6.) Code. Use the computer. It's not a television.

Excerpted from Lewis Lacook's posting of his "rules" for net.art to the Webartery mailing list, reposted to the Nettime list (February 14, 2002).^1

Codework refers to the use of the contemporary idiolect of the computer and computing processes in digital media experimental writing, or [net.writing]. Some of the prominent practitioners include Alan Sondheim, who has given the practice and genre its name, Mez (Mary-Anne Breeze), Talan Memmott, Ted Warnell, Brian Lennon, and John Cayley. These writers also use different terms to refer to work: Mez composes in a neologistic "net.wurked" language that she has termed m[ez]ang.elle; Memmott uses the term "rich.lit"; Warnell names some of his JavaScript poems "codepoetry"; Lennon refers to "digital visual poetics"; and Cayley produces algorithmic, generative texts, or "programmable poetry." Writers and artists who have taken up the general practice of codework heed the mandate - "use the computer; it is not a television" - and strive to foreground and theorize the relations between interface and machine and so reflect on the networked environment that constitutes and is constituted by a digital text. The precise techniques vary, but the general result is a text-object or a text-event that emphasizes its own programming, mechanism, and materiality.

Picture e.e. cummings, bp Nichol, or Emmett Williams upgrading their medium and exchanging their typewriter keys for the units of programming languages, and the result would in part resemble the contemporary mode of experimental writing and net.art called "codework." So, Mez, for example, expressly strives

2 uze computer kode kon.[e]vent.ionz spliced with irc emoticons and
   ab[scess]breviations....
2 spout punctu[rez]ationz reappropri.[s]ated in2 sentence schematics....
2 illustrate the x.pansion of software potentialities of co:d][iscours][e
   in an environment x.clusively reliant on it.

Mez, The Art of M[ez]ang.elle.ing. Also, see her introductory description to the data][h!][bleeding texts, which was short-listed for an ELO writing award (2001). ^2

In practical terms, the difference between Emmett Williams, "Meditation No. 1" or e.e. cummings's, "r-p-o-p-h-e-s-s-a-g-r" and a net.wurked text by Mez, Warnell, or Memmott is the difference between the typewriter and the computer, the difference of what the medium allows.

For another manifesto, see Blood Puppets and all. Though not individually noted as such, all excerpts from Mez's codework quoted in this essay are to be read as "[sic]." ^3

Broadly, codework makes exterior the interior workings of the computer. One formal purpose is to bring the function and code of the computer to a kind of visibility. That is, to illuminate the many layers of code - the tower of programming languages that underlies the representation of natural languages on the screen. For all of the differences among particular instances or events of codework, they all incorporate elements of code, whether executable or not. Code appears in the text, then, in whole or in part, in the form of a functioning script, an operator, and/or a static symbol.

As John Cayley's essay in this ebr release will indicate, there is a reiterative component to my initial description of codework. Cayley, Sondheim, Memmott, and others have outlined, discussed, and queried this branch of new media experimental writing, in forms ranging from the short gallery review to the listserv posting and conference presentation. We are, however, only now approaching a second wave of critical discourse on the subject, and some of the descriptive foundation still needs to be articulated as we move on to consider some of the more important questions and issues raised by the practice: the relations between natural and programming languages; the link between contemporary codework poetics and earlier, "avant-garde," found-object artistic practices; the schism between formal aestheticism and socio-cultural politics; and the question of cognitive transformation.

Codework participates in a larger movement that we might call the "art of code," in which the code used to produce the work seems to infiltrate the surface, the former domain only of natural languages and numeric elements. For example, on her recently released album, Head Slash Bauch, Antye Greie-Fuchs (AGF) reads lines of code deconstructed into syllables. [Click here for a sample] She intermingles English, German, elements of markup languages, and the language of code, such that "layer readme slash p ID blockquote slash layer" resonates aurally, symbolically, and technically. The art of code is not limited to any one particular media environment, and its use as a medium reflects the expanding symbolic database that is at once artistic and communicative. So, too, does it reflect the changes in natural and machine languages and the changes in our evaluation of both. That these encounters with code, however fleeting, partial, or incomplete, should necessarily result not only in revised cultural forms and practices, but also in anxieties about intrusion, contamination, and uncontrollability, is evident in Jessica Loseby's illustrative net.art work Code Scares Me. Loseby's installation uses Flash actionscripting to "domesticate" the monstrosity of code and directly thematize the fear of invisible and unknowable code, disturbing because she considers it to be "a language that is both hidden and alien to me."


