«Creativity Support for Computational Literature By Daniel C. Howe A dissertation submitted in partial fulfillment of the requirements for the degree ...»
Lastly, an implementation of RiTa for one or more mobile platforms could be of significant utility as the importance of these devices for personal and/or artistic expression continues to grows. Further, the resource constraints of today’s mobile platforms are analogous to those of web browsers, both placing strict limits on memory allocation, CPU utilization, and resource file sizes. Thus much of the work accomplished in optimizing RiTa for the web could be easily repurposed for mobile contexts. The open-source Android environment represents an initially promising candidate for such an effort, as it already contains a nearly complete J2SE stack and built-in speech-recognition capabilities. Such a port would again require a new graphics implementation as described above, though the fact that Android supports the OpenGL ES standard could significantly simplify this endeavor.
While potentially a longer term project, mobile platforms represent, in our opinion, a particularly promising realm of exploration for the further development of artistic support tools.
6.2.2 Auxiliary Tools I've often speculated, bitterly, as to why there is no word processor with the kind of filters and effects that are standard features in any of hundreds of graphic or audio manipulation programs; why page layout programs don't just 'know' how to typeset poetry with either traditional or post-Mallarmé sensitivities. [Cayley 2009] In addition to augmentations of the toolkit as discussed above, there are a number of ways in which the development of related tools might be pursued. One such direction might involve integration with one or more open-source word processing package to provide access (at the interface level) to a variety of literary filters and extensions, similar to the spell-check or thesaurus options found in current packages. Such an integration has been requested in the past by less technical users, outside of the educational context, who have expressed trepidation concerning RiTa’s programmatic interface. While supporting these users is clearly important, the fact that a majority of existing creativity support research has targeted graphical user interfaces has caused us, at least thus far, to hesitate before embarking down this path.
Another augmentation, or alternative, to RiTa’s text-based programmatic interface would be a visual programming interface similar to the Max/MSP environment discussed in chapter 3. An initial prototype of this functionality has been implemented and is currently being tested (see Figure 19 below).
Figure 19: The RiTa live-coding environment (prototype).
6.2.3 Evaluation It is no longer sufficient to evaluate whether people can use a given design to achieve a task easily and efficiently. We also need—sometimes primarily—to understand how the design resonates aesthetically, emotionally, socially and culturally, both with particular users and with a larger audience. And this implies that we need new sources of assessment on the one hand, and that assessments need to be multi-layered on the other. [Gaver 2007] 188.8.131.52 Further Study Further study is clearly needed to assess the effects of different tools on the degree of creativity exhibited in project outcomes, both for students and artists. Unfortunately such evaluation has generally proven to be both difficult and resource intensive. For an in-depth discussion of the issues involved and a variety of proposed approaches, we recommend ‘Creativity Support Tool Evaluation Methods and Metrics’ [Hewitt 2005], a report produced as part of the 2005 NSF-sponsored Creativity Support Tools conference. Similarly, further evaluation of the learning that occurs in courses like ‘Programming for Digital Art and Literature’ (or Mark Guzdial’s ‘Media Computation’) would be extremely beneficial, especially were it to include a longitudinal component (following students beyond a single semester or year). One potentially useful experiment would be to teach similar material in parallel courses using a range of different toolsets (e.g., a general-purpose language like Python or Java, a RiTa-like approach, and a environment customized for education, e.g., Scratch or Alice) and compare students’ relative confidence, self-efficacy and comprehension of core concepts after each. Of course the practical difficulties of designing and implementing such an experiment are significant; any proposal along these lines would likely require significant investment from one or more institutions. For those interested, Tew et al.’s  “Impact of Alternative Introductory Courses on Programming Concept Understanding” describes some initial experiments along these lines at Georgia Tech.
184.108.40.206 Metrics for programmatic support As mentioned in previous sections, evaluation metrics for creativity support tools to this point have primarily focused on usability as manifest through graphical user interfaces [Nickerson and Landauer, 1997]. We can see this bias toward the visual and the surface even in the introductory paragraph of the NSF-sponsored Creativity Support Tools conference
report which reads (in its entirety):
Paradigm shifting breakthroughs make for great stories, but normal science is equally important in the evolutionary development of science, engineering, and medicine. Large and small breakthroughs are often made by scientists, engineers, designers, and other professionals who have access to advanced tools. The telescopes and microscopes of previous generations are giving way to advanced user interfaces on computer tools that enable exploratory search, visualization, collaboration, and composition. [Hewitt 2005, italics ours] Here we can appreciate what has been referred to “the tyranny of the visual” [Arlen 1979].
Instead of focusing on the power of computational methods to help generate new creative techniques and outcomes, the focus is on the power of the user interface, and on tools for augmenting vision and visualization. While such approaches are clearly worth pursuing, they are not the only avenue for new research. In fact, this overwhelming focus on the visual interface suggests that new researchers in the field might perhaps do well to focus their attention elsewhere; on the aural, the haptic, the linguistic, or the conceptual.
In addition to this overwhelming focus on the visual, usability has generally been evaluated in the context of productivity in traditional tasks, even when performed by creative professionals. As Hewett  comments, “researchers often focus on serving professionals such as business decision makers, biologists exploring genomic databases, designers developing novel consumer products...” In contrast, we argue, further attention to exploratory creativity is also warranted, focusing on contexts (e.g., the arts) in which creative goals are less-well-defined, and users work at a range of proximities to computational mechanisms. As Jennings notes, “new media arts practitioners and researchers should be regarded as valuable contributors not only as users needing better creativity support tools (CST) to enhance their own creative process, but also as the designers of experimental and innovative creativity support tools” [Jennings et al 2005].
