User:Andre Castro/2.1/semniar-2012-10-10-summary

Project Proposal Draft

Short Description

For many of us our relation with spam emails consists in moving them from the inbox to the junk-mail folder, or even better, let them be detected and moved by spam-filters. However, if we take a look some of the unsolicited emails we receive we might find them curious, and might even amuse ourselves with what we find. We might consider them as not too removed from a literary creation, where considerable amount of inventiveness, and awareness to the current global situation and society soft-spots are added to produce persuasive messages.Besides curious glitches are also abound, such as the heavy employment of stereotypes, the unconventional uses of the English language, Google-translate misinterpretations, typos, word obfuscation, recurrence of text fragments, to name some. Being it such an idiosyncratic and rich area of text production I believe it should be embraced as a literary genre.

Consequently I want to experiment with the possibility of using it as a literary device, a template for writing and communication.

<strick>investigating its writing techniques, structure and categories. Based on that knowledge I intend to develop a system where spam methods will be applied to construct a text, over email messages, in collaboration with its receivers - anyone who has received the 1st invitation email and replied to it.</strick> <strick>If we take a close look at some of the spam emails that keep arriving at out email accounts </strick>

However, for various and legitimate reasons, we don't pay much attention, to spam email messages. Would we start to read, and even to respond to unsolicited emails, if rather than trying to persuades us to join some money winning (or loosing) scheme, they would seduce us into collaborative creation, into shared writing processes, or even simply to communicating with someone distant, unknown and different from ourselves? I would like to attempt to answer this question positively.

Methodology

My strategy for attempt giving a positive answer formulated above will consist in appropriating spam as a literary and communication device. In order to do it, I will device a system summed up in the following diagram.

DIAGRAM

1) To initiate the process I will take on the role of a spammer, by sending one unsolicited email message to a large number of addresses from unknown individuals.

2) If a receiver answers, her message will be replied from my side by employing a bot, who will analyze the received message at different levels, e.g. letter occurrence, word occurrence, keywords, and phrases, and produce a reply that will both relate and "respond" to the receiver's message, but also push into other directions. I try to tell short stories through these text fragments.

This circulation of messages between bot and the human receiver will go on a few times, after which it will be interrupted.

3) At this point the bot will step back function as a proxy forwarding the messages between two human receivers, who happen be in a similar stage of writing process.

The bot will server always as mediator between humans, either between me to the receivers and back, or between two receivers

This is possibility

Outcome

As a first raw outcome I want to all the communication history and content between myself(and bot) and the receivers. If possible I also would like to keep track of the receiver to receiver communication.

At the current moment I

A first test email

email

=

FOLLOWING PRACTICAL STAGE

• dependent on the responses
• possibily: developing various forms of making the process and results public
• reassemble the same elements into a different configuration

Project Methodology

For the development of this project it seems necessary for various topics of research to inform the project, in order for this to materialize.

Topics of research

• Spam: Collect spam; categorize it; analyze its structural functions; identify writing techniques used. [So far I have assemble a collection of spam and categorize it; you can look it up here
• Electronic literature (works that are the fruit of human-machine interaction). More specifically I am in algorithmic narrative generation, and chat bots. [ I began exploring the chat bot Eliza, hacking the code behind it, so it can respond convincingly to email]
• Email communication
• Audiences' engagement in fiction world: suspension of belief

Once I have gained knowledge and experience in these topics I believe it will be possible to put the various pieces together into the project realization.

Project realization

The method I plan to employ in developing this project consists in devising a pre-written screenplay, where certain variables a left open. This script will be given to a writing robot who, based on it, will have the task of producing responses to the receivers who answered to the robot's emails. Based on each responder's answers, combined with the bot's own choices, the screenplay variables will be filled and the subsequent actions taken, based on those variables.

step-by-step

What follow is description of how I imagine the project's central mechanism acting.