For Loseby, code is initially understandable only in terms of impenetrable darkness. It lurks beneath the surface of the text, but it is not in direct dialogue with that text: it is read and yet not read at the same time. The fear, further, is that code is autopoietic and capable of eluding the artist's attempts to domesticate it and bring it into order: "I imagine it unlocking itself in my absence," she notes, conjuring a vision of code compiling itself, generating its own output, and moving toward self-organization. In this instance, code is `scary' because it is both unknown (`foreign') and known (understood to have emergent properties).

An art of code, though, would almost necessarily suggest that code can be beautiful instead of alienating. For instance, Geoff Cox, Alex McLean, and Adrian Ward argue at length for "The Aesthetics of Generative Code" in a paper that suggests that the beauty of code lies in its performance, functionality, and execution.

I have elsewhere explored the question of the theoretical differences between analog and digital text; see "Reveal Codes: Hypertext and Performance," Postmodern Culture 12:1 (September 2001). ^4 This kind of affect for the elegance of code is shared by Ellen Ullman, though she holds out as well that elegance is linked to operation. So, too, does an aesthetic appreciation of code for these programmers require a knowledge of its operation, such that form and function (execution) are thought together. Ted Warnell's visual, computational poem, "Lascaux.Symbol.ic" serves as a particularly apposite illustration of this fusion, displaying as it does a JavaScript with both operational and visual style. Similarly, in "If () then ()", Jutta Steidl claims that code has and is capable of expressing an aesthetic, but she questions whether programming languages can express the inexpressible and whether they are capable of speaking to, and generating, literary affect. By emphasizing the affective deficiencies of code, she maintains an ontological distinction between programming language and literary language, even as she equates some of the historical properties of the literary, e.g. `beauty,' with the functionality of code.

The art of code and the practice of codework has a socio-cultural history, more specifically origins within origins, and it is not limited to our contemporary moment. Its genealogy includes many instances of codes used as a medium for art, including Oulipo's Algol code poems and the use of computer instructions in their texts; the long-term tradition of generative aesthetics and poetic programming, such as Tristan Tzara's `algorithm' for Dadaist composition (including in a similar vein La Monte Young's and John Cage's instructional scores); ASCII art; the composition of Quines; and Perl poems.

Also see Loss Pequeño Glazier's short discussion of the beauty of code in Digital Poetics: The Making of E-Poetries (Tuscaloosa: University of Alabama Press, 2002), 103-110. reviewed in ebr by Brandon Barr ^5

There are other analogues: though not limited to an online environment, the formal, aesthetic, and political principles of codework as a general category are echoed throughout, and indeed informed by, net.art discourse: the two often have a common basis in hacker culture; Open Source advocacy; anti-corporatist politics; authorial and textual self-reflexivity; and software technics. Such a broad shared platform suggests that net.art and the information arts, and not the Eastgate and Brown schools of hypertext, provide the socio-cultural, historical, and textual context for the branch of experimental media writing classified as codework. But codework is not a homogenous, monolithic, or pure genre. In his definitive introduction to the practice, and to a special issue of American Book Review, Alan Sondheim has carefully outlined a taxonomy of three types of codework: "Works using the syntactical interplay of surface language"; "Works in which submerged content has modified the surface language"; "Works in which the submerged code is emergent content."

The last, which has found a fairly wide-spread audience in net.art and digital art circles, exhibits a pronounced aesthetics of destruction and failure. In this branch of codework, the buried, or deep, code, is text and content, and because this code occasionally has damaging effects - varying in intensity and seriousness - the practice associated with it is sometimes classified as virus art. The Digital Craft organization has recently issued an exhibition catalogue on the "I Love You" virus that explores the possibilities of considering the programmers of this and other computer viruses as artists and linking them to the poetic appropriators of virus code active in net.art and digital poetry circles for the last five years.

