[go: up one dir, main page]

Academia.eduAcademia.edu
Assembly Lines: Web Generators as Hypertexts Elizabeth Losh University of California, Irvine HIB 188 Irvine, CA 92697 19498248130 lizlosh@uci.edu ABSTRACT This paper treats web generators as a form of online hypertext and provides a taxonomy of some of the most popular generator types. It also characterizes the ideology of participatory culture associated with the generator phenomenon – along with its subversion through parody – and how specific processes that produce user-generated digital ephemera often mimic the constraints of familiar software applications and online forms. It includes a consideration of some of the knowledge-sharing practices of PHP coders and of members of the general public who exchange information about and from these digital phenomena. Figure 1: The Church Sign Generator Categories and Subject Descriptors The basic instructions are simple: visitors to the web page choose a sign, input text, and then receive a digital photograph with their words emblazoned on a church sign. H.5.4 [Hypertext/Hypermedia]: User Issues. General Terms The author of this generator explicitly invites the visitor to engage with his online content to gratify a form of wish-fulfillment for interactivity that would be prohibited in the offline world, because social norms make such manipulation of the private property and public messages of others taboo: “Ever seen those signs in front of churches with the moveable letters? Ever wanted to rearrange the letters to make your own church sign? Well, now you can. Choose a design below, add your text, and a personalized church sign photo will be generated for you! Save it, send it to a friend, put on your website, or use it however you like. Enjoy!” [37] Human Factors, Design Keywords Web generators, hypertext theory, participatory culture, PHP hacking, Web 2.0 1. INTRODUCTION To users, popular web-based “generators” seem both to create original verbal or visual online texts and to output data that conforms to the conventions of a specific and recognizable preestablished genre. Thus, they are both generative and generic. For example, the Church Sign Generator allows users to insert their own phrases into a stock photograph of a roadside church sign (Figure 1). In the past, the webmaster of the site had photographed humorous church signs in the Austin, Texas area and had posted them to his web log. As the blog gained more readers and contributors, he began to receive doctored photos from imitators with invented messages to potential congregants and the public. So he decided to enable more of his readers to participate in activities that otherwise required some expert knowledge of proprietary software, such as the popular program Photoshop. Other conventionally forbidden activities of defacement that can be indulged in through web generators include writing on blackboards, painting graffiti, and giving women tattoos. These web generators can also allow users to personalize signifiers of authority, education, and privilege by allowing visitors to put their own names on law enforcement badges, diplomas, military dog tags, and access passes. Similar hard goods have been available for a long time from cottage industries associated with themed entertainment, novelty products, monogramming, and photographic services. Like the earlier generation of custom artifacts, digitally generated objects can create institutional liabilities when they are taken to be authentic and thus invested with the power of cultural capital and even – in certain situations – unwarranted contractual or political legitimacy. In one famous case, a graduate student in Informatics at Indiana University received a takedown notice from the Department of Homeland Security for creating a web generator that enabled users of the site to print out what appeared to be credible looking boarding passes from Northwest Airlines; he was also asked not to post official correspondence from the agency on his web log. [24, 43] Despite the political dangers of replicating Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. HT’07, September 10–12, 2007, Manchester, United Kingdom. Copyright 2007 ACM 978-1-59593-820-6/07/0009…$5.00. 115 this student’s code, mirror websites and other forms of technological assistance attempted to subvert the government’s attempts to control the generator’s dissemination and to invite further critique of the security vulnerabilities of procedures for the Transportation Safety Authority that such a simple form of coding exposed. 2. TEXTUAL COMBINATIONS 2.1 Poetic Machines Very early in the history of literary and linguistic computing, programmers turned their attention to the synthesis of convincing syntax in natural language. By 1971, computer scientists were trumpeting the successes of poems like “The Meditation of IBM 1094-7040 DCS,” which were culled from fragments gathered from random searches of poetry anthologies. [7] Even college students in a 1978 study could not tell which poems were written by a computer and which were written by a human being. [46] Electronic hypertext has long been associated with combinatorial procedures by artistic communities associated with Dadism, Oulipo, Fluxus, L=A=N=G=U=A=G=E poetry, and other aleatory forms [49]. Furthermore, these twentieth century movements were preceded by poetic practices about chance operations and textual recombination from many different national literatures