Proposal Development Funding Cover Sheetsites.lsa.umich.edu/.../336/2018/12/...Application-edited-for-website.p… · Proposal Development Funding Cover Sheet . Instructions: Please

Proposal Development Funding Cover Sheet

Instructions: Please submit the complete application package to [email protected] by 5:00 pm on Thursday, February 1, 2018 (late applications will not be accepted; there are no extensions). We recommend writing the narrative in Word and saving the file as a PDF for the final submission. Using Word will enable applicants to track on word counts; reviewers will not read responses that exceed the word limits outlined. This cover sheet, the budget worksheet, the signature form, CV's and the narrative PDF may be separate files from the application but must be submitted in one single email; a complete application package consists of all of these items. Narrative Application: Please observe the word limits and detail for question responses as outlined on the website and below. The notes included with each question reflect the kind of information the reviewers will be looking for and are meant to serve as a guide for answering the questions. Answer question 1 on this form, the cover sheet. Questions 2 - 4 are answered in a separate narrative document created by the applying team and saved as a PDF. Reviewers are especially interested not only in the value of your project, but specifically what collaboration adds to your project. The Narrative Application is one piece of the complete application package, please see above and the website for a list of all required items. 2. Describe your research project including the questions that motivate the research, your approach and the anticipated outcomes. (1000 words) 3. Describe your collaborative process. What will you work out or hope to accomplish during the Spring Term proposal development phase of your project? If you have previous experience with collaborative scholarship or if you have a model for collaborative work in mind, please explain. What is the anticipated role of graduate student collaborators, and what is your mentoring plan for graduate (and undergraduate, if relevant) students? (1200 words) 4. What contribution(s) will the project make to the humanities field(s) in which it intervenes, and to the humanities more broadly? How do you anticipate results will be communicated and to what audiences? (600 words)

Project Title

1. Identify the Research TeamPrinciple Investigator (PI) - name, title, dept affiliation(s) & unique name

Faculty Team Members - names, title, dept affiliation(s) & unique name; one team member per line with their pertinent information

Anticipated Graduate Student Team Members - include dept affiliation(s) & unique name - names are not required but if the team is funded, graduate student team members should be ready to BEGIN work on May 1.

Anticipated Undergraduate Team Members - if applicable

Application Narrative 2. Describe your research project, including the questions that motivate the research, your approach and the anticipated outcomes. (1000 words)

The algorithm is now a central problem and topic of inquiry among scholars of culture.1 In computer science the word refers to a step-by-step procedure for accomplishing a given task: “lather, rinse, repeat” is an example found in textbooks.2 Yet as human experience is digitally mediated, encounters with texts, media, space, form, and even reality itself are now produced by a step-by-step procedure executed by a computer, to significant effect. For the user, the reader, the citizen, and the audience, sense is computed—hidden electronic scripts are followed and calculations are performed to determine what rises to awareness. “That we are now turning to algorithms to identify what we need to know is as momentous,” Gillespie writes, as having relied on “the word of God.”3

Algorithmic Culture Demands New Modes of Research and Creative Work

In an “algorithmic culture,” our everyday habits create an ever-expanding but inaccessible archive.4 Most of what we type, photograph, say, or do is now logged, sorted, analyzed, republished, and repurposed toward ends we cannot see. This argues for a new kind of collaborative scholarship; we must overcome the complication that for the critic, only the products of these computations are accessible, not the operations or operands that produced them.5

Indeed new media “objects” like the digital algorithm cannot be subjected to traditional critique because they cannot be fully observed. Today the botheration of the male gaze is joined by the “coded gaze,” yet there is no man behind the curtain (or at least he is very difficult to find).6 Algorithms are the contemporary recipes for reality, but even if the magic formula for Google were revealed, it would be an abstraction complex enough that an engineer could not predict what result would obtain in a particular case.

