The Question Posed by Computer Generated Texts

Computer generated texts are primarily composed of a set of words chosen by a human that are randomly selected again by the computer and then added to a specified location in a defined sentence. The majority of computer generators take this basic definition and expand upon it vastly, adding many different sentence forms or specific topics, requiring that certain words rhyme, ect.  The addition of these mechanisms are closely related to Tariq Ali’s Ideas on Human Curation. While his article is focused primarily on the lack of “common-sense” among these text generators, he also brings to light the human curation of these text generators. The term “human curation” in this sense, refers to a human or group of humans reviewing the output of a text generator and choosing the “good” examples to display to the public as art. He argues that human curation can create meaning out of seemingly meaningless random texts, and asks the question “if machines generate literature, and then humans heavily edit the literature before publishing it, then was the final output ‘really’ computer-generated?” (Ali) Although it is not explicitly stated, I feel his argument actually brings to question the idea of author intentionality which we have discussed in this class.

Author intentionality states that the author’s intention when creating the art should be valuable to your own interpretation of the piece. However, if there is no true author, is there a true intention for the piece by any of its supposed authors? In Thomas Wendt’s piece on author intentionality and interpretation he argues that the most important factor linking these systems for electronic literature is the context of the work. For example, take a computer generated but human curated piece of poetry, the most important factors in your interpretation of the piece would have to be in the piece’s contextual clues. How it was displayed (if displayed digitally), who curated it, who wrote the code, what text(s) the computer is drawing words from to generate the poem, and other contextual clues will fill in the interpretive holes that not having a true author leaves. (Wendt)

The contextual clues of my project are key to understanding it and the artistic value it represents. My project is a generator that creates a short fake assignment given by either famous scientists or people from this class. These fake assignments are very difficult sounding and humorous, as I intended for them to be, so from a stranger’s perspective it is a whimsical little generator with little to no deep expression. However, the true meaning of my project lies in the contextual clues in my personal life outside of the project. I was creating and finishing this generator in the midst of recovering from a concussion, therefore even the most remedial assignments seemed impossible for me to complete. I decided to express my frustrations by creating this project which brings humor to the idea of impossible homework but also displays my inner frustrations in trying to complete this work. The reason for the italic black font is that it is hard to read at the low-level of back light that is required when coming back from a concussion, while the dark red background is symbolic of willpower, the same willpower needed to work when it is challenging. I got the ideas for this kind of contextual art by reading through the article by P. D. Juhl on computer poetry challenging author intentions. He believes that computer generated texts are art before they even generate a poem or text. This idea rests on the fact that a human wrote the code specifically for a reason and that the true art is seen through viewing what they intended the generator to output, and less focused on what the generator actually outputs. (Juhl) I hope you can see how I reflected his idea through my own work, thank you for reading!


Ali, Tariq,. The “Commonsense” Problem In Computer-Generated Works,, The DEV Community Blog, Aug 26, 2016.

Juhl, P. D., Do Computer Poems Show That an Author’s Intention Is Irrelevant to the Meaning of a Literary Work?, Critical Inquiry Vol. 5, No. 3 (Spring, 1979), The University of Chicago Press.

Wendt, Thomas,. Intention vs Interpretation: What Matters?, UX Booth, May 7th, 2013.

The Quest Generator

The Quest Generator

This project is called The Quest Generator is a program that creates a set of three quests for an aspiring knight trying to win honor, glory- and some all-important gold. However, the master of arms can assign him quests of different difficulties and with different weapons to prove that he is worth of becoming a knight. The most important quest of all is to defeat the mighty dragon that has been laying waste to the kingdom, and this encounter can end in a couple different ways, some deadly, some quick, some surprising. Hopefully, this quest simulator will remind you of text based role-playing games like Zork or the Oregon Trail, and while the user does not have any input on situations in the quests, each playthrough can feel unique even though you know some of the outcomes that are possible.

This quest generator has well over a billion different combinations, so it is very unlikely that without serious effort that you will be unable to see the exact same output. When Mark Marino wrote his article, “Critical Code Studies,” he quotes Donald Knuth who said “Instead of imagining that our main task is to instruct a computer what to do, let us concentrate rather on explaining to human beings what we want a computer to do,” and I think that this is an important distinction to make about computer created projects. They may be imitations of human created works, but we should not look at randomly generated outputs to generate meaning. Knuth continued, “the practitioner of literate programming can be regarded as an essayist, whose main concern is with exposition and excellence of style.” Programs like the Quest Generator are imitations of text-rpgs, and it is interesting seeing the different sorts of quests that the knight to face and live to tell the tale about. Sometimes the outputs of early computer programing were meaningless, and the ways in which the created stories meant that the outputs would have no logical sense. The Quest Generator will always create a sequence of quests that have a logical progression, and unlike the early computer generated programs, there is a structure and progression to the quests that make it feel more like reading an authentic human game or story. It also creates a huge number of outputs, and not even considering the different outputs from the number of hits and monsters, there can be over 9 billion outputs. This is one of the most interesting aspects about this project, and when Turing and Strachey created and love letter generator, the textual output was disappointing but it created a vast number of combinations. When they wrote about the program, Strachey said “the chief point of interest, however, is not the obvious crudity of the scheme… but in remarkable simplicity of the plan when compared with the diversity of the letters it produces.” I think that is one of the most valuable attributes about The Quest Generator, and this small program of only 136 lines is able to create more stories than I could ever be able to come up without charting it out.