For a more extensive and detailed account of the precursors to codework, see Florian Cramer, "Program Code Poetry." ^6 The net.art team Jodi might be understood as codeworkers in a similar fashion, with the obvious difference that Jodi's projects are not contagious or crippling in the way that a computer virus would be.

Florian Cramer also comments on "virus art" and notes that "computer viruses like Melissa and I LOVE YOU, small bits of text written in computer control code, strike me as perhaps the most dense and interesting examples of contemporary literature in the Internet." See Combinatory Poetry and Literature in the Internet (4) and Language, A Virus? Alan Liu's book, The Laws of Cool: The Culture of Information (Stanford UP), concludes with an extended discussion of the aesthetics of destruction and virus art. ^7 This type of codework, then, concerns itself with troubling the distinction between form and content, between surface and depth, such that one generally has to look at the source code to discern the content, functionality, even "meaning" of the work. To understand this particular practice of codework, we might take a general claim about all hypertext made by Mark Amerika and make it more specific: creative content and the source code, he says, are "one and the same thing" (Hypertextual Consciousness).

Jodi's work leads us toward what I take to be the most practically useful heuristic for critical investigation: a binary structure for codework that draws a distinction between code that is operational and has depth and code that is isolated on the surface of a text.

For a discussion of net art and its "aesthetics of failure," with a specific focus on spectatorship, see Michele White, "The Aesthetic of Failure: Net Art Gone Wrong," Angelaki 7:1 (April 2002), 173-93. Of relevance here as well is N. Katherine Hayles's analysis of print and digital textuality in terms of surface and depth, respectively; see, for example, "Print is Flat, Code is Deep: The Importance of Media-Specific Analysis," Poetics Today (Fall 2001). ^8 The codework that has depth exhibits what Cayley will call an "aesthetics of compilation," addresses the machine, and therefore "works," wields a performative power not equaled by the codework that treats code as a found object, a ready-made that operates as a static work of art. For Cayley, the performative power of the text authentically and genuinely resides with operable code: the difference is that between Roland Barthes's theorization of the potentiality of language and the potentiality of an algorithmic poem, which changes the system in a materially visible way. That is, the codework that has depth, that is pure rather than mere "decoration or rhetorical flourish," signifies within the realms of both natural and programming languages; it can continuously function and be legible within both systems; and it is capable of altering either one. No mere game of techno-cultural reference or "language of the tribe," as TS Eliot might say, this practice of codework differs structurally, metaphysically, and practically from the codework that incorporates static, non-functional elements of code into the surface, or "Interface text." Specifically, this working code has a "genuine" rather than "pretended ambiguity" of address; it is simultaneously addressed, in other words, to the human and to the machine.

Cayley also powerfully critiques the critical tendency to refer to the intermingling of natural language and codes as a creolization, but this is not to discount the linguistic changes signaled by the codework of the surface. The codework texts that remix natural language (usually English) and programming languages result in a kind of hybridized, electronic English, a language not simply suggestive of digital and network culture but a language of the computer and computing processes. When English is filtered through the languages of the machine something like this type of codework emerges; it is an English that has been manipulated and encrypted into an electronic computer-speak. Florian Cramer terms this hybrid of natural and machine languages "post-combinatory," suggesting both a fusion and an integration on the one hand, and a separation, or an incomplete mixture on the other ("CP" 5). This emergent language is ubiquitous within contemporary advertising and mass media: it is the idiolect of mobile phones, pagers, instant messengers, and chat settings, even appearing in a recent back-page New Yorker cartoon by Roz Chast (February 4, 2002), "The I.M.s of Romeo and Juliet," which updated Shakespeare to a scene of bedroom-to-bedroom Internet messaging, complete with the abbreviating and punctuating elements of much current codework. The basic semantic formula includes these abbreviating and punctuating elements: substituting numbers for letters; n = `in' and `and'; use of the dot, brackets, hyphens, slashes, and the doublepipe. One can only assume, too, that this type of codework will continue to incorporate the alphabetic, non-alphabetic and metacharacters of programming languages, such as the "??" ("hookhook") operator, as they move into widespread use.