1. A first email is sent to a bulk of email address, containing a dubious, but somehow enticing and poetic message
2. Once a receiver responds, and consequently enters the game, s/he will be enticed to keep on answering, to keep on writing. The writing bot, making use of its capabilities, will write a reply that relates to receiver's answer, but also opens other narrative paths.
3. Several emails will be exchanged using this device. The more the receiver writes, the more the bot will get to know him/her, but also more wild and hopefully intrigued its replies will be. Meanwhile, all the written material produced both by receivers and bot is collected (to be put to later use).
4. At a certain point, the receiver who is in the game will be disconnected from the bot and connected to another human receiver, who happens to also be in the same writing stage. The central node - the bot - will remove itself as the central node in the communication, to replace it a connection between the two receivers will be established.

General Methodology of Work

The methodology I have employed in the constructions of my works generally begins by encountering a problem or frustration. I come across something in which I see potential, but due to some blockage, it is under-explored. As an example, I found a great wealth of audio material available at Internet Archive, however its interface tended to conceal most of the material. This frustration led me to ask how could the diversity and quantity of material become more apparent and accessible.

A stage of experimentation and research generally follows. At this point I try to broaden my question, find possible angles to approach it, find other approaches to similar problems, engage with my subject and experiment. The Engage with the subject seems crucial. It is a step towards the object of research, trying no to look at it from a distance, but to get close to it, to establish a dialog. This happen more clearly in Data Factory when I tried to engage direct-marking companies' employees into talking with me and clarifying my questions a bout direct-marking and users' privacy. Loose experimentation seems also important to my work methodology. From it a more precise idea about the materials I am engaging with is formed, and the final approach begins to surface.

The following stage is one of concretization of the elected approach, bringing it from its experimental stage, through various prototypes to its final from. It is at this point when most decisions are made, such as what to include or exclude, in what context should the project exist, how it will function, what will be its appearance. In the example I have been giving, I decided for an online radio, exclusively based on audio material from IA, that would group the sound materials under given programs with specific genres.

The resulting object is rarely, a solution for the problem which initiated the work process. Rather than providing an answer it will be an imaginative solution that ought pose more questions, and show the problem under another, perhaps unexpected, perspective. Hopefully such approach will challenge the value of the object in question, and allow it to reveal itself under another form.

Description of my works in the past year

To begin with I started by writing on the performativity of code. Departing from Katherine Hayle's notion of code performativity I went on looking at how does code perform. How does code, rather than being a descriptive tool, has the capacity to act, to change things in the real world, similarly to a performative speech-act, such as a judge stating: "I pronounce you guilty of the crime", and often with very direct and material effects.

Simultaneously I developed and online map of the streets of Rotterdam. This map - a striped-down version of the original map of Rotterdam - only displayed the city street names against a black background, on the browser. By clicking on each street a request was made to the City of Rotterdam's Archive to provide information on that particular street's name origin. That information was then displayed alongside the map.

What I seemed to be creating here was an interface to access city's archive through a different channel. By having the indexes of the archive - the street names - displayed on the screen I might see and ask for information on streets that I wouldn't be normally interested in. It opened up the possibility for a serendipitous navigation through the city name's history.

The same serendipitous exploration made its way to the next project, Radio Liberté Egalité Beyoncé (RLEB). RLEB was an online radio-station with a schedule composed of audio content fetched form Internet Archive. One-hour long cycles presented the listener with ten-minutes chunks of various sound materials, picked at chance the archive, from within a specific categories, such as french talk-shows, hiphop, poetry readings, ambient music, etc. The design of website where RLEG was hosted presented a drawing of radio, with which the listener had to interact in order to get the radio playing.