Marino, Mark C. “Critical Code Studies.” Critical Code Studies | Electronic Book Review, Accessed 25 Sept. 2017.

Simanowski, Roberto, et al. Reading moving letters: digital literature in research and teaching: a handbook. Bielefeld, Transcript Verlag, 2010.


Superhero Mashup

My project “Superhero Mash-up” was inspired by my love for superhero, action movies and randomness. My biggest inspiration is randomness. Imagining the endless possibilities when you involve randomization and options is hard, but is made simple with computer programing. Randomness was such a huge inspiration for me because of the importance and effect it has on the world. As William S. Burough said in “The Cut Up Method of Brion Gysin,” ” There is nothing new under the sun.” That statement is true to a point. With the many powers of randomness you can take what’s old under the sun twist and manipulate it until it a few ways, and make it appear to be new. Randomness is so fascinating because it has the power to take an ordinary or familiar thing and make it interesting and original. The cut up method is a perfect example of that. You can take a book, poem, or article that you have read before and randomize a few things to make it into a new creation. This is similar to what I did with my own project. People continuously argue about what superhero would win in a fight versus another superhero (usually superman Vs. Batman). No one could ever agree on an answer because it is hard to simulate the power level and attacks of the superheroes. This made me think about a question: is it the super hero or the super power? What if your favorite super hero got stuck with another superheroes power for the day? How good of a hero would your favorite hero be now? This is what inspired me to use the power of randomness to get the ball of imagination rolling. I like how much power the article gives random. The quote that stood out to me the most is that ” All writing is in fact random” (Buroughs,3). I enjoyed this quote so much because I mentioned something similar to it in my blog post when liar to it in my blog post. It made me think about how everything is derived from other things.

My project also makes me think about the article ” Death of the author.” With the power of randomness and computer programing my superhero’s can have any power possible within the limits of the vocabulary I give it. This opens up the interpretation of the reading. The combination of superhero’s and super powers are over a million. As said in the article this turns the interpretation to the reader and no longer in the hands of the “author.” As mentioned in the article, ” the birth of the reader must be at the cost of the death of the author.” I have realized through creating this story that after giving the computer program the necessary tools to be successful at creating a story, I am no longer the author only the creator. The author has become Randomness and the producer is the computer.



Burroughs, William S. “The Cut-Up Method of Brion Gysin.” The New Media reader(2003).


Roland Barthes, “The Death of the Author”  Image-Music-Text

Pokémon Journey Generator Artist’s Statement

There is something inexplicably exciting about becoming an active player in a beloved fictional universe. Being able to let my mind drift to a world in which I am loved unconditionally by a team of pocket monsters is pretty much what keeps me going on rainy days, so I created the Pokémon Journey Generator in order to make this brand of escapism even more accessible to my already easily distracted mind. Upon each re-roll, my generator creates a new character, that character’s Pokémon team, and a basic backstory for the reader. Most of this information is pulled directly from the games, including the character titles, regions, potential enemies, and of course, over 700 listed Pokémon.  It’s good, simple fun, and the sheer number of Pokémon options seems to make it so no two Pokémon are never repeated in any given created team (which, I admit, was a pleasant surprise for me as a creator).

When working on the project, I felt as though I was visualizing an image more than I was producing a text. This could perhaps come from the fact that my source material is primarily stored in my brain as a video game franchise and television show, which definitely changed how I went about creating the code. Initially, I had intended to write out a base paragraph and determine what sections would be lists when I was done, but found that it was significantly easier to produce this project in reverse: I made my lists first, and the filler text flowed in naturally afterwords. This aspect of my project reminded me of William S. Burroughs’ argument in his essay,  “The Cut-Up Method of Brion Gysin”, particular the idea he proposes on page 90: “you can not will spontaneity”. In this passage he explains that writing, particularly writing using the cut-up method, is often good due to chance and happy accidents, and furthers this point by saying: “cutting and rearranging a page of written words introduces a new dimension into writing enabling the writer to turn images in cinematic variation” (91). This concept felt familiar to me because of the way I went about my project: it came from a visual basis and was forced into a text. Though it is obvious that my tracery project is by no means a sophisticated piece of classical literature or poetry, which seems to be Burroughs’ focus,  it utilizes language and ideas that are fundamentally unfamiliar (considering it is based off a fictional universe), and turns them into an easy to approach alternate reality for whoever is reading . This is because it is so visually oriented, therefore emulating the ‘cinematic variation’ that Burroughs claims can be obtained by randomizing text.