and creative eras in the history of poetics. In contrast to these relatively self-evident processes, in which input matches output at the level of the legible text on the digital object, there are types of web generators in which the algorithm of compiling the code from natural language is obscured or in which user input has no effect on ouput. In the case of such “black box” operations, it is difficult for the user to intuit how the rules that govern the system generate a given result. [5] For example, the IKEA (R) Name Generator outputs a sequence of Scandinavian looking syllables when a virtual button is pressed, but it is not immediately apparent what determines the phoneme order that ultimately appears on the screen. [18] Other generators, such as the Pirate Name and Mafia Name generators explicitly admit to be driven by random number functions and thus their webmasters refuse to assume liability for the name generated [37]. Thus, some generators may incorporate elements of randomization, while others move mechanically through a predetermined sequence of pseudo-outputs, while others are largely user directed. In this case, the degree and manner of user interactivity can vary considerably among these web generators: the user may type in text, pull choices from a menu, tick multiple choice bubbles, move virtual objects, upload JPEG images from a personal computer, take a screen shot, push a button, or merely refresh a web page. Commercially produced magnetic poetry for use on refrigerators and magnetic boards popularized the intention and agency associated these practices. In contrast, web generators often emphasize chance operations in the activity of poetic composition and the offloading of labor to elsewhere. For example, the Dada Poetry Generator suggests cutting and pasting text from “an online article or newspaper for best results.” [18] In contrast to this authorless text, other poetry generators commemorate the styles of specific poets, such as Emily Dickinson [42] or William Carlos Williams [51]. 2.2 Deconstruction Engines Web generators, like hypertext literature, also have been linked to critical theory associated with poststructuralism and other European schools of thought that question the supposed order and rationality upon which Western philosophy is based by pointing out the logical circularity of these cultural assumptions, which take their own premises as conclusions. Furthermore, many of these postmodern philosophers, following Saussure, often draw attention to the arbitrary nature of signifiers in linguistic systems. As George Landow and others have argued, hypertext and postmodern theory are intimately linked, in that both seek new ways to make meaning out of seemingly de-centered texts that are no longer moored to the set order of the printed page in space or of verbal speech as an event with specific unidirectional temporality. [26] Yet all of these online sites are somehow recognized as belonging to a common web genre that is tagged “generator,” to the extent that there is common metalanguage to describe all these webbased services as belonging to the same category. There is even a web log exclusively devoted to announcing, displaying, and organizing these websites, which receives thousands of unique visitors a month [48] Of course, web generators are usually placed in a very different category from more elite forms of web art or electronic literature that encourage an inherently much more selective audience to interact with texts or images that are clearly framed in the context of aesthetic value, artistic expression, poetic ineffability, and often resistance to commodity culture. However, what Bill Seaman has characterized as “recombinant poetics” functions at the level of popular discourse as well [40], as does N. Katherine Hayles’ “technotext.” [15] Furthermore, electronic literature often draws attention to the procedural quality of how readers interact with electronic texts, even it may do so without reference to the input/output functions that generators make manifest. For example, Jim Andrews’ “Stir Fry Texts” and Judd Morrissey’s “The Jew’s Daughter” generate combinatorial texts that change as the viewer mouses over sections of words. Although web generators may be more likely to foreground specific sets of instructions to the user, both web generators and web-based electronic hypertext draw attention to the ways that users interact with language rather than merely compose or consume it. Despite its challenges, Landow argues that hypertext, like postmodern theory, is legible and rhetorically effective. Landow sees a form of cultural convergence taking place in which software development and poststructuralist theory are producing analogous if not homologous texts. Landow frequently cites the work of Derrida, Barthes, and Foucault as explanatory lenses through which to understand how hypertext operates. He also asserts that the poststructuralist reader is still oriented through a “rhetoric of arrival and departures.” [26] For example, in Hypertext 3.0, Landow says that linking “permits simple means of orienting readers by allowing a basic rhetoric of departure” whereby these readers will be directed to “a clearly defined point in the text” that operates through a rhetoric of arrival. [26] Web generators similarly signal the teleportation of the reader from one place to another; they also subvert or bring into question norms 116 three MIT students used SCIgen, a