Notes on platform and infrastructure

From Driscollwiki

Jump to: navigation, search

Notes on platforms

Today, Pricescope relies on the phpBB software. phpBB is a free software project... etc. This is the infrastructure.

We might also say that Pricescope is a 'platform.'

Star's infrastructure explicitly includes "settings" as a site of inquiry

Number of settings in typical web-based messageboard software boggles the mind (see: vBulletin manual)

Explain how it is infrastructure

Note the ways that Gillespie and Bogost & Montfort's recent writing about "platforms" extends this understanding of Pricescope as infrastructure to account for institutional discourses and computational affordances

Embedded strangeness Boring

phpBB -> Recursive public

Relationship of Pricescope as infrastructure to the chosen software package

When Pricescope's admins decided to migrate from idealBB to phpBB, the users didn't attribute changes/bugs to the underlying software, they attribute it to PRICESCOPE
In other words, it's not that the phpBB search function is broken - the pricescope search function is broken
On one hand, this demonstrates the extent to which platform characteristics are not visible at the interface level
On another, once installed and "live" on the server, the new code is more pricescope than phpbb. The admins may or may not keep up with the on-going conversation of phpBB developers. They may not apply any patches. They may not install any plugins. What we do know is that they modify phpBB so significantly that Pricescope is a completely unique infrastructure. phpBB offers a starting point but the ultimate infrastructure is all pricescope

Examples

YouTube, institution
Atari VCS, computation

Clear division between, for example

YouTube employees, YouTube users
Atari VCS game programmers, Atari VCS game players

Pricescope does not fit neatly into either category

Some of the characteristics of YouTube - central authority, moderation, users can't program it
But the software (first idealBB, later phpBB) provides a structure on which new software can be written - on-going discussion
- "Recursive public"

Gillespie's discourse analysis demonstrates how the fuzziness of "platform" enables YouTube to continue to self-identify as a venue for popular discourse ("broadcast yourself") while making its strongest pitch to advertisers and "content owners"

See quote from CEO Hurley, "entertain and engage ... television" (9)

Web-based messageboards (idealBB, phpBB, vBulletin) are most like BASIC as described by Bogost & Montfort here:

Publishers of early books of BASIC code "were treating BASIC as if it were a platform - even though it wasn't even really a single language, but a family of a related languages, implemented differently" (4)

Platform studies might consider BASIC and other software systems as platforms despite the differences

Briefly mention the Andreessen quote used by both Gillespie + B & M?

Gillespie uses the term "infrastructure"

"an infrastructure that supports the design and use of particular applications, be it computer hardware, operating systems, gaming devices, mobile devices, and digital disc formats" (3)
and Bogost & Montfort quote him

Gillespie, The politics of 'platforms', 2010

Consideration for 'platform' in "populist appeals" and "marketing pitches"

Key example: YouTube

Various meanings of platforms, used inconsistently, "media platform"

Technical platforms
Platforms from which to speak
Platforms of opportunity

Managing a tension between

Facilitating user expression
Seeking limited liability for what is expressed

"As these providers become the curators of public discourse, we must examine the roles they aim to play, and the terms with which they hope to be judged." (abstract)

YouTube, self-described

"media platform"
"distribution platform" (2)

YouTube is one of a handful of "primary keepers of the cultural discussion as it moves to the Internet" (2)

"...increasingly facing questions regarding their responsibilities: to their users, to key constituencies who depend on the public discourse they host, and to broader notions of the public interest" (2)

Gillespie identifies "platform" circulating both in the self-identification of services like YouTube and in the discourse of users, journalists, and critics (2-3).

It may well be the case that "platform" has domain-specific meanings (such as that of Andreessen) but here we're looking at the meanings it accumulates in an extant discourse

"‘platform’ is a claim that arguably misrepresents the way YouTube and other intermediaries really shape public discourse online" (3)

"Platform" is "specific enough to mean something, and vague enough to work across multiple venues for multiple audiences" (3).