Continuing along Burroughs’ vein of thought, I think what makes my project ‘good’, or at least an entertaining and wish-fulfilling read, is that the information does not always match up factually to a major Pokémon fan, but still maintains grammatical accuracy and remains believable within the world of Pokémon. For example, a player from the Sinnoh region would be very unlikely to catch the legendary Pokémon Articuno, which is native to Kanto, but my generator allows that to happen. It pushes the boundary of the world of Pokémon and therefore creates an even larger picture than each individual game possibly can, since it ultimately creates a mental map encompassing all seven regions and all 700 or so Pokémon in one brief block of text. This, ultimately – as Burroughs suggests – was somewhat accidental. It was difficult to trim down such a giant world into a pre-written paragraph for the purposes of this assignment, so I instead allowed my brain to drive me in what I felt was necessary to list, and those instinctual decisions, rather than formulaic planning, are what allowed the generator to become as vast as it did.

Also related to the massive scope of Pokémon as a universe, I found myself thinking about Roland Barthes’ essay “The Death of the Author” when creating my project. In the essay, Barthes’ explains that the consumers of text ultimately define the meaning of text apart from its history or author’s intentions, as texts become works of their own as soon as they are physically created. I found this particularly relevant to my topic, as I felt like a contributor to a world that has existed in so many people’s hearts since the 90s, which has never been written or worked on by a single person, as it spans across games, television, manga, etc. Given this fact, it seems obvious that there is no “author” to be credited for the creation of this world, so it is worth questioning what separates the ownership of the franchise from the ownership of my individual project. I was specifically interested in the ideas raised by the quotation on page 146, which states: “a text is not a line of words releasing a single ‘theological’ meaning… but a multi-dimensional space in which a variety of writings, none of them original, blend and clash. The text is a tissue of quotations drawn from the innumerable centres of culture.” This concept felt particularly relevant to my project, as I am quite literally taking words (often made up) from a world and contributing my own jumbled, random version of it. In this way, my text is indeed a “tissue of quotations”.

Ultimately, I had a lot of fun messing around with the code for this project, and even more fun refreshing the page for an hour after I was done.

Works Cited:

Barthes, Roland, and Stephen Heath. “The Death of the Author.” Image, Music, Text, Hill and Wang, New York, NY, pp. 142–148.

Burroughs, William S. “The Cut-Up Method of Brion Gysin.” The New Media Reader, edited by Noah Wardrip-Fruin and Nick Montfort, The MIT Press, Cambridge, MA, 2003, pp. 89–91.

“List of Pokémon by evolution family.” Bulbapedia, Accessed 14 Sept. 2017.

“Pokemon Trainer 2.1 List of Trainer classes.” Bulbapedia, Accessed 15 Sept. 2017

Artist statement: A post-apocalyptic story generator.

The most interesting part of word generators to me is that they can create such great jumping off points for authors which is what led me to creating my Post-apocalyptic story generator.  This project was influenced heavily by the Fallout videogame series.  For example many of the creatures and enemies encountered in the story are reminiscent of what you might find roaming the wasteland in Fallout’s universe and I wanted to recreate the atmosphere. An important aspect of the Fallout universe is that it has this sense of dark humor that i really wanted to capture. For example one of the species you can be born as in the generator is an elephant hybrid.  I chose the font because it has an old computer like feel to it and i wanted the reader to feel as if they were reading the story off of an old military computer.

While working on the generator it occurred to me that  my intention of creating a Fallout style atmosphere may not be received at all by the audience but rather they may interpret the setting and general feel of the story as something completely different. At first it aggravated me because the randomness of the generator makes it difficult to build a world which I am used to doing in regular writing. It later occurred to me that the point of a word generator i not to necessarily build a world or story to my liking but rather be a world that the audience can interpret for themselves. This is a key concept in Roland Barthes “The death of the Author”. Barthes establishes the idea that the reader is more important than the author in writing and exemplifies that idea by saying, “The reader is the space on which all the quotations that make up a writing are inscribed without any of them being lost; a text’s unity lies not in it’s origin but in its destination.” (Barthes, 148). Word generators are excellent for depiction of this idea as it’s very difficult to even get across what you mean and since you can’t convey your own ideas as well as you could through conventional means of writing it allows the reader to take what they wish from the generator and attach their own meaning to what they read. This is also gets at the idea of ownership in authorship because it takes the author out of the equation and allows the work to just exist without any connotations being attached to it. In a sense having no author or no meaning from an author allows a work to be free from certain biases that it otherwise would be attached with.