program that generates random computer science papers in ACM format – complete with graphs, figures, and citations – to fool the organizers of WMSCI 2005, a supposedly peer-reviewed conference in Orlando, Florida, known for its spam-style solicitations of those who work on technologyrelated issues. [2] Titled “Harnessing Byzantine Fault Tolerance Using Classical Theory,” “Synthesizing Checksums and Lamda Calculus Using Jog,” and “On the Study of the Ethernet,” these papers demonstrated that computer science also had its buzzwords. Such automatically generated papers have also been successful at deceiving mathematics, emerging technology, and new media conferences. about the conventional rules of linear written discourse and the fixity and authority of print culture. However, web generators have also been used to satirize the theoretical and scientific pretensions of postmodernism, by suggesting that postmodern criticism could be written just as coherently and authoritatively if manufactured by chance operations by a machine. Those who applied the Dada Engine to the project of creating a Postmodernism Generator that manufactures convoluted writing with polysyllabic diction and generously footnotes at the push of a refresh browser button claimed that it was also an homage to the hoax of physicist Alan Sokal. [9] Sokal, who has proclaimed himself to be an advocate of “reason, evidence, and logic” achieved fame outside academia by submitting an article entitled “Transgressing the Boundaries: Toward a Transformative Hermeneutics of Quantum Gravity” to the journal Social Text. [45] Editors decided to publish what Sokal later revealed to be a spoof that parodied the relativism of contemporary critical theory. [45] Andrew C. Bulhak, creator 2.3 Speech Acts In contrast to Bulhak’s critique of postmodernism, which focuses on the insignificance of web-generated expressions, as it presents a critique of the wasteful rhetorical expenditures of expert discourses in the humanities, many web generators emphasize the performative functions of language. In other words, generators for speech acts assume the cultural force of linguistic expression rather than its ineffectiveness or disengagement. Ironically, this power of language can also be seen as linked to making explicit the arbitrary nature of linguistic signification and the fact that these utterances are not part of everyday speech. Many generators emphasize how language can wield forces capable of reordering the social dynamics that might remain constant during more quotidian exchanges or even of transforming the circumstances or human actors within their cultural and physical environments. For example, the Mobster Threat Generator assumes that particular utterances can have life-altering consequences. of the Postmodernism Generator, takes Sokal’s joke further by removing human intentionality and decision-making from the process of composition entirely. He makes his criticism of postmodernism even more explicit in a scientific paper on his generator, in which he compares the system’s output prose of pseudo-postmodernism to a simulation of “mental debility,” as might be observed in the “ranting of a paranoid schizophrenic street preacher, or perhaps a USENET ranter.” Bulhak also suggests that the script could also be modified to mimic the writing in “eccentric pseudoscientic/religious pamphlets.” [8] Situating computergenerated discourse in the realm of mental illness – or of mental health – is not new. As the programmer of the Postmodernism Generator, Bulhak acknowledges his debt to “Weizenbaum’s ELIZA (which simulates a psychiatrist) and Ken Colby’s PARRY (which simulates a paranoid mental patient).” [8] In particular, curses and blessings occupy another significant subgenre of web generator sites. Many of the cursing sites avoid crass obscenity or colloquialism in their generated texts, and some assume mock religious, historical, or literary linguistic pretensions. There are curse generators that draw content from religious scriptures, and several that use Shakespeare’s text. For example, the Biblical Curse Generator appeals to website visitors with the following pitch: “Lost for a smart remark to see off your enemies? Unable to deliver that killer insult? Put an end to ‘I was speechless!’ misery with the amazing Biblical Curse Generator, which is pre-loaded with blistering put-downs as delivered by Elijah, Jeremiah and other monumentally angry saints.” [40] Although this is a simple push button generator, visitors to the site are given opportunities for a more participatory experience on another page that offers a monthly caption competition for images in which clergy or lay worshippers may appear ridiculous in their photographs. One interesting feature of Bulhak’s paper about the Postmodernism Generator is that, as he reflects upon the experience, is his reluctant to take his assertions about postmodern culture into the realm of politics as Sokal does. Sokal asserts his identity as a “leftist” and “feminist,” while also descrying the loss of political ground to the right that academic relativism brings. [45] Bulhak’s argument, in contrast, never gets far outside the college campus as its rhetorical context; he also notes that “this script was originally written with the intention of generating bogus practice examination papers to be distributed in lectures for the purpose of scaring students.” [8] Instead, Bulhak often keeps his assertions in