is a blog about design, technology and culture written by Khoi Vinh, and has been more or less continuously published since December 2000 in New York City. Khoi is currently Principal Designer at Adobe, Design Chair at Wildcard and co-founder of Kidpost. Previously, Khoi was co-founder and CEO of Mixel (acquired by Etsy, Inc.), Design Director of The New York Times Online, and co-founder of the design studio Behavior, LLC. He is the author of “Ordering Disorder: Grid Principles for Web Design,” and was named one of Fast Company’s “fifty most influential designers in America.” Khoi lives in Crown Heights, Brooklyn with his wife and three children. Refer to the advertising and sponsorship page for inquiries.+
Everything They Ever Said
This attitude is apparent, too, in “The Complete New Yorker,” an exhaustive archive of the magazine’s first eighty years of publication that ships on eight jam-packed DVDs. Weighing in at some thirty-odd gigabytes, this collection is gorgeous and somewhat unbelievable: every word and image that the magazine printed between February 1925 and February 2005 — including the text of every article, the layout of every advertisement and the illustration and caption of every cartoon — is available here in high resolution for less than US$100. Taken on the indisputably rich merits of its content, it’s less a bargain than an outright steal.
And yet, the compendium falls down mightily on the job of providing a coherent user interface for users to access all those articles. Mightily.
I was lucky enough to receive a copy of “The Complete New Yorker” as a holiday gift this past December, an item I had pushed to the top of my Amazon wish list in spite of having clicked-through the online demonstration of its user interface beforehand. Even in that canned preview, the product looked kludgey and inelegant, so you could say that I knew what I was getting into. But the temptation of the content was too great.
The Flash demo makes little or no effort at hiding the fact that the collection’s search interface is confused, but upon closer inspection a better term might be ‘tortured.’ Its fundamental problem is that it attempts to present the search interface and the results listings simultaneously, attempting to emulate the casual nature of a content browser while remaining mired in the awkwardness of a parametric search.
Users select search terms from scrolling lists of authors, dates and issues, etc. in a top-left to bottom-right interface that suggests a logical sequence. But changes to parameters later in the sequence can alter selection criteria earlier in the process, causing considerable frustration. In effect, any of the parameters can be the primary axis upon which the other parameters further refine results, but there’s almost no intuitive indication that this is so.
State of Confusion
It might be true that a user can become well-enough acquainted with the interface to use it competently after repeated, fitful exposure, but they wouldn’t be helped by the lack of attention to detail apparent throughout.
For instance, the archive’s designers seem not to appreciate the value of clearly indicating state; buttons have no on- or off-state to speak of, so it’s impossible to tell at a glance whether they’ve been activated. Articles and pages themselves are displayed in a viewer with two modes, “flip mode” and “read mode,” which should really be displayed as a single toggle, but are instead rendered as two separate buttons, with no indication of which mode a user is viewing. There’s also an unfortunate and impolitic use of red as an emphasis color which inadvertently suggests errors or warnings where nothing is amiss: I’d wager there will be plenty of users who see the label for “Complete Archive” in bright, alarming red and immediately register a moment of panic.
Remember That One Article about That Guy Who Did That Thing?
It’s also difficult or, at best, unpredictable to search the archives without some specific knowledge of an article’s attributes. The first piece of content I searched for was an article I’d read several years ago about a 19th century confidence artist; I couldn’t recall the title of the article or any specific details, and in fact it turned out to be about an early 20th century grifter. Entering in “con artist” returned every item about artists in the database, including just about every one of The New Yorker’s thousands of cartoons. It took me ten minutes to hunt down my article successfully.
Clearly, “The Complete New Yorker” suffers from either a lack of understanding of or an under-appreciation for the art of search. In the end, that shortcoming must have been so apparent even to the publishers that they felt compelled to address it somewhat defensively and with unintentional comedy in the accompanying section of frequently asked questions:
This search is not as fast as Google. Why?
Because of the way this information is stored, archive searches may be significantly slower than online searches.
What It Needs to Be
It’s true that any search interface in this day and age will inevitably and perhaps unfairly be compared to Google, but even setting that aside, The Complete New Yorker’s interface remains a disappointment. I’ve only touched on a handful of the dozens of low-level interface missteps that complicate its use. The product is literally a gold mine of content, and yet in total, it remains difficult to mine it. In honesty, it’s my opinion that the fact that it’s spread across eight DVDs (really no different except in quantity from the CD-ROM format in which it might have shipped ten years ago) prevents it from ever becoming a completely elegant experience.
This product really needs to exist online, but the magazine’s slavish dedication to exact reproductions of every printed page prevents that. None of the articles are available as ASCII text. Rather, every page of every issue is stored as high-resolution images, a decidedly old media way of thinking about new media that perfectly embodies The New Yorker’s digital ambivalence. Content can’t be selected, copied or pasted, which obviously reflects a desire to protect against copyright violations. But it also prevents truly meaningful searches of the database; when a search term is entered, the engine scours only the article abstracts, so if a key detail happens to be missing from those abstracts, it won’t appear in the results.
To be fair, there’s something to be said about the historical value of seeing articles presented in their original context, next to the advertisements with which they were originally printed, and in the flow of the issues in which they were originally published. Given the choice, though, I would just as easily have spent money on a text-only archive available online that could be accessed with a much more powerful search interface. For that matter, I would have paid a monthly subscription fee for such a product with Google-like search capabilities. But, as I said, this was a gift, and for the fact that I have some kind of access to this archive, I’m grateful.+