People are hearing what they want to / need to hear?

Four categories of use (4):

Computational
- Gaming device, SNS, PC, etc
- "an infrastructure that supports the design and use of particular applications, be it computer hardware, operating systems, gaming devices, mobile devices, and digital disc formats" (3)
Architectural
- Literally, flat raised surface
Figurative
- Metaphorical position from which to speak/act
Political
- Set of issues endorsed by a candidate or party

Platform suggests populism, accessibility, level, openness (5)

Gillespie traces the inconsistent use of "platform" by Microsoft to describe a programming language, an operating system, an advertising initiative, and a set of multimedia standards (5).

"Platforms are platforms not necessarily because they allow code to be written or run, but because they afford an opportunity to communicate, interact, or sell" (5).

Andreessen calls it "a swirling vortex of confusion" (6)

ask "can it be programmed?"
Reveals a disciplinary boundary
This word, which has some specificity (not much) in CS, is taken up by non-programmers to describe aspects of online/digital activity and it bums Andreessen out

YouTube's discourse activates all 4 categories of meaning found in the OED

"computational, something to build upon and innovate from; political, a place from which to speak and be heard; figurative, in that the opportunity is an abstract promise as much as a practical one; and architectural, in that YouTube is designed as an open-armed, egalitarian facilitation of expression, not an elitist gatekeeper with normative and technical restrictions" (6)

"Platform" brings together advertising, users, and professional media producers without agitating the contradictions among them (7)

Gillespie's discourse analysis demonstrates that YouTube's rhetorical work is primarily targeted at "content owners" and advertisers

CEO Hurley compares YouTube to TV in 1941
"a new technology that has the ability to entertain and engage people on a massive scale" (9)
Where is the "broadcast yourself" empowerment rhetoric? Absent. Whose platform is this?

"YouTube and others can make a bid to be the new television, convincing media producers to provide their valuable content and advertisers to buy valuable consumer attention, on the back of user-generated content and all its democratic, egalitarian connotations, offered to them as television’s antidote." (9)

The stakes of this fuzzy use of "platform" raise when the term migrates out of marketing pitches and into policy-making (9)

Platform is a "structural/spatial metaphor" similar to common carrier, conduit (11)

Making a "semantic claim" (11)

On technical affordances

"these intermediaries provide have distinct technical affordances, designed to serve their particular clients and purposes" (12)

On algorithmic moderation

"site indexes that purport to represent user judgments [Most Viewed, Top Favorited, etc.] will in fact do so only within parameters unknown to users" (12)
Tools available only to a limited number of highly-capitalized stakeholders, "content owners", enabled copyright takedowns on a massive scale - Content ID (13)

"it is YouTube’s complex economic allegiances that compels it to both play host to amateur video culture and provide content owners the tools to criminalize it" (13)

"Despite the promises made, ‘platforms’ are more like traditional media than they care to admit" (13)

Bogost & Montfort, Platform studies: Frequently Questioned Answers

"'Platform studies' is a new focus for the study of digital media, a set of approaches which investigate the underlying computer systems that support creative work" (2).

This article is an effort to clarify 6 misconceptions about the platform studies approach

Platform studies entails technological determinism
Platform studies is all about hardware
Platform studies is all about video games
Everything these days is a platform
Platform studies is about technical details, not culture
Platform studies means that everyone in digital media will have to get computer science training or leave the field

Some initial notes of distinction:

"all computing platforms on which interesting creative work has been done" (2)
"computational platforms, the basis for digital media work" (2)

Platform studies entails technological determinism

"Hard" v. "soft" determinism a la Thomas J. Misa
"Media ecology" a la McLuhan
Cultural studies often ignores technology in favor of human expression/ideas
STS and "social construction of technology" (SCOT) are better but do not often deal with "how people respond to them after [they come into being]" (3)
- Cite Langdon Winner "Upon opening the black box and finding it empty"
"Technological determinism has become [...] a stock answer anytime the lid comes off the box" (3)

"we agree with [objections to hard determinism] because they actually support a platform studies approach: people make negotiations with technologies as they develop cultural ideas and artifacts, and people themselves create technologies in response to myriad social, cultural, material, and historical issues. If we were to believe that technology manifests itself and unidirectionally influences the course of human history, we would be cutting off at least half of platform studies: The study of how our technologies, our computer platforms, embody particular cultural concepts and ideals, how they too are created in a cultural context" (3).