Here is where the notion of a programmatic usability might prove useful, shifting our focus away from the interface and toward the authoring of creative processes. Such a focus might not only increase the utility and novelty of creativity outputs, but would facilitate a transition away from surface effects and instead toward a deeper understandings and engagement with process, such as has been argued for by proponents of procedural literacy.
As a trivial example, instead of realizing a new way to use a provided Photoshop filter, a procedurally literate user might design and implement their own filter, which not only better expresses their creative vision, but could then be shared with other filter users and designers.
As Noah Wardfrip-Fruin says in his introduction to Expressive Processing , “It is common to think of the work of authoring, the work of creating media, as the work of writing text, composing images, arranging sound, and so on. But now one must think of authoring new processes as an important element of media creation.” 220.127.116.11 Longitudinal Studies It is still an open question how to measure the extent to which a tool fosters creative thinking. While the rigor of controlled studies makes them the traditional method of scientific research, longitudinal studies with active users for weeks or months seem a valid method to gain deep insights about what is helpful (and why) to creative individuals. [Seo 2006] Lastly, creativity support tools research such as presented above could benefit greatly from the application of longitudinal studies, with students and, especially, practicing artists.
While short term evaluations have clear advantages, their efficacy is often limited as the success of creativity support software is determined over years, not weeks or months. Studies that examine the all the various phases in a tools adoption, from initial uses, to increasing familiarity, to eventual mastery, to use in teaching and/or mentoring, will be particularly useful in eliciting a more complete set of properties that either facilitate or discourage creativity in these different scenarios.
6.3 Final Thoughts In conclusion there are several points from above that we wish to reiterate. First, that perhaps for the first time, in part due to increases in the computing resources available to the average user, tool support for computational literature appears to be feasible, even when considering constrained-execution domains such as the web browser. Second, given such support, computational literature appears to be a productive new context for a wide range of users, not only for practicing computational artists and others with extensive programming knowledge, but also for students with variable backgrounds and skill levels. Third, this fact is particular encouraging as the context of computational literature also appears to quickly and naturally raises core ideas in both art and computer science, facilitating students’ creative expression while simultaneously advancing procedural literary, and exposing students to the core concepts that constitute ‘computational thinking’. Although not practical in all cases, it is also worth noting that the benefits of this approach appear to multiply when tools and accompanying pedagogy can be mutually informing, as was the case with the RiTa tools in context of the ‘Programming for Digital Arts & Literature’ course.
Additionally we wish to raise a few more speculative ideas for future researchers which, although not adequately tested, appear evident from our experience. First, that creativity support is not something easily simulated in a laboratory environment, and thus the field desperately needs more real-world applications, with careful and well-planned evaluation. Second, whenever possible, creativity support researchers should leverage synergies with other research programs stressing creativity, specifically procedural literacy, expressive programming, and the computational arts. Much of the creativity support research performed thus far has been limited by the lack of such cooperation, and by an over-emphasis on the visual elements of creative artifacts, whether the visual properties of creative outputs, or the interface elements implemented in new creativity support tools. It is important for researchers to appreciate the fact that artists have a long history of building and refining their own tools to better match their needs, and low-level control over the medium has generally been an important property of such tools. Further, such tools, even when designed and developed by a single artist for a specific project, have often proven to be of significant general utility for others working in related areas.
Lastly, we conclude by making explicit what the reader has likely sensed beneath the surface of this work; specifically that our interest in these topics is not motivated only by an impassive scientific curiosity, but instead by a deep and abiding interest in the topics at hand.
It is not by chance that this research focuses on the intersection of creative writing and exploratory computational practice, but rather because these have been central concerns in our lives, in our own artistic practice and in the work of the authors and artists that have moved us. As such, we can only hope that this writing inspires in the reader some small fraction of the aesthetic and intellectual excitement we have experienced in its creation.
RiTa example programs http://www.rednoise.org/rita/examples/ The RiTa Project Gallery http://www.rednoise.org/rita/rita_gallery.htm Programming for Digital http://www.rednoise.org/pdal/ Art & Literature course web
Gibson in his seminal book The Ecological Approach to Visual Perception. The concept was introduced to the HCI community by Donald Norman in his book The Psychology of Everyday Things from 1988. According to Norman (1988) an affordance is the design aspect of an object which suggest how the object should be used; a visual clue to its function and
use. Norman writes:
“...the term affordance refers to the perceived and actual properties of the thing, primarily those fundamental properties that determine just how the thing could possibly be used. [...] Affordances provide strong clues to the operations of things. Plates are for pushing.
Knobs are for turning. Slots are for inserting things into. Balls are for throwing or bouncing.
When affordances are taken advantage of, the user knows what to do just by looking: no picture, label, or instruction needed.” (Norman 1988, p.9) Norman thus defines an affordance as something of both actual and perceived properties. The affordance of a ball is both its round shape, physical material, bouncability, etc. (its actual properties) as well as the perceived suggestion as to how the ball should be used (its perceived properties). When actual and perceived properties are combined, an affordance emerges as a relationship that holds between the object and the individual that is acting on the object (Norman 1999). As Norman makes clear in an endnote in Norman (1988), this view is in conflict with Gibson’s idea of an affordance (explained next).