One aspect of word generators that is constantly overlooked is the creativity that they bring forth out of people. As I mentioned before i really wanted my generator to be a jumping off point for authors who were possibly suffering from writers block. The best way to avoid writers block to take a vague concept and run with it but that requires an initial concept that many have trouble creating. Word generators help fix writers block because they can create so many ideas and constantly do it. this is best said in Margaret Masterman’s, “The use of Computers to Make Semantic Models of Language”. Out of the entire article one very specific line stood out to me, “The computers advantage is that it does not tire;  it can produce an indefinitely larger amount of an indefinitely larger number of variants of any type of combination of words which the poet may desire to construct.”(Masterman, 36) The creation of word generators could perhaps be authors greatest tool as they could simply generate ideas whenever they like for writing. Many say that word generators are dangerous as one day all novels may simply be generated rather than skillfully crafted by authors but they are wrong, word generators are merely another tool in the belt of authors.

Works Cited:

Barthes, Roland. “The Death of the Author.” Image, Music, Text, selected and translated by Stephen Heath, Hill and Wang, 1968, p. 142-148.

Masterman, Margaret. “The use of Computers to make Semantic toy models of language.” Astronauts of Inner-Space, 1966, p.36-37.



The Institutional Response Generator: Your Hot, New, Go-To Tragedy Response HelperBot

The Tracery Project as a platform excites me because it gives an avenue for less technically advanced, but highly procedural minds to explore the predictability of texts. When I say “procedural”, I generally refer to “the procedural” nature of digital environments as described by Janet Murray in Hamlet On the Holodeck. She writes “the computer can be a compelling medium for storytelling if we can write rules for it that are recognizable as an interpretation of the world” (Murray 73). In other words, our ability to write rules for computers to generate content similar, or somehow meaningful in interpreting, the world around us is described by the quality of being “procedural”. She furthers that “the challenge for the future” is to make it such that “rule writing” is “as available to writers as musical notation is to composers” (Murray 73-74). The Tracery library created by Kate Compton is doing just that.

For my project, I created an Institutional Response Generator which aims to address the often sterile, shallow, hollow response of institutions (and especially academic institutions) to national tragedies affecting their constituents. Because institutions of higher education pander to a wide variety of bankrolling alumni and parents, they are inherently inclined to remain apolitical for the sake of keeping the peace. However, the onslaught of increasingly horrific incidents as well as an oppressed populace gaining increased critical consciousness thanks to pressure of progressive organizers has forced institutions to respond at least nominally to such incidents. The common response of an institution has several key traits, perhaps most notably, the lack of historical analysis regarding why the event may have occurred and lack of accountability on the part of the institution. The first aspect of this project that aims to address this is the intentional use of the passive voice in several of the possible generations (from “The #topic# is something we as #unifyingNoun# must address” to “sustained damage to several local mosques”, as opposed to having sentences with active agents and descriptions of the agents’ corresponding actions). Regarding accountability of the institution, the generator provides an entire branching mechanism to “dissociateFromGuiltyAgent” as well as to “avoidResponsibility”. The options within these branches represent common phrases institutions may use to redirect attention away from their own role in contributing to (or failing to stop) violence, and towards an individual or group that perpetrated violence. This is meant to be a commentary on the way we interpret our institutions as holders of values with mission statements, yet ask very little of them in terms of accepting a role in the creation of dominant culture and subcultures.

Another key feature of generic responses is the response of “advocateDialogueNotAction”. For any powerful institution, it will always be in the institution’s best interest to prevent real conflict and demands that require shifting who holds the power. Famous early wave feminist Sojourner Truth made the analogy, “I know that it is hard for one who has held the reins for so long to give up; it cuts like a knife” (Truth 65). To incorporate such notions into my project, I created (some Davidson-specific and some general) sentences that advocate for commonly advocated solutions that fail to meaningfully disrupt power relations, yet provide some semblance of lip service to the issue at hand. I attempted to replicate the dynamic in which such statements are framed as calls to action but actually seek to offer an alternative deliberately distinct from activism in a traditional sense.