Algorithms operate on data in ways that are both personalized and stochastic, the same computed Web page--or jail sentence--may never appear twice. Computed experience is built up from the bedrock of rationality, but it offers an incoherent, irrational plan for us. The many conflicting processes involved are fragmented, sometimes irritating, and for many purposes unknowable, even to their designers.7 To adapt a quote from computing, an artificial intelligence will beat you at chess without noticing the room is on fire. How will the humanities respond to a world running atop these technologies?

We Propose a New Collective Spanning Five Colleges

As our answer, this humanities collaboratory pre-proposal calls into being a new, radically-multidisciplinary collective of five professors and two doctoral students. Our task: to probe these omnipresent but otherwise invisible algorithms. Our tactic: To develop new forms of collaboration that will combine traditional humanistic scholarship with other creative practices. We represent the humanities, computer programming, architecture, design, music, and art -- five different colleges of the university -- and we aim to braid and intertwine them all.8

To justify this proposal development funding, our project is presently over-ambitious: we want nothing less than to interrogate this new politics of the dynamic, computed representation of the world. We want to develop a proposal to reveal algorithms at work and unearth elements of their inaccessible archives. The threat of the computer algorithm can also be our tool: We will entrain new critical cultural production to the new cultural production of computer algorithms. In Suchman’s phrasing, algorithms can be “both a method through which things are made and a resource for their analysis and un/remaking.”9 We will write algorithms that write algorithms, revealing themselves: the matryoshka of the digital.

Our intellectual grounding is as varied as the participants, but a cohering theme is “digital studies,” an area within the digital humanities. One strand of the digital humanities investigates computers as a resource for traditional humanistic work, while a second strand applies modes of humanistic inquiry to investigate computing itself. We are working firmly in the latter camp. Our group envisions a proposal for a series of collaborative artistic and scholarly projects that address the problem of knowing algorithms, divulging “algorithms and their others.”10

Toward FAT, Sensible Algorithms

We anticipate that we will transform algorithms into sensory experiences and thus reflect on their sensing of us. We wish to make algorithms sensible. As an ontologic inquiry we ask: how do our computers use us to co-produce what exists? As an epistemic inquiry this includes how we know what we know about algorithms and vice versa. The specific projects we propose may traverse terrains of algorithmic-ignorance and the absence of information in order to highlight its opposite. We expect to focus each of the proposed works on algorithms of wide significance. An algorithm might be important because of its prevalence: the Google RankBrain algorithm is used by over 70% of the US population whenever they perform a Google search. Alternately, algorithms might be important because of their consequences: the NSA’s SKYNET algorithm identifies terrorists for the US military’s drone strike “kill list.” When Edward Snowden leaked the fact that unsound machine learning techniques were being used in SKYNET, this led experts to estimate that the algorithm may have already inadvertently killed thousands of innocent people.11

These projects will instantiate and develop ideas from digital studies, including the subdomains of software studies, “critical making,” Web epistemology, and the

“ethnomethods of the algorithm.”12 Bridging the humanistic and the technical, these projects will draw inspiration from the diverse backgrounds of the faculty participants. We are inspired by the pioneering work of artists like the F.A.T. Lab (Free Art and Technology Laboratory), as well as the pioneering machine learning work of the FAT* (pronounced “fat star,” for Fairness, Accountability, and Transparency, where the star means “all” or “any”).13 As explained in the next sections, each piece will employ novel forms of collaboration to combine making, programming, and visualization with either performance, interaction, exhibition, or scholarly reflection.

[961 words] 3. Describe your collaborative process. What will you work out or hope to accomplish during the Spring Term proposal development phase of your project? If you have previous experience with collaborative scholarship or if you have a model for collaborative work in mind, please explain. What is the anticipated role of graduate student collaborators, and what is your mentoring plan for graduate students (and undergraduates, if relevant)? (1200 words)