And, as the forms of digital writing and art continue to change in relation to production environments and programming languages, it is interesting to speculate whether increasing institutionalization will mean that these high-level machine languages will be sanctioned for study as a "foreign" language within the humanities. The different kinds of institutional training programs developing now, such as those at UC Santa Barbara, UCLA, and Brown in the U.S., suggest that we will indeed see programming languages included within degree certification programs in literature and the arts at the undergraduate and graduate levels. At the moment, however, what we see in most codework writing and art practices is less code per se than the language of code: codework that integrates elements of code into natural languages and brings code to the surface as a medium for literary, artistic, and experimental composition. The codework practice of "netwurker" Mez, which again involves the use of a made-up code language as a mode of artistic composition and everyday communication, is paradigmatic.

An overview of Mez's work can be found in her recent JavaMuseum solo show (February 2002), commemorating her nomination as "Java Artist of the Year 2001." For a sample of the critical analysis of her work, see LaCook; Beaubien; Esdaile; Beiguelman; Stephanie Strickland; and Rossitza Daskalova, "Language Transformed by the Machine," CIAC Magazine 13. Also, Mez keeps an archive of clippings at http://www.hotkey.net.au/~netwurker/nav.htm. ^9

Mez's medium is a neologistic net.wurked language that she has termed m[ez]ang.elle, which has at least two narratives of origination. One is almost hallucinatory and never presented in straight English. In 1995, while working on an HTML document _cutting spacez," she realized that actively networked communicative circuits produce mixed and entangled data streams: "jumping fromme one terminal 2 the next//running three chat-rooms at once via three different terminal [behavior]z sew as 2 opt.tim[id no lonah]ize the time d-layzm chairz blurring b-tween as the monitorz flashed fiction wurdz that [k]neededObleed.ed e.vent[ingz]ualli into the cutting spaces doccoO."

See her rhizome.org interview with Josephine Bosma, February 1, 2000. She elsewhere terms this process the "de/reconstruction of language" but, in that this is a fairly broad statement, I am not convinced that this is the most useful description of her work. See her interview with Streetpress. ^10 The other narrative of origination is more concrete: mezangelle emerged from Mez's borrowing and manipulation of text from email lists and chat settings and specifically from a lengthy email exchange with Matt Hoessli from the CADRE Institute on the 7-11 mailing list, beginning in 1996 and continuing with experiments with members of the Webartery and trAce online writing communities.

Though her codework has "][r][e.volved" and become more visually complex in its presentation with Flash, shockwave, and JavaScript, her methodology remains basically consistent: she continues to work with text, frequently taking portions of a network communicative exchange, and then mangling them with machine language elements such as brackets, colons, slashes, hyphens, and the double pipe (||). One of Mez's techniques - one among many she shares with NN/Antiorp and other codeworkers - is to replace letters with symbols and with other letters ("2 4m a text"), in a colloquial style associated in a general way with IM and musical forms like hip-hop and urban youth culture (for instance, `z' substituting for `s' = `boyz'). Incorporating elements of emergent idiolect of the computer, computer languages, and digital culture, she also makes heavy use of the dot, as a connector that groups together words or partial words; to name hierarchies, as with domain names; to form abbreviations and word combinations; and to separate an object from its properties, as with OO programming conventions. The result is a hybridized neologistic language rich with semantic and poetic possibilities. With some graphical and lexical consideration, then, m[ez]ang.elle - with its play on `elle,' `angel,' `mangle,' and `angle' - suggests a pointedly feminist aesthetics and praxis of linguistic mutilation.