At this juncture, it is relevant to note that several of the “artistic” points of the piece reflect the code itself rather than only the result that the code generates on a public-facing screen. This study of the signifying features of code has an entire category of scholarship behind it. Regarding the JFK: Reloaded video game, Mark Sample and Cindy Poremba engage in a lively discussion about what it means to be a documentary video game as well as whether the code behind a product is in and of itself a critique-worthy feature of the product. For context, the JFK: Reloaded game is a first-person shooter game in which the player is prompted to reenact the assassination of former President John F Kennedy in the most historically accurate manner possible. Poremba notes, “By shifting the notion of documentary away from the inherent properties of recording technology, objectivity, and authority, and by framing it as a matter of social negotiation” the game complicates our understanding of “documentary” as “objective truth” (Poremba 8). However, Sample notes that perhaps, through “three lines of code commentary” in which the coders reveal misogynistic intentions in the creation of the game, the creators “absolutely undermine the entire stated pedagogical project” by revealing parts of their positional framing (Sample 2). I attempt to similarly (yet, perhaps more intentionally) play with understandings of the critical ways in which our framing plays into coding endeavors. Thus, I encourage those interested in the decisions I made to check out the rest of my Tracery grammar file to explore the other symbolic decisions I made in the creation and naming of the different components of the code.

Overall, my project fits into the larger sociological scholarship of what societal change looks like, the larger digital humanities scholarship of what the non-functional features of code signify, and the larger digital affordance conversation. I hope to expand on the project by providing additional examples, sources, and page formatting, and put it on a public-facing platform (aside from where it lives, here on my personal website) at a later point. Thanks for reading!


Murray, Janet. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. Cambridge, MA, The MIT Press.

Poremba, Cindy. 2009. “JFK Reloaded:  Documentary Framing and the Simulated Document.” Loading… 3 (4).

Sample, Mark. “A Revisionist History of JFK: Reloaded (Decoded).” Play The Past RSS, 2011,

Truth, Sojourner. “Two Speeches.” The Essential Feminist Reader, Modern Library, 2007, pp. 61–66.


Hot New Beauty Launch Coming Your Way!

Need a new go-to concealer? On the market for a high-impact lippie that’ll last all day long? Well, word on the street is that there’s a  New Beauty Product Launch hitting stores soon, and rumor has it that it’s going to be HUGE. Everyone is already buzzing about it on social media – I hear the beauty queen Kylie Jenner herself will even be at the official launch party! What exactly is all the fuss about? You’ll just have to visit the official launch page to find out! Oh, and don’t forget to use your discount code at checkout!

Alyssa Edwards is all about those discount codes.

My project, New Beauty Product Launch, is a playful satire on the overly bubbly, plastic, and bubble-gum scented beauty industry that I’ve been an avid consumer of for almost 8 years. When you visit my site, you’ll be treated to a hot-pink page featuring a gorgeous, minimalist font, along with a lovely, un-retouched image of the beauty realm’s lord and savior, Kylie Jenner. Each time you reload the page, you’ll get an advertisement for a brand new product, complete with a full product description, beauty guru endorsement, and best of all, a promo code. Even though every now and then you may get an advert for a brow gel with a duo-chrome finish (not likely to be a reality anytime soon), it’s okay, because big brands and their beauty-guru partners can and will claim a product does just about anything.

To develop the list of words and phrases the grammar could pull from, I watched several YouTube videos on makeup, and noted how products were described, who the top beauty influencers were, and what types of products were most popular. Popular beauty slang (which should be noted also overlaps heavily with drag slang) formed my striking list of discount code options.

Codes on codes on codes

At it’s core, my Tracery project is a digital form of Brion Gysin’s cut-up method. Instead of snipping pieces of paper up and scattering them around to see where the pieces may land, I chose to take a slightly more involved course by giving my JSON file the exact words I wanted it to pick from, thus ensuring each result reads grammatically, if not logically as well. It is this adjustment of the cut-up process that leads me to place my project as almost, but not quite surrealist. The tossing of my digitally produced dice “might help to coax something up from the unconsciousness”, if the audience is familiar enough with the industry being parodied. If so, this generator could “resonate” so much with someone that they are inspired to finally create that duo-chrome brow gel mentioned earlier (Burroughs, 89).

As Mark Marino discusses in his essay on Critical Code Studies, code can be “explicate[d] like we would literature” -analyzed separately from its output to discover and clarify a greater meaning or implication (Marino). Although my Tracery project’s code is quite simple, it reveals my biases, intentional, and unintentional. No where though, does it reach the potential for “humor, innovation, irony, double meanings, and a concentration on the play of language” (Marino). Just by combing through the words and phrases I used, one can ascertain my deep familiarity with the beauty world. Furthermore, my disdain for certain new trends can be seen in some of the made-up product descriptions, like “drag-level coverage”, along with the illogically huge/tiny shade ranges and discount percentages. By analyzing the style.css file, my view of the beauty world as an ultra-feminine one reveals itself by way of both the code and the output, displayed for all to see as the Barbie pink background and royal purple text. If given more time and a chance to develop my coding skills, I could potentially implement some of these ironies and double meanings Marino considers to further the literary depth, and perhaps merit, of my work.