The essay written alongside the RLEB began by tracing the archive's historical ties to power, functioning as a legitimator of power and knowledge. It mentioned the ideology which underlies any archive, even folkloric archive such as Youtube, an which can be best perceived through what is kept from entering the archive. Never-the-less important differences exist between a physical offline archive and a digital online one: the latter has the potential of a wider access, and also to let its item be taken by visitors, without risk of impoverishing it. These differences result in an invitation for appropriation and transformation of the digital archive. When one of its items is taken and transformed, a new version of it appears, and a certain confusion emerges over its origin - which one is the original and which is fictitious? When remixed versions surface, they change the archive into a space of dialogue and emergence of discourses, rather than a site aimed at legitimizing power and knowledge.

Data Factory looked at direct marking companies' acquisition of consumers data, with the aim to target individual consumers with products which they might be more inclined to buy. It consisted of a small archive of recordings from phone-calls made to such companies. During the calls I tried to sell them their business main ingredient - consumer data - with the catch being the fact that "consumer" here meant only myself. Through the reactions to this simple short-circuit the ideology behind such business model became more apparent. The calls were gathered together in a website, with a certain corporate style, in which they could be listened to, in parallel to images illustrating or ridiculing what was being said.

The essay The Work of Audiences departed from Dallas Smythe's work On the Audience Commodity and its Work. In Smythe's view, the audience job consists in creating demand for advertised goods. What advertisers buy when placing an advert on a magazine, TV-show of website is the audiences' attention and their capacity to buy what is being advertised. It happens that the labor of this immaterial work-force changed under the Web 2.0, as noted by Andrejevic. Audiences are no longer simply asked to generate demand for advertised goods, but also to put their creativity to work in producing many of the contents that surround those adverts, and also make the Web. By doing so audiences also develop a community around their creations, and produce information about their behavior, which will be used in targeting advertisement directed at them. In exchange audiences receive a small amount of control over their creations, which are ultimately property of the company hosting them. Following this line of thought I asked whether there was a difference from the work of audiences on platform like Youtube to one like Wikipedia. The situation seems to be that although Wikipedia also lives on audiences' free creative labor, they are given more control over their creations, and engage in discussions over the functioning of Wikipedia. Is this difference enough for the work of audiences to be classified as less exploitative than on Youtube?

tO dO

• include examples of Eliza hack
• include examples of interesting spam message (beginning)

Aymeric feedback

28.11.2012

0. Receivers are anyone in your project, or spammers?

There are anyone. Spammers won't participate, they will source material from which the bot is constructed. Say we take Mr. Wang, a spammer I have been exchanging emails with, I know a bit about Mr.Wang that I shape the bot, so that it impersonates Mr.Wang (he is middled age bank manager, living in Beijing, wants to make money out of the account of a dead client of his bank; Writes his long and repetitive emails, using elaborate language that doesn't really saying anything, the google translate glitches of his text). Mr Wang will become the bot, and also he could be given a whole new (online) life. (I am thinking as I am writing, but this should be included in the proposal)

1. make a diagram to make more explicit the step-by-step part

Yes, think it will help understanding

2. The introduction of the human receiver vs another human receiver via a preliminary bot chat is the strongest part of the project, the whole chat bot ai stuff does not deserve so much attention on its own, so try to balance it better in your text.

I agree!

3. Section "General Methodology of Work" is too vague, unclear, cant really figure out why it's there. Besides, you take as an example your phone calls, but I'd say that unlike with the spam stuff, you had an objective and played dumb to extract information by means of surprise and confusion, to actually get a better understanding of the data tracking industry. With the spam, by making spammers talk, there was also an objective, that is to use other techniques to make alive the spam mythology as a whole.

Maybe this is confusing. I believe the title General Methodology is not the most clear. This section should be a look back on past work and figuring out what general methodology is constant, and should also give examples of past work. Maybe it should be called some like Previous Works' Methodology

Without any clear objective (maybe I missed it), there is a risk that you will be creating a complicated text generator that looses its connection with the spam world, and the fact that you struggle to clearly explain what will be the outcome or final object more precisely at this point seems to confirm the existence of such a risk. This point should be nuanced depending of the answer of question 0.

Will try to nail down the outcome