Platform studies is all about hardware

"A platform is a computing system of any sort upon which further computing development can be done. It can be implemented entirely in hardware, entirely in software (which runs on any of several hardware platforms), or in some combination of the two" (3)

Examples

Atari VCS is entirely hardware
the Java Virtual Machine is entirely software

"Platforms contain other platforms" (4)

Publishers of early books of BASIC code "were treating BASIC as if it were a platform - even though it wasn't even really a single language, but a family of a related languages, implemented differently" (4)

Platform studies might consider BASIC and other software systems as platforms despite the differences

Platform studies is all about video games

VG consoles are easily seen as videogames

And the first platform studies book was about Atari VCS

But there are all kinds of platforms.

Everything these days is a platform

"Platform" is used widely but the "platform studies" series is focused on

"the study of computational or computing systems that allow develoeprs to work creatively on them" (4)

Relying on a quote from Marc Andreessen complaining about the dilution of the computational meaning of "platform"

"a 'platform' is a system that can be reprogrammed and therefore customized by outside developers" (as quoted on 4)
"flexibility of platforms provides them with creative potential" (4)

Andreessen

Their seem to mis-read Gillespie here

"if he reads the computational sense of 'platform' as outdated..." (5)
He doesn't. He's pretty clear that he is addressing the rhetoric of corporations like Google/YouTube and its reflection in journalism, criticism, and popular debate (5)

Are they concerned that Gillespie's serious consideration of the fuzzy use of "platform" will undermine their project?

"Some systems may be communication platforms or simply very large-scale computing systems, but may not be best understood as computing or new media platforms because they do not mainly support the development of general digital artifacts, including computer programs" (5).

Further quotation of Andreessen

Level 1, the "Access API", e.g. Flickr, etc.
Level 2, "Plug-In API", Photoshop, Firefox; on the web Facebook
Level 3, "runtime environment", "developers upload code and run it directly on the platform" (5)
- They offer Ning as an example in the "online world" (5)

"There are many ways to slice platforms, but certainly, the ones that are most likely to be culturally important are those that are most accessible to people, that have interesting capabilities, and that specifically welcome developers" (5)

Is a particular system "influential and important as a platform"? (5)

How "culturally interesting" is the use of it as a platform by developers?

Advocating studying "computational platforms" because they are the "overlooked basis of a half-century of computational, digital work" (5)

"Communication systems ... have been studied for a long while" (5)

Platform studies is about technical details, not culture

"[Techno-fetishism] fails to ask either practical or critical questions about technology" (5)

Extreme positions:

"Technically adept gadgeteer" (5)
"Technically ignorant critic" (5)

"Platform studies is about the connection between technical specifics and culture" (5)

From one end, particular graphics mode made certain games possible and made that making appeal to devs
From the other, how social, economic, cultural, and other factors led platform designers to put together systems in particular ways (6)

"Form and function, code, and platform, are fully embedded in culture" (6)

Platform studies means that everyone in digital media will have to get computer science training or leave the field

"The new media scholar is aiming to understand technologies well enough to connect them to culture, but not to invent new algorithms..." (6)

Winner suggests that SCOT people were hesitant to do platform stuff because new methods would challenge those using dominant methods (6)

How widespread is this feeling?
Is it a factor in comm?