the realm of disciplinary feuds between what C.P. Snow once called “the two cultures,” science and the humanities [43]. For example, he claims to have chosen critical theory because it is “easy to convincingly generate meaningless and yet realistic travesties of works in it . . . because of the combination of the complex, opaque jargon used in these sorts of works and the subjectivity of the discipline; similar automated travesties of papers in, say, mathematics or physics, would be less successful, because of the scientific rigor of these fields.” [8] Also noteworthy for this form of analysis is the Elizabethan Curse Generator, whose author indicates the influence of meme theory on his web work, in which “cultural software” serves as a category for analysis from a technological viewpoint toward reproduction and replication [1]. Although the Elizabethan Curse Generator is a relatively simple program, which consists of little more than a push-button interface, the associated web materials indicate a recombinant interdisciplinary collection of elements that also speaks to the heterogenousness of some of the textual practices associated with generator sharing and making communities. [46] There is a certain hubris in his assertions, given that scientific disciplines have also been fooled by the authentic appearance of computer-generated technical papers. For example, a group of 117 Anthropologists have long noted the generative power of curses and blessings, [23] so it is interesting that they have become such a prevalent generator type. Curses are not the only form of magical speech; there is also a blessings generators. For example the Worldwide Blessings Generator combines generated texts with images from the world’s religious traditions and objects of meditation. [6] Visitors to the site can also suggest that more blessings be added to the database by using an online form. On this form, the author of the Blessings Generator explains the syntactic rules of his blessings generator, in which utterances always begin with the word “may,” which is a structure also used in some curse generators. killers, gangsters, and ogres. Nonetheless, name generators also exist for popes, superheroes, Mormons, and others associated with exemplary lives and social purity. Naming can also have implications for privacy as well, which is always a key concern for those who take part in virtual communities or online commerce. For example, the security and technology expert Bruce Schneier has examined one random name generator as a way to actually generate false identities for purposes of crime, political resistance, or heightened privacy. [37] 3. TEXT IN VISUAL CONTEXT A major family of web generators combines image and text in ways that date back at least to the tradition of the emblem in Renaissance discourse, when interchangeable stock images or stock phrases were used by authors, illustrators, and printers as part of the popular culture of the literate public. [48] Cartoon captions, beer labels, collectable cards, and many other forms of participatory culture associated with fan communities [20] are also celebrated in such generators. Traditional of holidays are marked – such as Christmas, St. Patrick’s Day, Easter, and Halloween – with generated ephemera much like the personalized greeting cards that once could only be circulated among the affluent. (Such greeting cards from the first half of the twentieth century sometimes had an interactive component as well, which required the recipient to apply water or a lit cigarette in order to see the whole image or text message.) Insults and compliments are produced by a related and yet separate class of web generators. The Open Directory catalogues twenty-two separate insult generators, which actually represents a mere fraction of the genre. [33] In the case of insults, some involve web forms to further “personalize” the barb. On the opposite side of the semiotic spectrum, a multiple compliment generator combines proscribed parts of speech. Moreover, the Surrealistic compliment generator combines web generator genres by paying tribute to the early twentieth century heritage of other new media artworks [19] There are also other generators for positive speech acts with real-world effects, such as generators that create love poems or pick-up lines to persuade potential romantic or sexual partners to move toward greater social intimacy with the speaker or writer. 3.1 Design Applications Finally, no taxonomy that includes taboo forms of language in web generators is complete without pointing out that there are also a number of sites that transform what may be civil language on regular websites into scatalogical or sexually explicit texts, such as the Pornolyzer, which also serves as an online translating program for multilingual obscenity. [22] On the other extreme, there are also generators for euphemisms, which render the initial word more inoffensive. A Family Values Generator comes complete with a “censor” button that inserts asterisks in the generated profanity. Lev Manovich has observed that the power of automation is one of the chief features of the computational and aesthetic logic of the commercial design program Photoshop [29]. And yet – for many amateur users – Photoshop is not nearly automatic enough, particularly when complicated layering, filtering, or numerical functions are applied to the image. In contrast, web generators allow website visitors to modify