Lastly, the challenging question of authorship arises. There is no doubt that my role of author is “diminished” as Barthes writes – my intentions for this project would likely be an afterthought, if anything, to anyone stumbling across my piece. Although I took a hard stance on my role as the creator and author for my Twitter bot I blogged about last week, I must secede to Barthes this time. My text has no “secret…ultimate meaning” (Barthes, 147). Its ‘explanation’ lies in the “destination”, the audience to which it unfolds itself infinitely (Barthes, 148). Because my audience could originate from any possible context, there is no reason to “impose”, as Barthes calls it, my authorial intent on them. I trust that, if they are savvy to the makeup product realm, they will develop an informed interpretation , and if not, they’ll have a very quirky text generator to toy around with at the least.

Either way, I’m dropping everything and heading to the nearest Sephora to check out that new launch…


Barthes, Roland. N.p.: n.p., n.d. Hill & Wang. Web.
Burroughs, William S. “Introduction.” The Cut-up Method of Brion Gysin. N.p.: n.p., n.d. 89-91. Print.
Marino, Mark C. “Critical Code Studies.” Electronic Book Review. Open Humanities Press, 4 Dec. 2006. Web.

One Wild Night

In the creation of the tracery project, I am ultimately contributing to a long history of computer generated text. Text generation has been common practice since the 1960’s and continues to change the way we look at modern day text. Permutations allow for these changes, and can bring a new meaning to even the tritest statements. As expressed in “First-Generation Poetry Generators,” an article written by Christopher Funkhouser, “early works of text generation performed some type of permutation in that they transformed or reordered one set of base texts or language into another form” (Funkhouser 244). This excerpt directly relates to my project as it draws from a bank of text containing various different scenarios for how someone’s night could go. The inspiration for my project was drawn from the concept of The Hangover franchise. The Hangover movies are a ridiculous representation of the unpredictable turns one’s night can take in the presence of drugs. My project uses several of these scenarios illustrated in the movie and randomly generates a potentially disastrous night.

One Wild Night particularly addresses today’s party culture. This theme influenced my word choice for the grammers of this project. For example, #recluse# was used, rather than #quietnight#, to show how people are viewed as lame and antisocial for choosing to not go out and party every night. Additionally, grammers such as #modeofrecovery#, #neveragain#, or #partyperson# contain stereotypical actions of people after wild nights. This includes “vowed to never go out again,” or “decided to rally for the next party.” Furthermore, I included the #conflict# grammar, because where there are drugs, there are often altercations. One Wild Night attempts to generalize possible outcomes of one’s night, while relating to the Hangover, by showing the extremes of how it can take a turn for the worst when drugs are involved.

The tracery project is an example of a permuted text, as the ultimate meaning of the text can change drastically through every new generation. Funkhouser acknowledges that these texts may not be entirely coherent, but they are capable of giving new meaning to redundant language. Funkhouser could strengthen this argument by recognizing the “rules of reading,” as established by Peter Rabinowitz in the first chapter of his book “Before Reading.” The rules of reading are an understood code that authors are constantly challenging in order to sustain a sense of originality. These rules are also unknowingly followed by readers in order to get the best experience or understanding from a text. For example, the rule of notice states that there is an unspoken understanding of what one should pay attention to when reading a text. This pertains to elements such as symbols, motifs, setting, and title as they are instruments commonly used throughout literature to give a deeper meaning to a text, thus they are meant to be noticed. This relates to the tracery project, because in order to get a basic understanding of the intentions behind a randomly generated text, one must pay attention to static elements such as the title or the structure of the text. After recognizing these elements, the rule of signification explains how readers then have to take their observations and draw meaning from them. In the case of my tracery project, titled “One Wild Night,” a reader can see this title and make conclusions on why the text is structured the way it is. The next step in understanding a text is the rule of configuration, which is identifying a genre or structure in order to develop a sense of coherence. In my tracery project, as one reads the first sentence of the text they develop a clearer understanding for the latter parts of the text. Lastly, the rule of coherence is critical as it states that a reader should “read a text in such a way that it becomes the best text possible.” This is important as computer generated texts can be loosely structured and interpreted many ways, however this rule ensures that nonsensical statements can still be interesting.

