My CCCC presentation is brief enough to share here in slightly edited form; it’s a written-out-talk intended to precede a roundtable, so it may read a little breezily. Here’s a summary:
Readymades have re-emerged on the web, displacing the “web site” with the “web presence,” in the templates and skins of much web-based software, and also in the workaday of social networks. Three patterns dominate use: collection, juxtaposition, and combination. For theories of use and usability, there are four implications:
- Finding a single, clearly defined, easily anticipated use, as suggested by many usability practitioners, becomes more challenging;
- We can’t always assume users are interested and motivated;
- The nuts and bolts of use need even more attention;
- Readymades themselves are rare; the new readymade is designed for customization and individualized use (whether or not that is realized).
Now, the whole thing:
The readymade has returned to the web in a big way. Readymades are all over the web. Increasingly, the single-node web site as we know it has been displaced by web presences made up of sites and services which users select and aggregate to meet their needs. This is true for individuals, corporations, organizations, universities. We are all familiar with the mechanisms by which this takes place: weblogs, Twitter, Facebook groups.
In The Language of New Media Lev Manovich described this process of creating by selection and argued that it was the core mode of media creation. He writes:
New media objects are rarely created from scratch; usually they are assembled from ready-made parts. Put differently, in computer culture, authentic creation has been replaced by selection from a menu. In the process of creating a new media object, the designer selects from libraries of 3-D models and texture maps, sounds and behaviors, background images and buttons, filters, and transitions. (124)
Drawing on the work of Baudrillard and other scholars, Manovich also connected this to identity:
The process of art making has finally caught up with modern times. It has become synchronized with the rest of modern society, where everything is assembled from ready- made parts. Whether assembling an outfit, decorating an apartment, choosing dishes from a restaurant menu, or choosing which interest groups to join, the modern subject proceeds through life by selecting from numerous menus and catalogs of items. (126)
That’s exactly what happens with the web presence. Our online identities are built from stichings together of readymades from WordPress, YouTube, and other sources. For Manovich and many critics of consumerism, this is regrettable. And he suggested we could resist this movement by refusing customization, which he characterized as a “‘changing collage of personal whims and fancies’ mapped out and coded into software by the companies.”
For me, resistance by using the default isn’t viable. I’m interested in the questions of identity Manovich raises, but more interested in the use of readymades on the web, and implications for usability testing and user centered design in particular. Use has been a part of the question of readymades since day one: Duschamp didn’t pick a urinal arbitrarily. Today’s readymade is an engine of customization, all of the things Manovich describes and more. It is designed to be customized, connected, and fractal. And frankly I see no way we can not engage it, apart from walking away from the web. Quite a few technologies have been developed to facilitate the use of readymades, their selection and customization. Templating systems, skins and themes, and plug-in or modular architectures are all over the web, both embedded in software like WordPress. Selecting ready-made add-ons is core part of the functionality of social networks. Sites like Flickr and Google Maps outside their boundaries by offering code snippets which can be cut and pasted into other sites. Aggregation by RSS, Atom, SOAP, REST, or similar methods are obviously the most important technology which distributes readymades.
I see three main patterns of use:
First, collection. There are a huge number of sites based in whole or in part on collecting found objects. Delicious, Reddit, Digg, and a host of others for pages and sites: anything with a URL is a readymade which can be found and refound by Delicious-type sites. Ratebeer and Cork’d do the same for beer and wine. LibraryThing does it for books: you can literally find books in others’ libraries and adopt the information they’ve entered about the book as your own. And of course collecting applies to the friends or contacts of social networks. So one of the core uses of readymades is simply accumulating a store of them, like a commonplace book or a designer’s catalog of examples. You may have seen people celebrate their 1,000th link on delicious. For me this shows the value of collection of these readymades is similar to the pleasure collectors find in baseball cards, art, or other things.
Via aggregation, these collections become new readymades themselves, ready to be collected, repurposed, and reused. So collection is not just a diachronic list, or a set, but something which can be monitored and updated in a host of ways.
The second pattern of use is juxtaposition. One of my favorite parts of Facebook is the live feed [killed the day of this presentation—bwah], because status updates are mixed with relationships, game scores, and a million different things. You get the “stack of FYC papers” or “annoying question from a student” next to a question from an FYC student written on your wall. And as with collection, these juxtapositions themselves can be reembedded as readymades via aggregation.
The final use pattern is combination: intentional juxtaposition or grouping. Flickr provides an excellent example in the sets it allows users to create. Other services actively support mashups and other forms of combination, technologically and otherwise. I’m thinking of things like Creative Commons licensing here. Weblogs, arguably, are engines of combination, as they enable the quick collation of very different readymades into single posts.