images or add text easily to pictures in ways that would otherwise involve more knowledge of the Photoshop tool menu. What Manovich calls “the logic of selection” is obviously also tightly constrained, because the cut and paste features of web generators are limited. Furthermore, process-intensive operations are impracticable when even hot linking is discouraged. Nonetheless, because these generators are more clearly associated with design practices of recreation and leisure than those of work, users may consider them as part of the digital practices of informal sociality and ludic interaction. All these speech act categories are exemplary of what experts in computers and other technologies call “automagical” thinking, [10] an adjective that combines the wonder associated with mechanistic processes with that used for supernatural effects. Virtuosity in the stage-managing of interface and programming design can make the output appear to be particularly responsive to the user’s unarticulated needs. In this way, complex technical processes are also hidden from users or operators and thus are experienced as phenomena without rational explanations. Although the most popular web log that categorizes these generators describes them as “software that makes software,” [48] it may be more accurate to understand web generators in the context of coding practices associated writing functions that are generally for middleware and particularly for PHP. Of course, web generators have been written using a variety of computer languages, scripts, and coding practices, and include hacker communities that share knowledge around Perl, JavaScript, and even Pike to build these interactivity experiences for users. Thus web generators are at the center of a number of online gift economies in which digital media and the information for encoding are exchanged among separate communities of users. In other words, not only do visitors to these websites share the texts and images that are generated among many others who are similarly in non-technical constituencies, but other users – who are more expert about “unit operations” in a variety of senses as 2.4 Protocols of Naming There are other kinds of generators that can be understood through anthropological frameworks or speech-act theories as well. For example, many web generators are designed to create possible names, which may be for the user, for others in his or her face-to-face social circle or online cohort, or – more often – for entertainment purposes exclusively. Given the fact that screen names constitute a primary identifier for members of online communities, creating or choosing names is often a significant practice in digital culture. It is interesting to note that the names produced by web generators frequently represent the identities of transgressive personalities, such as mafia hit men, pirates, serial 118 Publishing source code is not always a politically neutral act, particularly for generator programs that involve the random or extremely large numbers that are often used for encrypting data. For example, on one generator-maker’s home page one of the dictums that is generated at the bottom of the user’s screen is “If You Can Put It on a T-Shirt, It’s Speech.” [46] It might initially appear that the T-shirt reference is simply a metaphor included in the generated choices for its appropriateness to the subject matter, because it suggests images of customization and manufacturing that are consonant with the generator experience, but the history of this quotation in intellectual property battles is actually much more complicated. The actual sentence was uttered by Carnegie Mellon Professor David Touretzk in his testimony for the defense in the Universal Studios v. Shawn C. Reimerdes case. The t-shirt in question was printed with the computer source code for descrambling commercially produced movies in DVD format. Ian Bogost describes them [5] – congregate around web generators to swap code and trade ideas for customizing interfaces. 4. REMIX CULTURE In another major family of websites, there are generators that use no written text or very little and thus rely on enticing the user to combine images, video, or sound clips as a part of an interactive sensory feedback experience. For example, playful inquiry into modern art appears to be operative in the Jackson Pollock Generator and the Modern Art Generator. At its best, this can epitomize what Lawrence Lessig has described this as a “remix culture” [27] in which recombination itself can be recognized as a creative act. However, remix culture certainly contains potential controversies and social tensions, as Lessig himself has shown. For example, there are an extremely large number of avatar generators, which hearken back to paper dolls and other interactive two-dimensional forms of child’s play with paper goods. Michelle White has observed, however, that avatar creation often involves recombinant activities that create anxieties about power that can include considerations about the ownership of intellectual property or the orientation of a viewing gaze that potentially compromises online privacy. [50] As Bruce Scheier has pointed out, if randomly generated numbers can be owned and maintained as the sole property of particular legal actors, what prevents anyone from stumbling into infringement through a similarly random process? Certainly in their blogs, many generator creators also indicated sympathy with the recent case of the the Digg rebels who later came to be known as “09-ers” who published another video encryption key and also created popular specially marketed goods and apparel that were emblazoned with the secret number, sometimes ironically in cleverly encrypted forms, to publicize their open source cause. 