In addition to the rules of reading, tracery also relates well to “The Four Essential Properties of Digital Environments,” transcribed in article “Hamlet on the Holodeck.” Written by Janet Murray, “Hamlet on the Holodeck” elaborates on the properties on digital environments. These properties include digital environments being procedural, participatory, spatial, and encyclopedic. Tracery is no exception to these principles. Tracery is procedural in the sense that it has the ability to execute a series of rules. Certain grammars are always followed by the program, and while the material of the text can vary greatly, the procedure by which the text is made is constant. Tracery can also be defined as participatory, just to a lesser extent than other digital environments. Tracery is still participatory in the sense that you have to reload the page to generate new text, however you do not have an influence on what text is actually produced, as in the case of some twitter bots that respond to questions directly posed to them. However, Tracery does embody the characteristic of being encyclopedic. It has the capacity to store countless combinations of phrases and grammars, which can be utilized to create unique experiences with every refresh of the browser. This characteristic really sets it apart from the limits of the book and “Hamlet on the Holodeck” effectively explains this affordance as it claims, “The encyclopedic capacity of the computer and the encyclopedic expectation it arouses make it a compelling medium for narrative art” (Murray 84). This quote captures the beauty of tracery as it is truly a tool with great potential to create many different forms of literature.

This project has given me a better understanding of what generated text is and what sets it apart from the customary physical book, and has allowed me to explore a new genre of writing in an interesting and comprehensive way.

Works Cited

  • Murray, Janet H. “Chapter 3.” Hamlet on the Holodeck: The Future of Narrative in Cyberspace, The MIT Press, Cambridge, MA, 1997, pp. 65–94.
  • Funkhouser, Christopher. “First-Generation Poetry Generators.” Establishing Foundations in Form,
  • Rabinowitz, Peter J., and James Phelan. Before Reading: Narrative Conventions and the Politics of Interpretation. Ohio State University Press, 1998.

Artist Statement: DSCR Generator

For my tracery bot, I decided to create a text generator that is a parody of the typical Dark Sous “challenge runs”.  It takes some of the elements from the games, such as the different classes and weapons, and combines them with humorous constraints such as only being able to level up one’s “words-per-minute” stat, or having to date every enemy before killing them. I wanted to take some of the seriousness that is usually associated with the series and turn it into a more light-hearted interaction between a player and the game.  As one could probably tell, I am a very big fan of the series and appreciate a lot of the intricacies of the game and its mechanics but I also think that people occasionally take the “difficulty” aspect of the game too seriously and this leads to them missing a lot of the nuances which make the series so great.

In a similar fashion to Olivia’s Horror Movie Plot Summary Generator, my challenge run generator employs aspects of the cut-up method to force the player to step back and change the way they look at Dark Souls as a platform just as the cut-up method forces readers and writers to rework how they perceive words and their uses.  Tristan Tzara said that “Poetry is for everyone” and Dark Souls can be for everyone as well. They do not have to only view it through a perspective that focuses on the difficulty of the game. They can focus on discovering the lore of the world they actively participate in. This includes the characters, covenants, weapons, armor sets, and environments they will come across. The player can also focus on the design of the game whether that be the artwork or the discoverable interconnectivity of different locations in the game as they find shortcuts between areas.

In her book, Hamlet on the Holodeck, Murray introduces us her four essential properties of digital environments. She states that they are procedural,  participatory, spatial, and encyclopedic. I believe that my generator fits three of the 4 properties Murray discusses.  It is procedural because it has to follow a very specific set of rules provided by JSON in order to even generate a set of text and if I did not program it properly, it would not function. There is a defined process in place that allows my generator to create over 1 million different combinations. I believe my generator is also participatory because I created a set of lists as the “input” which it can choose from to generate a text. An individual using my generator will also have to prompt it to generate another text. Finally, my generator is encyclopedic because I can choose to go back and expand the input lists within the code practically infinitely to allow it to generate an even greater number of possible combinations. The digital environment in which my generator is found allows it to store, organize, and recall upon a seemingly endless amount of information and is only limited by my own imagination.


Works Cited:

Burroughs, William. “The Cut-Up Method of Brion Gysin.” The New Media Reader, The MIT Press, Cambridge, MA.

Murray, Janet H. “Hamlet on the Holodeck.” MIT Press, 16 July 1998,

Artist’s Statement: “Master Letter” Generator

In addition to her prolific poetic career, Emily Dickinson wrote hundreds of letters throughout the course of her life. The majority of her correspondences are standard, informing friends and family of the trivialities of her day-to-day life. The exceptions to this pattern are three letters known as the Master Letters, written to an unknown correspondent referred to only as “Master” (Dickinson 140). It is widely accepted that the letters were intended for Emily’s lover because of their passionate, effusive, and heartbroken language, although this conclusion has never been proven. My procedural generation rewrites Dickinson’s Master Letters in a digital environment, challenging scholarly definitions of authorship and literary convention.