In this respect, as well as with their shared concern with the bio-politics of the body, there is a striking resemblance between Mez's aesthetics and those of noted hypertext author Shelley Jackson. See Jackson's website, Ineradicable Stain. ^11 After reconfiguring the data elements of her source texts, she directly and indirectly re-channels them to the forums from which they emerged, broadcasting them either through listserv posting or through web-based net.art installations. As her manifestos and statements of artistic principles would suggest, this codework language is something of a formal experiment, but it also reflects on the medium and its artistic potential.

That this "net.wurked" component of her artistic practice should be understood as axiomatic is evident from Mez's frequent and reiterative email postings and statements of methodological principle on the subject. For example, Mez notes in the artist's statement accompanying her solo show on Java Museum that her writing has at its base an entire archive of ongoing, collaborative, and "multifarious....directed email/irc exchanges and performances." Further, her list-based commentary on The Art of M[ez]ang.elle.ing stresses that one of her artistic aims is "2 network 2 the hilt N create de[e]pen.den[ting]cies on email lizts for the wurkz dis.purse.all." She introduces her critically acclaimed data][h!][bleeding texts by noting that ".these t.ex][e][ts r _code wurk_ remnants d-voted to the dispersal of writing that has been n.spired and mutated according 2 the dynamics of an active network." And, in the introduction to her curated gallery exhibition at Incubation 2002, she describes her technique as "largely dependent upon an electronic method of production that is exclusive to a networked environment, and shares some characteristics with the very format that houses it [such as n- tegration of email/Web browser/IRC jargon and stylistic blueprints."

Because codework participates in a long-term discourse of referential layering and linguistic wordplay, Mez's techniques have also been preceded by experimental, "avant-garde" Language poetics, concrete poetry, and visual poetry. Although the symbolic database and mode of distribution differ, in both can one find a use of brackets, white space, punctuation, and techniques of excision and negation that is at once excessive and indeterminate. We can, however, again speak of the difference or particularity of Mez's theoretical and practical aims in relation to her poetic precursors, particularly with respect to the mechanism it uses for composition and transmission - its context, and her relationship to that context. The sheer ubiquity and viral spread of both her codework texts and her codework practice echoes back to the very networked environment in which it is situated.

This insistence on the production and delivery environment of her mezangelled texts brings to our attention the necessarily collaborative nature of codework, appropriated as much of it is from listservs and chat settings. Mez and Alan Sondheim, for example, not only build codework texts from communicative exchanges but they also incorporate the responses of their audiences and fold them back into their codework texts, in a continuation of the collaborative production of the text and a fulfillment of the logic of email itself. The reply function, in other words, is coded into a text that is distributed across networks. In this sense, the relationship between codework-text (system) and its generative field and ultimate audience (online environment of chat settings and listservs) could also be understood in terms of feedback, whereby system and environment are both altered as a result of their interaction.

Feedback, Principia Cybernetica Web. ^12

The writing of Mez and other codeworkers compels an extended technical dissection, or a line-by-line parsing that would explain the significance of her use of each and every element of different programming languages such as Unix, Perl, and Java (she does not tend to borrow from machine or assembly code but rather from the higher-level languages that are closer to English in structure and vocabulary). However, the fundamentals of that kind of analytical-tutorial work have already been established. Florian Cramer, one of codework's most prolific and important critics, has circulated intricate and extended annotations of one of Mez's "wurks," _Viro.Logic Condition][ing][1.1_, and its thematics of the virus, infection, and the mutual contamination of the organism and the computer.

Cramer text sent by mez breeze in a personal email, "n.sights & n.][elec][troductions" (June 14, 2002).^13 Some of his most illuminating commentaries consider the significance, within this specific text, of Mez's use of square brackets; Perl header lines "::"; and the Unix commandline double pipe "||" - the logical `or' condition - all of which open up multiple grammatical and semantic possibilities within a "line" of mezangelled poetic prose. A rigorously formalist analysis akin to Cramer's is often not equally revelatory or satisfactory with Mez's current work. In this sense there is a difference between her early and more recent practices: "mezangelle" has evolved to become a manner of writing, an idiosyncratic style that bears the signature of its author. It is not generally executable code, and it often does not even allude to the functional component of her code elements, but instead functions as a mode of poetic composition and general communication.

