant in my opinion—exploring information.
Prezi, much like other authoring tools, still suffers from a lack of widely recognized best practices and commendable design principles. Here’s to hoping I may provide a bit of useful input to their fashioning. More information is in the Prezi itself, and some files for download can be found at my lab.
This is not going to be a post about critiquing a product, even though there is ample opportunity to do so with Prezi. Some of its idiosyncracies are baffling, to say the least, and the kind of design decisions Prezi forces upon its users are a tough pill to swallow at times. But hey, I made it work and so can you. ↩︎
I fully intend to post some tutorials about how to create the right flash files for Prezi in the future. The one tool that makes it possible, other than flash itself, is pdf2swf. ↩︎
Having made the case that we need to look for examples of inventive information architecture on Prezi that acknowledges the functional paradigm of its medium, today I present you this example by Travis Hitchcock. Prominent indicators link pieces of information (nodes, if you will) to each other, maintaining a hierarchy of information that literally and figuratively scales well.1 I find this quite a compelling case for discovering details about typography through zooming.
By the way, the embed code that Prezi churns out really does not play nice with tumblr. I don’t know if all the divs and classes are meant to be helpful so I can apply my own styling or if they are necessary for the Prezi to work (and I’m too lazy to check) but in any case the result is not pretty. I always have to hope that things work out on the front end once I hit the publish button. ↩︎
I pondered a while how I should approach this Prezi presentation about information architecture by Peter Morville. For one thing I really like the content. It gives you a nice impression of what information architects do. But then I feel that it is being let down by the structure. Ironically what we see is an introduction to information architecture that suffers from poor information architecture.
Mind you, I don’t mean to bash the author nor the tool. I just think that it’s time that for Prezi, as for other types of presentation tools and media, we start to appreciate their properties and develop information architecture that caters to their strengths and weaknesses. This particular example, unfortunately, fails to do so.
The problem with the structure of Peter’s Prezi is that zoom or size is not meaningful at all. As is the case with too many Prezi examples already, the information is accessible only through a linear, monosequential roller coaster ride that may induce motion sickness to boot. Try to navigate the Prezi without the guiding help of the path that Peter created. Once the background image loads there is a visual cue that you start from the bottom left and follow a maze, but that does not get you everywhere you need to be to fully understand the content. In effect, you are bound to linear consumption in a way that is worse than what I came to expect from print.1
A page in a book usually features linear design.2 And yet, by typographic conventions such a page already has better information architecture and usability than most Prezis. On a page I can scan the text flow for headlines or keywords and skip to the part that I want to know about. I can’t do that with a Prezi that goes out of its way to hide information away in arbitrary places.
Prezis often look like mind maps that Mr. Magoo would draw under a microscope. Unlike headers in print, the size of the typeface does not give the user a clue about information hierarchy. No indents are there to tell us where a new line of argument may begin, no grid is there to guide our uptake. Most of the time we don’t even have a clue about how far we have progressed in absorbing the content. This isn’t just bad architecture. I’d say that in the worst examples there is no structure at all. It’s Labyrinth of powerpoint slides.
Now, I do not claim to have the definite solution to the puzzle that this new technology poses. I did in the past propose one possible framework to structure information on the infinite canvas on Prezi in a meaningful way, but I can imagine others that work just as well, if not better, at making the form of the content meaningful. Making the form have function. I’d say this calls for Information Architecture with a capital I&A, does it not?
In Peter’s defense: the title itself says that this version is only alpha 0.1 though ↩︎
Text need not be a monosequential medium. Footnotes are an obvious example. Even in print or handwriting do we have highly sophisticated information architecture that defies linearity. I shall revisit the theory of hypertext some other time, but in the meantime I highly recommend you ponder in awe its precursor, the information architecture masterpiece that is the Talmud. In there you find centuries of dialogic discourse among Hebraic scholars condensed into multilinear typography. Yeah. That just happened. Thousand years ago. ↩︎
It’s always good to learn from some one elses mistakes, so if you want to avoid mine, please bear with me. There is a lesson to be found in the following story.
I recently tried to explain the concept of immersion in a Prezi. There was a little contest on their homepage and I thought: “Hey, they are asking what people think about Prezi + iPad and you wanted to explain a concept that is a feature of the user experience for both. Why not create a contest entry about immersion?”
To me immersion is a key defining feature of great interface design and I think the one thing that both the iPad with its touch interface and Prezi with its zooming interface can bring together is the elimination of borders between users and content.
No more mouse fidgeting, but a responsive interaction with virtual elements on a screen by touch is one powerful user experience. The technology fades into the background and you feel connected with the content to a point where the interface feels as if there was none. Freely zooming around in a virtual environment can be an immersive experience by itself, bringing both interface designs together should be a homerun.
Bummer. Prezi only accepts swf or pdf as vector files. No scalable vector graphics, much to my dismay. Worse still, swf import is a tad buggy. I manage to create a decent looking scalable virtual world stacking pdfs with matching backgrounds on top of each other - mind you, there is no transparency support in pdf. Apart from a typo and suboptimal typesetting I finalize the project just in time to beat the deadline. Then my mac dies on me. Here’s where insult is added to injury:
When I check my submission on a different machine, all of a sudden the experience is much less convincing than before. The closer I zoom in on the pdf objects the less fluid the zooming and panning motion becomes. The embedded movie looks jagged. Instead of being invisible the technology makes watching the Prezi almost unbearable. Surely not very immersive.
I can only assume that embedding several pdf on top of each other is a resource hog. Even though it looked good on my tricked out machine at first, it does not really work on every kind of hardware. And immersion only works if you do not get distracted by any glitches in the matrix.
So the moral of the story: Unless you control for every step of the way of your delivery, there is always an open flank for murphy’s law to catch up with you. Unfortunately nature always sides with the hidden flaw. Keep that in mind when you try to innovate.
Here’s the Prezi in question. Your experience may vary.
In this experimental showcase I have applied an optimality approach of communication architecture to Prezi, a design tool that is based on the concept of the “infinite canvas.” The constraints that govern the design process in Prezi are rather straightforward and intuitive, most of them are functionally motivated.
I tried to embed a set of semantic rules in the communication paradigm of Prezi, the most important of which is to make scale a semantic category, rather than a dimension of information sequence or even decoration. I hope that the experience speaks for itself - there are four layers of meaning, where the zooming in and out of details renders different information legible or intelligible.
The embedded Prezi-presentation works best when following the predetermined sequence at first and when viewed in full screen. I have decided to still zoom down a gratuitous step onto the text boxes for those who might have trouble reading. I am sure that there is a lot to improve in this showcase and that there are many other principles to be found when designing with Prezi, but I hope you like what I could come up with so far.