Roland Barthes, in his 1968 essay “The Death of the Author,” disputes the traditional practice of attempting to derive the meaning of a text based on both the author’s intention and his history, “his person, his life, his tastes, his passions” (143). Instead, Barthes claims that once the words are physically written, they become independent of the author (142). Each word has a distinct set of meanings and connotations to each reader, a “tissue of quotations drawn from innumerable centres of culture” that the author cannot anticipate (146). Further, the author does not actually create something new, but because he relies on a lexicon of pre-defined, pre-created words, he is simply “[imitating] a gesture that is always anterior, never original” (146). The words and sentences are brought into existence when they are processed and interpreted by the reader, whose previous experiences conjure feelings, musings, and images that give the text meaning (148).

According to Barthes’s essay, I, as the reader of the original Master Letters, define their meaning, not Dickinson. The Master Letter” Generator visually demonstrates the process that Barthes describes. First, I read through the three Master Letters and chose the sections that I would like to combine to create the project’s skeleton, the parts of the sentence that do not change between generations. Peter Rabinowitz’s essay “Narrative Conventions” outlines the rules, or procedures, by which a reader gives a text meaning. They include rules for how to “draw significance from the elements that are brought to our attention,” rules for making patterns emerge, and rules for making a text cohere (Rabinowitz 44-45). In making the aforementioned choice for the skeleton, I demonstrated my ability as a reader to determine which sections of a text are important and vital to remember, a process that Rabinowitz calls the rule of notice (43).

Next, I chose the words from the text that would be replaced randomly upon each generation. The original word conjured a string of words in my mind that were related to the original word, based on things like what I have read in the past, where I have traveled, or any number of experiences I have had. For example, I replaced the original “Daisy” from the line, “Oh, did I offend it – . . . Daisy – Daisy offend it” with a list of thirteen alternatives (Dickinson 167). I did not consciously realize it, but all of the alternatives I chose were related to nature (pansy, violet, cat, horse, river, lake, stream, etc.). Barthes would say that this is because the word “Daisy” conjured thoughts of nature in me, and Rabinowitz would call this the rule of signification. This follows for every list of alternatives that I produced. Each list demonstrates the complex, individual web of connections that is evoked in each reader, the web that gives the text meaning.

Although I enjoy the way that my project visually demonstrates Roland Barthes’s theory in action, it further complicates assigning authorship; in creating a project that demonstrates the death of the author, I have created a new text. In demonstrating my own authorship, my ability as a reader to decide a text’s meaning, I have simultaneously forfeited my right to assign meaning. The one who gets to decide the meaning of my procedural generation is its future reader. If my ability to determine the future reader’s interpretation of the text were not impossible enough already, the 9^59 possible combinations for an original letter make predicting the meaning for not only one, but all of those combinations, unfathomable. Therefore, my project reveals how, at least in some circumstances, authorship is a cycle. A reader interprets meaning from a text, gaining authorship, and then, if he or she writes it down, forfeits that authorship to a new reader. Barthes does not acknowledge the existence of this cycle, but I find it valuable to the comprehension of my project.

The physical form of my generator, as mentioned in respect to its ability to visually demonstrate Barthes’s and Rabinowitz’s theories, is vital to its significance. This particular digital environment is spatial because it “represents navigable space” (79). That is, the background of the page represents the paper on which Emily Dickinson would have written her letters. However, the generator is also encyclopedic in its “range of possibilities offered us as interactors in the seemingly limitless worlds of digital narrative” (90). The user would have to refresh the page 9^59 times in order to get even one repeated letter. The ability to create a number of combinations for this letter that seems never-ending to the human user is a characteristic found only in digital environments. I chose the spatial background of yellowed paper in order to evoke the original letter’s author, Emily Dickinson. The encyclopedic regeneration of new letter after letter continuously reminds the user that what they are seeing is a century and an environment removed from the original text. These two characteristics in combination urge the reader to question, like Barthes, who really writes the “Master Letter” Generator.


Works Cited

Barthes, Roland. “The Death of the Author.” Image, Music, Text, selected and translated by Stephen Heath, Hill and Wang, 1968, p. 142-148.

Dickinson, Emily. Emily Dickinson: Selected Letters, edited by Thomas H. Johnson. The Belknap Press of Harvard University Press, 1986, p. 140, 141, 159, 167, 168.

Murray, Janet H. “Chapter 3.” Hamlet on the Holodeck: The Future of Narrative in Cyberspace, The MIT Press, 1997, p. 65-94.

Rabinowitz, Peter. “Part 1: Narrative Conventions.” Before Reading: Narrative Conventions and the Politics of Interpretation, Ohio State University Press, 1987, p. 15-46.