In that mezangelle incorporates the punctuating elements of programming languages, such as the dot and the bracket, it partially revives the old, and well-digested, postmodern (or, "(post)modern") language games played with parentheses and slashes for largely similar general purposes. As Mez notes of her use of the now-expanded database of punctuating "elle-E.ments," the aim is to "condense/dilute/refresh wordage and imagery meanings/established codes/cues of associative meaning/s" (The Art of M[ez]ang.elle.ing). So, if there were a critical exhaustion with the kind of linguistic double plays afforded by a set of parenthesis, Mez counters by using punctuation particular to the apparatus - Talan Memmott calls this set of punctuation "technical ideogrammatics" - such that there is no true single word to destabilize or negate. Witness three signatures from her current Webartery listserv threads: Cur.][O][va.ture;][co][De][e][p.rivation; and app][lick.ation][end.age.

By making widespread use of brackets and periods to split words (somewhat hypertextually) into multiple component parts - letters, prefixes, suffixes, and phonetic elements - Mez's strategy is to disturb, disorient, and defamiliarize, to shift "the units of information and communication from the usual and expected to the cryptic" (rhizome.org interview). Such a cognitive disturbance is also partly facilitated by her use of dynamic text elements, as one can find in her data][h!][bleeding texts. Her repetitive technique - the encoding of text into this new mezangelled form is more or less consistent throughout her various texts - produces patterns of dissolution and reiteration that continue to disorient, even once one has more-or-less mastered the rules to the point of perceiving the patterns at all. For Mez's art and writing, the political purpose of this aesthetic practice is precisely to imagine this kind of interruption and disturbance of the economy of informational transaction.

Given that Mez operates in an expanding field of codeworkers, most of whom also work with "technical ideogrammatics," whether they be executable or not, establishing the specificity of her artistic project should necessarily reach beyond punctuational difference. Such an analytic approach would be thwarted quite quickly: for example, Mez's putatively particular use of bracketed expressions in "m[ez]ang.elle" can be analogized to Talan Memmott's embedding of command structures in his Lexia to Perplexia ("PER[(p)[L(EX)]]ia").

Deena Larsen briefly links Memmott and Mez in a short essay on e- poetries in Currents. ^14 The components of Mez's work similarly invite comparisons to Netochka Nezvanova's style and artistic practice, which also involves a surface play of code, listserv posting, and ever-changing display signatures, avatars of a sort (Nezvanova performing as NN, Antiorp, Integer, and Fifo). In both can one see an artistic personality produced by and coded into the text. One exemplary and distinctive formal feature of NN's texts is the substituting of the "!" for "I," which in effect gestures toward a negation of the self (See Beatrice Beaubien and David Johnston on the parallels between them). Rather than splitting symbolic hairs and claiming particularity on the basis of NN's use of the operator "!" (= not) or Mez's reiterative use of square brackets, then, it is possible to take a wider look at Mez's placement within the field. First, there are quantitative differences. Her seemingly inexhaustible productivity - surpassed one can imagine only by Alan Sondheim - has directly contributed to her prominence, ubiquity, and the frequency with which critical commentary on her work appears. She has been able to capitalize on her numerous awards and commissions - all prominently detailed on her web site - precisely because she is both networked and extraordinarily skilled at networking. Her institutional currency also derives from her manifestos and from her frequent articulation, theorization, and defense of her project as a project; in that it comes bundled with its own explanation and critical commentary, the mezangelle project legitimates itself as poetic and political praxis.

Mezangelle, too, makes for distinctive sound bites, both because of its quotable form - it is not as easy to extract self-contained lines from Warnell's or Cayley's codework - and because of its thematics, which resonate with theoretical investigations in the field: the organic-machinic divide, infection, gendered subjectivity, data, and networks. Her work is thus situated within the apparatus in two respects: formally and materially on the one hand (through its technologies of inscription), and conceptually on the other. As a whole, her mezangelle project is reflective on the apparatus and on its cultural context. Again, this last description applies also to Memmott's Lexia to Perplexia ("PER[(p)[L(EX)]]ia") and Warnell's