5. THE RULES OF THE GAME 5.1 Best Practices 5.3 Copyright and the Creative Commons Although web generators are in a class of digital objects for which scholars do not generally have the same anxieties about cultural preservation and documentation that hypertext literature inspires in groups like the Electronic Literature Organization, visitors to generator websites are still irritated by incompatibility with particular browsers or irrupted lines of code. How generator designers gather and respond to user feedback is generally not transparent to critics of digital culture. Nonetheless, some generator designers provide copious information about fixes and bugs to forestall user frustration. Web generators often carry copyright announcements, and yet at the same time many pages with web generators also make arguments for the fair use of either their content or the content of others. For example, the Road Sign Generator bears a copyright symbol, but it also claims to be released under a Creative Commons Attribution-NonCommercial-ShareAlike 2.5 License. [21]. The Postmodernism Generator uses a similar code. [9] Although some web masters do express anxieties about unintentionally participating in infringing activities. For example, the Church Sign Generator explicitly asks its visitors not to approach the author about inserting copyrighted material, although monitoring violations would be possible given the large collection of registered trademarks and copyrights that have been claimed. [37] The Elizabethan Curse Generator gives credit to the high school English teacher who compiled the list of Shakespearean expressions used in the program, but the programmer also points out that this list should not be considered the teacher’s intellectual property since he is re-purposing the works of Shakespeare, even though selection may be more arguably capable of creating proprietary new works than randomization. 5.2 Public Source Code Web generators that represent advertising or marketing interests, such as the IKEA(R) Name Generator [18] tend to be understandably reluctant to reveal trade secrets. However, many of the other generators have made the source code public in the interest of promoting collective intelligence [28] and facilitating bottom-up design processes that follow the bazaar model that is described by Eric Raymond [35]. For example, the creator of the Church sign generator has released the PHP source code, [36] as has the creator of the Card Catalogue Generator, [4] albeit less prominently. Release of the source code is not always a signal of continuing participation in a collaborative community, however. Sometimes when the programmer releases the source code, he or she will do so with the specific caveat that further help with implementation will not be provided. For example, on a site that provides a generator that converts digital images into ASCII characters the author writes, “no help will be given for people who want to set it up on their own websites. There is no way to run it without a webserver application, so don't ask to be sent the program.” [36] 6. THE VIRTUAL STATE Some web generators can also be understood in the context of what Michel Foucault has called “governmentality” in that they make manifest features of a particular mentality of rule or technologizing of the political subject. [13] Pull-down menus and online forms are a key characteristic of the electronic bureaucracy of e-commerce and e-government, which Jane Fountain has traced back to the Clinton administration and government initiatives to provide broader access to the public to the Weberian system of 119 creating and maintaining files, if only at the level of data entry [14]. These online forms on government websites could be said to have set user expectations for interactivity for much of Web 1.0. For example, the Library Card Generator [3] and Barcode Generator [4] may be representative of the recurrent theme of bureaucracy and organizational systematization that characterizes many web generators. Of course, in the case of the Barcode Generator making UPC labels, users generally lack the actual piece of equipment that could read the code and verify that the label that appears as output is an accurate translation of the input content. Like translation generators, or those that produce other kinds of code, certain forms of knowledge, the need for which the generator seemingly subverts, are still required to authenticate the system’s accuracy. 