Error: SQLSTATE[HY000]: General error: 1030 Got error 28 from storage engineError: SQLSTATE[HY000]: General error: 1030 Got error 28 from storage engineError: SQLSTATE[HY000]: General error: 1030 Got error 28 from storage engineError: SQLSTATE[HY000]: General error: 1030 Got error 28 from storage engine What Makes A Good Poster?: Design Observer




11.08.06
Jessica Helfand | Essays

What Makes A Good Poster?


"Designing Effective Posters," Design by Jeff Radel, PhD., Department of Occupational Therapy Education, University of Kansas Medical Center.

At its core, the school science fair of my youth was an excuse to engage in two of my favorite activites: making things and showing off. This was arts-and-crafts for credit — science credit, no less — which was a dream come true for scientifically impaired children like me. It was also a good reason to load up on art supplies — in this case, lots of poster board and markers, cotton balls and tape. Today I look back on this stage in my early artistic development as my diorama period, a time in which my scientific aptitude (non-existent) was trumped by my pre-teen bravado (quite persistent), and the empty shoebox was my medium of choice.

Eventually I satisfied my desire to show off by studying theatre and my need to make things by studying graphic design. With some minor deviations along the way, I've remained clear about where these disciplines diverge. And as I became more serious about a life in design, I let go of the cotton balls and the Elmer's glue, those long-ago staples of my science projects. Poster board, I learned, does not a poster make.

Apparently everyone does not understand this distinction.

Posters, the original public art form, date back as far as the 15th century. Their arrival ushered in a period of change that began with replacing the Town Crier and would later come of age, in the 18th and 19th centuries as mechanization (and color presses) allowed for better printing and wider distribution. Over the course of the last century, posters have reflected the efforts of world-class painters, printmakers and graphic designers. And while something of an endangered species in a world driven by time-based media, the poster remains an enduring hallmark of modern visual achievement, reflecting the changing social and cultural values of a generation: from Jules Cheret to Peter Max to James Victore, the history of the poster is the history of modern civilization.

Meanwhile, in a parallel universe, there have been numerous efforts, at schools around the globe, by scholars preaching the gospel of good poster design with nary a design professional in sight. Laugh if you must, but do a search on "what makes a good poster" on Google, and barring the Information Architects (Mike Lee, we applaud you), the results may surprise you.

A Professor in East Asian Studies at at Lewis and Clark State College recently launched a website in which she extols the virtues of posters for scholarly presentations. Her efforts are being duplicated in campus-wide efforts by scholars at other institutions to do the same — and while there is nothing legally objectionable about this process (yes, I know, you don't have to be certified to practice graphic design) it assumes a level of rudimentary knowledge that is extraordinarily off the mark. IT'S AS IF FIVE HUNDRED YEARS OF POSTER DESIGN HISTORY NEVER EXISTED. Worse, there's a science fair element at play here (lots of poster board, and professors posing idly by their printouts) that lends itself to ridicule, but that's actually not my intention here. Rather, in the spirit of collegial critique, I'd like to examine some of the more critical formal and procedural flaws in this otherwise noble initiative.

Why consider presenting research as a poster? Herewith, some reflections.

No. 1 : Increased Participation by Scholarly Society Members.

Translation: Audience.

I have never designed, nor have I ever witnessed a student or a colleague design a poster without an extraordinary amount of editing, weeding out, narrowing down, and critically evaluating everything. Such rigor is fundamental to the design process, and usually results in a better poster — a better anything, for that matter. Central to this editing process is the basic understanding of who you're designing for. A Fortune 500 company or the Girl Scouts? There's no question that scholarship plays a part in rigorous thinking, but scholarship, in and of itself, does not directly lead to clear (or even abstract) formal execution. That process demands knowing who you're talking to. Period. And good kerning. But I digress.

No. 2 : For many researchers, the poster format is superior for the presentation of their data.

Translation: Form and Content.

Much has been written in the design press, of late, of the 8.5 x 11-ization of the surface, demonstrated most recently (and quite brilliantly) by last year's graduating MFA students at Yale. But to look at the visual evidence here, it appears that "poster" is being loosely used to refer to a xerox printout hung on a bulletin board, without taking into consideration the content and the particular form that content might take. (And frankly, even 8.5 x 11 can look great. These don't.) Certainly a poster produced, say, in Italy in the 1930s is going to look different from something made in Haight Ashbury in the 1960s. But what if they're both public health posters? What if they're not? In this view, the poster spins on a fascinating, collaborative equation that merges form with function, style with substance. This explains why all posters are not created equal. Nor, frankly, should they be.

No. 3 : The individual presentation time of a panel, i.e. 15-20 minutes is often not enough time.

Translation : Scale.

Posters generally succeed best when they're big, when their sheer scale results in a kind of physical impact that proves striking, arresting and memorable. This is not to say that posters can't be small: on the contrary, the reduced typographic broadsides of the early-to-mid Nineteenth Century are as compelling as the celebrations of minimalism produced in Europe in the years after World War II. What these posters share is a kind of attention to detail, that is typically possible in the hands of a skilled designer. (See "kerning," above.)

No. 4 : Poster sessions are a consideration to foreign colleagues.

Translation: See (No.1), Audience, above.

To be fair, Professor Levine offers some compositional tips that are not without merit — she advocates sketching beforehand and paying attention to white space, among other things — but overall, there's rather a missed opportunity here. As an academic herself, it seems she might avail herself of the expertise likely available on her own campus — the kind, for example, that students might offer. No doubt the design students would benefit from the experience, and Levine and her peers would get better posters. Sure, a poster titled: "Curable Cancers and Fatal Ulcers Update: Changing Attitudes Toward Cancer Disclosure in Japan" might not conjure the thrilling appeal of, say, a poster for a David Mamet play, but the basic rules still apply. Content. Form. Scale, Audience. Typography. Image. Hierarchy. Composition. And as required by so many scientific posters, Charts and Graphs — the poster-sized visual articulation of which are, in all fairness, deserving of their own post. (But not by me.)

So, here's a thought for really designing effective posters: why not launch a poster task force, with real designers on board? The Design Workshop at Parsons, for instance, allows students to realize a design project by providing free architectural and construction services to nonprofit organizations. A similar approach — teaming design students with scholars in the humanities — would invariably benefit those academics looking to elevate their work in poster form. And it would minimize the science fair quality that, while charmingly anachronistic, is kind of goofy. Scholars deserve better. Designers deserve better. And the poster, quite frankly, deserves nothing less.





Jobs | December 11