Our collaboration model will be the artists collective. We envision an ambitious multidisciplinary group spanning five colleges. This isn’t disciplinary variety for variety’s sake, but rather a gamble that if we combine experts whose normal work practices involve making, building, doing, programming, performing, and/or designing (henceforth “making”) we may be able to make significant headway on the intellectual challenges facing the humanistic scholarly study of algorithmic culture. Collaboration plays a special role in this endeavor because the proposed project is grappling with a research domain where a wide variety of very technical skills may be required. Of course any advanced research project requires rarefied skills of some kind: our aim isn’t to discount other humanistic modes of work and sources of expertise, such as interpretation, criticism, historicization, curation, theorization, or preservation. Rather, it is an attempt to use intentionally unusual connections to create what Lovink calls “uncanny networks” that link people together as nodes of intellectually productive difference.14

Students (including undergraduates in the future full proposal) will be partners with us in the collective. They are involved because they have unique skills and perspectives to contribute, and not for use in a supporting role. The phases of the project specified below (see especially phase 5) have been crafted with mentoring and the needs of graduate students in mind as well as faculty.

We expect this to be a rocky road. We are combining collaborators from very different intellectual traditions in order to create useful frisson from their conflicting habits and expectations. Our training is riotously different. Some of our faculty come from a solitary “studio” or “practice room” tradition, while others are at home in libraries, offices, or labs. Some of us make books while others make things. Among our

group we teach classes with wildly differing formats, sizes, and pedagogies. Our students and our intellectual products would normally not interact. We use different jargon. Although subsets of the faculty on this proposal know each other, this particular group has never collaborated before.

This is ambitious, but it may be what is at stake is the future of humanistic work. We are intentionally pursuing a challenging collaborative form as a response to the Provost’s Office charge that the future of the humanities at Michigan be innovative, and in response to Sidonie Smith’s call that a future humanities be organized around an “ensemble” ethos. Our Proposal Development Timeline

If funded, we will use the May-June period as a proof-of-concept to see if a collaboration will work. Our group has some advantages that suggest bootstrapping a productive collaboration may be possible. Obviously we share an interest in the proposal topic. In addition, we have each grappled with instantiating something inaccessible by giving it a new form. Some of us call this an interest in “materiality,” others describe it as “embodiment,” or an interest in “invisible infrastructure,” “visualization” (or “sonification”), etc. Whatever the term, there is a common impulse. We also share an interest in culture and creativity. Members of the group have a higher-than-average tendency to cross disciplinary boundaries. There is also an ethics surrounding the importance of this topic that unites us.

We anticipate holding twice weekly co-working sessions with some participants joining remotely via videoconferencing. At every step we will also make the most of digital collaboration tools (for instance, our group will certainly have a Slack channel). Each of the following phases will take 1-2 weeks.

(1.) Acclimatization: Our first stage refers to “acclimatization” to the collaboration itself. At the first meetings this will involve presentations of our favorite inspirations related to the topic. We will then shift to briefly present our own relevant work, skills, and backgrounds--to include discussions about how we usually work and how we would like to work together. We will finally move to bring particular objects and experiences (e.g., via recordings) to share with the group in an effort to ground our work in the modality of critical making. In parallel with acclimatization, during this period we will also investigate the feasibility of specific kinds of making for our full proposal. This includes compiling the available digital fabrication resources and the skills that team members bring to the group. (2.) Target Acquisition: In our second stage we will research and report out to the group our findings about particular algorithms deployed in the world that seem worthy of an intervention. During this period we will also (in parallel) investigate the feasibility of using the humanities collaboratory space in Hatcher as a site to either host, work with, store, or display objects related to our full proposal. The specifics of this will not be known until digital fabrication is investigated as discussed above.

(3.) Prototyping: In our third stage we will employ collaborative design methods to imagine the potential forms of our full proposal’s work. We may not develop actual prototypes at this stage -- we might employ paper-and-pencil prototyping or storyboarding as well as other collaborative design techniques. At this time, in parallel we will task each other with outside research that supports the full proposal (literature reviews, viewings).

(4.) Specification: Our fourth stage will involve detailing the requirements that will need to be met in order to realize our prototypes in final form and therefore our proposed projects. This may involve the recruitment of additional team members, and will certainly involve a consideration of staffing, budget, and timeline. In parallel we will individually be given responsibility for writing a piece of the full proposal.