I see several implications of these three patterns of use have for usability theory:
First is the notion of a single, clearly defined, easily anticipated use which Duschamp manipulated. Wide differences in intended and actual use, and radical repurposing, are still happening today. Twitter began with “What are you doing?” and retains that as their tagline. But the intended use, shared status, has been extended to an engine for microblogging, interpersonal communication, group communication, or distributed research. Jon Udell recently pointed out that Delicious could be used as a database, if you enter tags which were key-value pairs. My personal favorite redirected use was Collin Brooke and Derek Muller figuring out how to use trackbacks to achieve forward linking in the CCC Online Archive, so articles showed not only works cited but works citing.
Usability theory goes back and forth on this. Throughout his considerable body of work, from 1993’s Usability Engineering forward, Jakob Nielsen presents a very limited vision of use, focusing on tasks quite limited in scope and, in analyses, broken down into atoms easily managed by testers. Nielsen is also intensely pragmatic, setting aside nearly all notion of entertainment. In his seminal 2000 DWU, he comments, “most web projects should be to make it easy for customers to perform useful tasks.” This intensely pragmatic perspective continues in 2006 Prioritizing Web Usability. The problem here, obviously, is uses of the web which fall outside pragmatic or business use, or combine them with entertainment, communication, or other uses.
Contrast that to Joshua Porter, who confront these problems head on in Designing the Social Web. Unlike Nielsen and others, Porter notes just how difficult it is to isolate and identify the primary activity of a site or service. Though he sticks to that framework, writing, “The applications people find most compelling allow them to excel at a single activity”, he is more tolerant of envisioning use as complex. He suggests differentiating between goals, activities, and tasks. For example, Amazon’s goal is procuring basic goods; activity is shopping; tasks are comparing products, performing a product search, adding to shopping cart. Porter also recognizes that even in a work context, the best forms of uses are often enjoyable uses: the goal itself can be fun. On the one hand, he restores enjoyment and subjective satisfaction to its place in usability. On the other, the amount of time given to enjoyment is scant, and his focus remains more traditional components of usability such as efficiency, effectiveness. Because of the uses of readymades I’ve discussed, I think the bar’s been raised for complexities of use. I think we need to move even further in the direction that Porter has begun to take us.
Second, usability theory often assumes that users are motivated: that they approach their tasks not only with clear goals in mind but that they perform a task willfully. I think the exchanges of readymades on Facebook, in particular, show this isn’t always true: how many of us have installed applications, perhaps unwillingly, to avoid ignoring a friend’s gifted drink, or cause, or piece of flair? Or added friends who weren’t really our friends, or even contacts? Just the other day I saw that a friend had added the “20th century artwork” application—itself a readymade fashioned from the Gift Creator application. That resulted in this message: “So and so accepted a Marcel Duchamp, Fountain, 1917, replica 1964, Tate Collection from Doug Dunlop using 20th century artwork.” Did So and so want this application? My recollection is that she prefers other art to modernist work. I speculate that this seemingly against the grain use is smoothed over for her by the readymade framework: her familiarity with the act of collecting other Facebook objects paved the way for the fountain to become a part of her life stream.
Third is attention to the nuts and bolts of use: how does it work? Can people use this function of readymades? It’s second nature for a lot of us to say, put this in your reader, or look what happens when we mash up these data sets. We can engage readymades, in the uses like collection I’ve described, with little or no understanding. Next year’s call for papers for CCCC focuses on remixing, asking us to engage uses quite similar to the three I’ve identified above. I have to wonder, though, how many CCCC attendees are comfortable making the remixes the call asks. Let’s say we wanted to educate them how to do so: how would that work? On the one hand, it’s hard to imagine a world without cut and paste. On the other, learning how to make a mashup can require some pretty specific and complicated knowledge. It’s one thing to select and exchange of readymades on Google or Facebook. It’s quite another, I think, to do so in at a professional conference, apart from the readymade-enabling structure those applications provide.
So we need to admit and understand the reluctance of many people to use readymades. Whether resistant to consumerism, fear of misunderstanding, or for other reasons. Paradoxically, part of the reason extensible, customizable readymades are all over the place is because simple readymades don’t appeal to everyone. The desire for a coherent identity, as in visual identity, means that organizations aren’t happy with, say, using Ning to build a social network: they feel it’s necessary to find one which is as customizable as possible. The readymade for customization includes a supporting infrastructure. This makes the skin, theme, or plugin approach, all of which depend on readymades, in high demand. In turn, that reinforces the ubiquity of the logic of selection.
For me, this adds up to a shift in use, and in the readymade itself. The readymade is no longer the isolated, independent, object as found—not a fountain on a pedestal, but an object engaged deeply in complex systems of use. Now it’s up to us to adjust our apparatus for understanding use to meet the changes brought to use by the reemergence of the readymade.