7. CYBERCULTURE AND SATIRE Digital culture itself is often a topic for web generators, in that they draw attention to the lowest common denominator of interactivity or user-generated content. Although Pierre Lévy [28] and Henry Jenkins [18] have celebrated the potential of a synergistic participatory culture that capitalizes on social media and transmedia story-telling platforms for user-generated texts, there are many ways that convergence can be stultifying. Most obviously, a customizable template is still a template, so that user experience in social media venues is still constrained by technological and organizational restrictions that potentially also limit emergent behavior. A related subgenre lampoons the corporate logic of the new digital economy. For example, the Dot Com Prediction Generator or the Apple Rumor Generator demonstrates public skepticism about hype surrounding the fates of technology companies. Given the existence of the MSN Search Spoof Generator, it appears that at least one technology corporation has produced a satiric generator of its own. 6.1 Pull-Down Menus and Online Forms The Evil Guide Plan uses many of the conventions of the pulldown menu, which are already familiar to users who shop and register for services online and thus give personal information about themselves in order to complete basic transactions. The opening text of the guide reads, “Your evil plan is nearly complete. Simply fill in your answers in the appropriate blanks below and then get ready to call your press conference. You may want to photocopy this page first, in case you change your mind later and want to create a different evil plan.” [11] In this generator, pre-set options exist for critical categories like “motive” and “objective,” and the narrative can be filled out in stages with “supplemental information” about the evil person’s “base of operations” or “tragic past.” 7.1 Web 2.0 Generators Despite the fact that “Web 2.0” is a relatively recent phrase to describe the cultural shift in digital practices toward two-way communications, personal publishing for niche audiences, and reciprocal file-sharing and remix practices, several different Web 2.0 generators have already appeared that ridicule hyperbole associated with those who are capitalizing on this technological trend. In addition to mocking vapid catch phrases, some Web 2.0 generators also parody elements of the design aesthetic that has come to be associated with Web 2.0. Writer of interactive fiction and hypertext critic Mark Marino actually created a kind of metagenerator that combines elements of other Web 2.0 generators, which is called the Web 2.0 app GeNerAtor. At the push of a button, Marino’s page generates a silly name, color scheme, list of features, and a full mash-up of possible kinds of functionalities, files, and people to share with. There are also more open-ended online forms in web generators, some of which have been appropriated by new media artists who are more closely associated with the academy and institutions of educated taste. Alternate Reality Game designer Jane McGonigal created Place Storming to encourage academic researchers to popularize their work and connect research activities to more engaged forms of discourse. To develop the metaphor of creating your own superhero identity, McGonigal first has visitors to her site fill out an online form that generates a profile. [33] Academics are asked to name their “roving band of superheroes” and identity their “superpowers” and “mission.” For example, the Web 2.0 Buzzphrase generator uses the familiar scaled tags motif to organize hyperbolic fragments, such as the following sample: “Cry out, blogosphere! We shall transcend borders. This will change everything. 2.0 is the new New. The buzz is loud and clear. The words aren't what they were. This is newer media. Float this. An AJAX-driven GUI. Single. Word. Sentences! Faster. Faster! Hack it.” [34]. Unlike the Postmodernism Generator, which was dominated by elaborate sentences with polysyllabic words and subordinating clauses, the Web 2.0 Buzzphrase generator truncates language and scatters it on the page in isolated memes. At the bottom of the main page, the creator of this generator credits a real Web 2.0 company, Flock, as “inspiration.” 6.2 Privacy Issues about surveillance and policing conduct on the Internet also play a role in the policies adopted by the creator of web generator pages. Many generators show the messages from a given number of the most recent users of the site, which creates an environment in which virtual graffiti may defame particular individuals or violate the privacy of others who are creating messages not intended for public consumption. For example, although the creator of the Church Sign Generator says that there are no serious obstacles to posting the last ten signs generated, issues about community standards make posting this user-generated content problematic. [37] Indeed, as Marino points out, it can be difficult to create a generator for humor value that is as ridiculous as some real-life web applications can be. As Marino writes of the non-fictional company Ning, “With Ning, Web 2.0 has reached the height, nadir, and infinite loop of its own generationality by offering a Web 2.0 sites that generates other Web 2.0 sites (as perhaps all Web 2.0 sites do).” [30] Ironically, there is even a Privacy Policy Generator from the Direct Marketing Association, in which operators of websites can generate standard disclaimers about how information may be used by the company and by third parties. [12] 120 [17] Hess, G. Dada Poetry Generator. Poem of Quotes. http://www.poemofquotes.com/tools/dada.php, 2004-2007. If the naive understanding of a web generator is that it is “software that creates software,” the recursiveness of Web 2.0 cultural products may make it particularly ripe for representations in web generators. As this paper has argued, there are also forms of “cultural software” [1] involved in PHP hacker and user communities that merit serious attention, in that they suggest interpretive frameworks with which to understand the sensory, social, political, and