"Our concept"

"Platform studies investigates the relationships between the hardware and software design of computing systems (platforms) and the creative works produced on those systems, which include but are not limited to video games - digital art, electronic literature, recreational and playful programs, and virtual environments are all built upon platforms, too" (6)

Emphasis on "creative production" (6)

Star, "The ethnography of infrastructure"

Methodological paper about studying infrastructures with ethnography

Infrastructure can be "mundane to the point of boredom" - easily overlooked (as in B&M's description of 'platform')
Difficulties of dealing with very large data sets (e.g. transaction logs)
Understanding interplay of online/offline behavior

"This article is ... a call to study boring things ... It takes some digging to unearth the dramas inherent in system design creating, to restore narrative to what appears to be dead lists" (377)

Numbers, technical specs, hidden mechanisms
"Indirect readings of such dry documents can also be instructive" (378)
Looking for historical changes (in code, for example, looking at versioning - perhaps in sourceforge, github, or other online repositories)

Very similar call as B&M to attend to "overlooked" aspects of digital media:

"It is easy to stay within the traditional purview of field studies: talk, community, identity, and group processes, as now mediated by information technology" (378)
"There are much fewer on the effect of standardization or formal classification on group formation, the design of networks and their import for various communities, or on the fierce policy debates about domain names, exchange protocols, or languages" (378)

Settings and code are not typical anthropological sites, they represent "embedded strangeness" (379)

Forgotten, background, frozen in place

"We need new methods to understand this imbrication of infrastructure and human organization" (379)

"Infrastructure ... is by definition invisible, part of the background for other kinds of work. It is ready-to-hand" (380)

But infrastructure is not equally 'background' for all, e.g. a person in a wheelchair does not consider stairways a "seamless subtender of use" but a "barrier" (380)

Infrastructure is a relational property, always in the context of some activity (380)

Water system is infrastructure to the cook preparing dinner
Water system is target for repair to a plumber
Water system is a variable in a complex urban system for a city planner

"Infrastructural inversion"

"foregrounding the truly backstage elements of work practice" (380)

Properties of infrastructure

From:

Star, Susan Leigh and Ruhleder, Karen (1996) 'Steps toward an ecology of infrastructure: design and access for large information spaces', Information Systems Research, 7: 111-34.

Embeddedness

"Infrastructure is sunk into and inside of other structures, social arrangements, and technologies" (381)

People might not distinguish different pieces or aspects of infrastructure
e.g. Server, software, template, skin, etc.

Transparency

"Transparent to use" (381)

Doesn't have to be re-invented for each task
Transparency is not equal for all participants, what is easy/transparent for one is not necessarily true for all

Reach or scope

"Spatial or temporal" (381)

Infrastructure has reach beyond a single event or one-site practice" (381)

Learned as part of membership

Artifacts, org arrangements are taken for granted by communities of practice (381)

Becoming a member is to naturalize the infrastructure

On a messageboard this includes use of search function, knowing where to post about certain topics, referring to stickies, embedding images, sending PMs, etc.

Links with conventions of practice

Infrastructure shapes and is shaped by existing conventions (381)

Even if this isn't ideal from a designer's point of view

Embodiment of standards

Standards enable infrastructures to transparently interlink (381)

Built on an installed base

Struggling with and inheriting the limitations of earlier infrastructures (382)

e.g. backward compatibility
Related to conventions of practice but more about actual pre-existing infrastructures

Becomes visible upon breakdown

Invisibility and transparency becomes visible when it breaks (382)

Backup and precautionary procedures make infrastructure visible

Is fixed in modular increments, not all at once or globally

Change takes time (382)

Happens at different rates locally
Systems extend across spaces
e.g. different browsers, etc.