(5.) Evaluation: During this stage we expect to consider how we might gather measures that indicate our project is a success. Of necessity, this will involve a serious discussion of how faculty and students manage experimental, collaborative projects and expectations, as well as thinking through tactics that faculty (especially untenured faculty) and students might use to ensure that they receive credit for humanistic work that is not a traditional “output” in their chosen field. This will require a frank discussion of career risks that come with experimental work. We will also use this time to develop detailed plans for dissemination. In parallel, we will individually edit each other’s assigned sections of the full proposal, eventually finalizing it. The Hatcher Library Space

The physical distance between our participants is a significant barrier to this collaboration, and in part the Huron River explains why more faculty don’t collaborate across some of these administrative boundaries at the university. We expect the Hatcher Library collaborative space to be essential to make this collaboration work. We plan to use this dedicated space to co-locate our participants from North Campus: the Stamps School; the School of Music, Theatre, and Dance; and Taubman, with the participants from Central Campus: the College of Literature, Science, and the Arts and the School of Information. (This co-location is a rare occurrence and thus worth noting.)

We suspect that our eventual full proposal will feature the dedicated space for the humanities collaboratory in Hatcher library as the centerpiece of our collaboration, as our project also involves making objects and (in consultation with the collaboratory administration) we would like to use the spatial and material dimensions of our interests to greatest effect via the humanities collaboratory’s dedicated space. [1,198 words]

4. What contribution(s) will the project make to the humanities field(s) in which it intervenes, and to the humanities more broadly? How do you anticipate results will be communicated and to what audiences? (600 words)

Our approach will be iterative and experimental. As this is a pre-proposal, the ideas in this section are at an early stage. Nonetheless, at this time we envision a series of sub-projects as outputs with each project likely having a different form. Using the language of art, we will call each of them a “piece” although they may not be artworks. A particular piece might be an interactive experience. A common output across all pieces could be the virtual or physical “objects” that relate to them or constitute them (a map, software, a model, something fabricated, a recording, a database, a new instrument). These “objects” would be paired with a Web site for each piece.

It is likely that every piece will have some associated algorithm. This algorithm might be one that we create to reflect on an existing system that we do not control. Alternately, the piece may deal with a published algorithm that we reprint, enact, or transform in some way. Each piece may also be paired with an accompanying scholarly article or commentary. Publications will target venues such as the Journal of Digital Humanities, Computer-Human Interaction, Media-N, Computational Culture, Critical Inquiry, and Representations. For pieces where this is suitable, we will pursue appropriate exhibition, performance, and/or archiving.

One important goal of this proposal is to innovate in the format of our creative intellectual contributions. We propose that for each project we will employ performance, online interaction, exhibition, and/or scholarly reflection. We will then try to take these format(s) out of the disciplinary tradition where they are comfortable and/or combine them in unusual ways (e.g., “performing machine learning” or “programming scholarly reflection”).

We will now close by presenting one preliminary example piece drawn from prior work. We will not be held to this example, as the goal of our collaboration is to develop new projects we cannot envision now. In one piece we might apply Donath’s notion of “data portraiture” to Facebook’s “news feed” algorithm. Facebook identifies people and status updates that it determines are not likely to be interesting to you and hides them, filtering out as much as 90% of what a user might see.15 In the past we have queried Facebook for particular users, creating an estimate of what is likely to have been withheld. In a gallery or lab we could design and exhibit a personalized visualization of this information, then ask: Do you have a preferred alternative filtering or ordering rule? We hope to investigate the idea of a parallel, vernacular or “dream” social sort.

As a second potential element of this piece, we could build upon Fuller’s concept of “flecks” of identity as the primary compositional element within new media. The Facebook news feed makes filtering decisions in part because of each user’s behaviors and inferred demographics.16 We could make these explicit (e.g., reveal that the feed is currently filtered because of factors like “detected” sexual orientation and current bank account balance), then query supply-side advertising exchanges (such as OpenX) in real

time to price the categories involved. With some effort, this could also give us a mechanism to value “dream” or “vernacular” algorithms and compare them to each other or to Facebook’s typical operation.