economic features of hypertext and interactive media more generally. [18] IKEA(R) name generator, http://www.minimarketing.it/ikea_gen.htm [19] I.N.X.J.U./Banjo Ruthless Creations, The Surrealist Compliment Generator, http://www.madsci.org/cgibin/cgiwrap/~lynn/jardin/SCG/, 1994-1996. [20] Jenkins, H. Convergence culture : where old and new media collide. New York University Press, New York, 2006 8. ACKNOWLEDGMENTS [21] Josefsson, S. Atom Smasher's Road Construction Sign Generator, http://atom.smasher.org/construction/ Thanks to Mark Marino for his insights about the genre of the generator and to Gerard Vlemmings for his initial response to my question about definition. [22] Kim. The Pornolyzer, http://www.pornolize.com/. [23] Kratz, C.A. Genres of Power: A Comparative Analysis of Okiek Blessings, Curses and Oaths. Man, 24 (4). 636-656. 9. REFERENCES [24] Krebs, B. Boarding Pass Hacker Breaks Silence. The Washington Post, http://blog.washingtonpost.com/securityfix/2006/11/boarding _pass_hacker_breaks_si.html, 2006. [1] Balkin, J.M. Cultural software : a theory of ideology. Yale. University Press, New Haven Conn., 1998 [2] Ball, P. Computer Conference Welcomes Gobbledygook Paper. Natutre, 434. 936, April 2005. [25] Krieger, M. Ekphrasis: the illusion of the natural sign. Johns Hopkins University Press, 1992. [3] Barcode Label Printer: Free Generator & Sales of Barcode Printers, http://www.barcodesinc.com/generator/index.php. [26] Landow, G.P. Hypertext 3.0: critical theory and new media in an era of globalization. Johns Hopkins University Press, Baltimore, 2006. [4] Blyberg, J. blyberg.net » Catalog Card Generator, http://www.blyberg.net/card-generator/. [5] Bogost, I. Unit operations : an approach to videogame criticism. MIT Press, Cambridge, Mass., 2006. [27] Lessig, L. Free culture : how big media uses technology and the law to lock down culture and control creativity. Penguin Press, New York, 2004. [6] Bombzilla.com. Worldwide Blessing Generator, http://worldwideblessing.com/. [28] Lévy, P. Cyberculture. University of Minnesota Press, Minneapolis, Minn., 2001. [7] Borroff, M. The Computer As Poet. Yale Alumni Magazine, 1971. [29] Manovich, L. The language of new media. MIT Press, Cambridge, Mass., 2001. [8] Bulhak, A. On the simulation of postmodernism and mental debility using recursive transition networks. 96/264, Dept Computer Science Technical Reports, Dept Computer Science, Monash Univ, Melbourne Australia, 1996. [30] Marino, M. Generating Web 2.0 WRT: Writer Response Theory, http://writerresponsetheory.org/wordpress/2007/04/21/genera ting-web-20/ , 2007. [9] Bulhak, A. and Larios, J. The Postmodernism Generator. Communications from Elsewhere, http://www.elsewhere.org/pomo, 2000. [31] Marino, M. web20_generator, http://www.bunkmag.com/bunklets/web20/. [10] Cool, J. The Datastructures of Everyday Life (The Limits of Automagical Order). The Participant Observer, http://cool.org/?p=20, February 24, 2007. [32] McGonigal, J. and Kelly, R. Place Storming. Vectors Journal of Culture and Technology in a Dynamic Vernacular. [11] Darksites.com. Evil Guide Plan, http://www.darksites.com/evilplan.php. [33] Open Directory. Recreation – Humor: Insults. http://dmoz.org/Recreation/Humor/Insults/ [12] Direct Marketing Association. Privacy Policy Generator http://www.the-dma.org/privacy/creating.shtml, 2002. [34] Ordoveza, P. Web 2.0 buzzphrase generator, http://what.was.the.question.whyblog.org/buzz/, 2007 [13] Foucault, M. Governmentality. The Foucault Effect, Studies in Governmentality, University of Chicago Press, 1991. [35] Raymond, E. The Cathedral and the Bazaar: musings on Linux and Open Source by an accidental revolutionary. O’Reilly, 2001. [14] Fountain, J. Building the Virtual State: Information Technology and Institutional Change. Brookings Institute Press, 2001. [36] Roos, P. TEXT-IMAGE.com, http://www.text-image.com/, 2005. [15] Hayles, N.K. Writing machines. MIT Press, Cambridge, Mass., 2002. [37] Sanders, R. Church Sign Generator, http://www.churchsigngenerator.com/, 2007. [16] Heller, D. Magnetic Poetry. English Journal 87 (4). 89, 1998. [38] Schneier, B. Random Identity Generator. Schneier on Security. 121 [45] Sokal, A. A Physicist Experiments with Cultural Studies. Lingua Franca, May/June 1996. http://www.schneier.com/blog/archives/2006/06/random_ide ntity.html, 2006. [39] Schock, J. The Mafia Name Generator, http://gangstaname.com/mafia_name.php. [46] Stone, T. Elizabethan Curse Generator. http://trevorstone.org/curse/ [40] Seaman, B. Interactive Text and Recombinant Poetics. First Person: New Media as Story, Performance, and Game, MIT Press, 2004. [47] Strenski, E. and Esposito, N.G. The Poet, the Computer, and the Classroom. College English, 42 (2). 142-15, 1978. [48] Vlemmings, G. The Generator Blog, http://generatorblog.blogspot.com/, 2005-2007. [41] Ship of Fools.. Biblical Curse Generator, http://ship-offools.com/Features/Curses, 2003. [49] Wardrip-Fruin, N and Montfort, N. The New Media Reader. MIT Press, 2006. [42] Smith, M. logopoeia » emily dickinson random epigram machine, http://logopoeia.com/ed/, 2005. [50] White, M. The Body and the Screen. MIT Press, 2006. [43] Snow, C.P. The Two Cultures, And A Second Look. New American Library, 1964. [51] Write an Instant William Carlos Williams Poem, http://ettcweb.lr.k12.nj.us/forms/williams.htm. [44] Soghoian, C. Takedown Orders. slight paranoia, http://paranoia.dubfire.net/2006/11/takedown-orders.html, 2006. . 122