It is likely that what is valuable to most people is often not valuable at all to Facebook advertisers. This piece would work to make this tension concrete with specific examples that would vary for each person, and the real-time structure would also mean that there would be no way to know what would happen each time the piece was used.

[594 words]

Notes

1 Lorraine Daston, “Whither Critical Inquiry?” Critical Inquiry 30, no. 2 (2004): 362; Malte Ziewitz, “Governing Algorithms,” Science, Technology, and Human Values 41, no. 1 (2015).

2 See also T. Wangsness and J. Franklin, “‘Algorithm’ and ‘Formula,’” Communications of the ACM 9, no. 4 (1966): 243.

3 Tarleton Gillespie, “The Relevance of Algorithms.” in Media Technologies, eds. Tarleton Gillespie, Pablo Boczkowski, and Kirsten Foot. (Cambridge, Mass.: MIT Press, 2014).

4 The term “algorithmic culture” is from Alex Galloway, Gaming: Essays On Algorithmic Culture. (Minneapolis: University of Minnesota Press, 2006); see also Ted Striphas, “An Infernal Cultural Machine: Intellectual Foundations of Algorithmic Culture.” (Paper presented to The Digital Humanities in Theory, Drexel University, Philadelphia, Pennsylvania, USA, November 7, 2012).

5 Rogers calls this “Web epistemology,” meaning the World Wide Web. See: Richard Rogers, Digital Methods, Cambridge, Mass.: MIT Press, 2013.

6 For the “coded gaze,” see The Algorithmic Justice League (https://www.ajlunited.org/). On the man behind the curtain: Nick Seaver, “Algorithms as culture: Some tactics for the ethnography of algorithmic systems.” Big Data & Society July-December 2017: 1-12.

7 Wendy Hui Kyong Chun, Programmed Visions: Software and Memory.(Cambridge, Mass.: MIT Press, 2012). Recent work in “explainable artificial intelligence” attempts to counter this trend, but the proposers are skeptical.

8 Stephen Ramsay and Geoffrey Rockwell, “Developing Things: Notes toward an Epistemology of Building in the Digital Humanities.” in Debates in the Digital Humanities, ed. Matthew K. Gold. (Minneapolis: University of Minnesota Press, 2012): 75-84.

9 Lucy Suchman, “Configuration.” In: C. Lury and N. Wakeford, eds., Inventive Methods: The Happening of the Social. (London: Routledge, 2012), pp. 48–60.

10 Paul Dourish, “Algorithms and their others: Algorithmic culture in context.” Big Data & Society, July-December 2016: 1–11.

11 See: https://arstechnica.com/information-technology/2016/02/the-nsas-skynet-program-may-be-killing-thousands-of-innocent-people/

12 Matthew Fuller, ed., Software Studies: A Lexicon, (Cambridge, Mass.: MIT Press, 2008); Matt Ratto, “Critical Making: Conceptual and Material Studies in Technology and Social Life,” The Information Society, 27, no. 4 (2011): 252-260; Richard Rogers, Digital Methods.; Malte Ziewitz, “A not quite random walk: Experimenting with the ethnomethods of the algorithm.” Big Data and Society, July-December 2017: 1-13.

13 F.A.T. Lab is now defunct. See: http://fffff.at; For FAT*, see http://fatml.org 14 Geert Lovink, Uncanny Networks: Dialogues with the Virtual Intelligentsia (Cambridge,

Mass.: MIT Press, 2003).

15 On data portraiture, see Judith Donath, The Social Machine, (Cambridge, Mass.: MIT

Press, 2014): ch. 8. This example is an extension of PI Sandvig’s joint prior work with Karrie Karahalios and Motahhare Eslami.

16 Matthew Fuller, Media Ecologies: Materialist Energies in Art and Technoculture. (Cambridge, Mass.: MIT Press, 2005), 148

** Other Support(Department, External

Grant, etc.)

Proposal Total(Include expenses paid

by Other Support)

Non Student Compensation and Fringe Benefits

PI Current Annual Salary/Full Time Rate (FTR) xxx

PI Summer Salary Support (auto calculated as 1/9th or 12,000 whichever is less) 12,000

Faculty/Librarian/Curator etc. Team Member Compensation

<list compensated (non student) team members below; total not to exceed $8K per person, $32K total>

8,000

8,000

8,000

8,000

5

6

7

8

Total Non Student or PI Team Member Compensation 32,000

Fringe Benefit Rate 30%

Total ‐ Summer Support 57,200

Graduate Student Team Members

(compensated as hourly research assistants)

# Hourly Research Assistants (0, 1, 2) 2

Estimated Total Hours Per Graduate Student [no more than 360 hrs per asst (9 wks * 40 hrs p/w)] 128

Hourly Rate (recommended $15‐$25 per hour) 25

FICA 8%

Total ‐ Graduate Student Team Member(s) Compensation 6,912

Undergraduate Student Team Members

(compensated as hourly assistants in research)

# Hourly Research Assistants (0, 1, 2)

Estimated Total Hours Per Undergarduate Student [no more than 360 hrs per asst (9 wks * 40 hrs p/w)]

Hourly Rate (recommended $10‐$12 per hour)

FICA 8%

Total ‐ Undergraduate Team Member(s) Compensation ‐

Travel: Where, how many attendees, purpose

Research Trip

‐

Research Trip

‐

Total ‐ Travel Related Expenses ‐

Misc. Other:

(equipment purchases such as computers are prohibited)

Supplies/Materials 150

Printing/Copying

Hosting

Other (Describe) 200

Total ‐ Misc. Expenses 350

GRAND TOTAL: ‐ 64,462

= Input Values (Collaboratory Funds) = Formulas

= Input Values (Other Funding)

Source(s) of Other Support

domain names

MICHIGAN HUMANITIES COLLABORATORY

BUDGET SUMMARY

May‐June 2018

1

Using the tab key after entering data in each yellow field will

take you to the next input field.

MICHIGAN HUMANITIES COLLABORATORYProposal Development Application

Funding Cycle: Spring Term (May-June) 2018

The Chair/Director/Associate Dean (in the case of a LSA Chair/Director team member)/Dean of each unit in which the applicant holds a funded appointment must sign this document.

The collaborative team of faculty will be funded to conduct intensive research and inquirytowards the development of the collaborative project idea and preparing the Project Funding Proposal if the team determines that is the appropriate course of action. Collaboratory Proposal Development funding supports work conducted during May and June, 2018

By signing below the Chair/Director/Associate Dean/Dean signifies awareness that the faculty team member has applied for Collaboratory Proposal Development Funding and agrees to support the terms of the fellowship (intensive Spring Term research/development and compensation per the terms of the approved proposal/budget).

PROJECT INVESTIGATOR (PI) ____________________________________________________________________ Printed Name

________________________________________________________________________________________________________ Chair/Director/Dean Unit 1 Printed Name Signature Title


FACULTY TEAM MEMBER 1 _______________________________________________________________________ Printed Name






**This form may be signed electronically, we suggest routing the form via email for electronic signature to ensure all required signatures are on one form.

Christian Sandvig

Beth Yakel Senior Associate Dean for Academic Affairs

Nojin Kwak Chair

Sophia Brueckner

William Calvo-Quirós

Alexandra Stern, American Culture Chair

Guna Nadarjan Dean, Stamps School


________________________________________________________________________________________________________ eltiT erutangiS emaN detnirP 1 tinU naeD/rotceriD/riahC

















John Granzow

riahChciveruGleahciM

Catie Newell

riahCraaHnorahS

Proposal Development Funding Cover Sheetsites.lsa.umich.edu/.../336/2018/12/...Application-edited-for-website.p… · Proposal Development Funding Cover Sheet . Instructions: Please

Documents