Infrastructure and methods

Sites:

Sites include "decisions about encoding and standardization, tinkering and tailoring activities, and the observation and deconstruction of decisions carried into infrastructural forms" (382)

Techniques:

Historical, literary analysis
Interviews, observations, systems analysis, usability studies

"Ethnographic sensibility":

"an idea that people make meanings based on their circumstances, and that these meanings would be inscribed into their judgements about the built information environment" (383)

Scale problems

Traditional ethnographic model no longer fits when there are geographically dispersed groups of hundreds of people and computer terminals (383)
Aggravating an existing limit on the scale of ethnography
Trying to scale up traditional methods: "exhausting" for the researcher, time suck (384)

"Tricks of the trade"

Tricks for "reading" infrastructure

"unfreezing some of its features" (384)

Identifying master narratives and 'others'

Many information systems employ a "master narrative", speak with a "single voice" (384)

e.g. "flesh colored" (385)

Tasks: identify that which has been made "other", or unnamed (385)

Pulling diverse activities/interests into a monolith ("the US believes...")
Personification ("science seeks...")
Passive voice ("data shows that...")
Deletion of modalities (context is erased and a fact is made into truth)

Example: medical forms

Heteropraxial: differently used according to local specifics
Heteroglossial: different voices inscribed into a seemingly monotonous form

Surfacing invisible work

Look for processes of encoding + embedding work in the traces left behind by coders, designers, and users (385)

May have to go "backstage" to data storage, ontology (385)

In system design, there is occasionally a tension between making things visible or leaving them tacit

Paradoxes of infrastructure

"Why do people persist in using less functional, but more routine actions when cheaper alternatives are nearby?" (386)

Need to see encounters with infrastructure in the context of a larger workflow

Small changes might cause ripples that disrupt non-infrastructure tasks

The thorny problem of indicators

Three ways to read information infrastructure

A material artifact constructed by people, truth status of info presented is not relevant, only impact
A trace or record of activities. Information (and truth status) are more relevant. Logs, emails, classification/categorization, ontologies - somewhere between cultural artifact and semi-automated research assistant
A veridical representation of the world - info system is a "mirror of actions in the world"
- For example, substituting the posts on a messageboard for taking fieldnotes (388)

One tactic to be cautious of:

Substitution of precision for validity (388)
e.g. Focusing on indicators, categories over "big questions" (388)

Bridges and barriers

Citing Winner (as did B&M)

"whether and how values are inscribed into technical systems" (388)

"Applying the insights, methods, and perspectives of ethnography to this class of issues is a terrifying and delightful challenge for what some would call the information age" (389)

Star, Bowker - "How to Infrastructure"

"the infrastructure that subtends new media: overwhelmingly for our purposes this means the Internet in all its guises" (151)

Infrastructure in commonsense way "runs 'underneath' actual structures" (151)

"that upon which something else rides, or works, a platform of sorts" (151)
Common sense definition crumbles when we see "multiple, overlapping and perhaps contradictory infrastructural arrangements" (151)
- And add people to the mix
- One person's infrastructure (railroad) is another's structure (railroad engineer)

Three themes of this paper:

"the moving target of infrastructure, from easy-to-use black box to active topic of work and research";
"the breakdown of infrastructure that opens the taken-for-granted";
"the relative usefulness of infrastructure for different populations" (151).

"Infrastructure is not absolute, but relative to working conditions. It never stands apart from the people who design, maintain, and use it" (151)

Designers try to make it invisible
But repair, maintenance needs will return it to visibility

"Infrastructure is an often neglected aspect of communication studies" (151)

Again - similar to the call of Bogost + Montfort
Exceptions: policy, regulation, history of specific media technologies

"Can [infrastructure] be anything?" (152)

Another similar concern to B&M

"A historical, archaeological approarch to the development of an infrastructure ... needs complementary sociological, regulatory, and technical studies" (152)

Sociohistorical analyses of infrastructure

By the 1990s, STS researchers turned their attention to computers and other IT (152)

Woolgar called this the "technical turn" in STS

Hughes on "reverse salients"

Tech, social, political sticking points which can slow the development of an infrastructure (153)
Solution can be of different register than the problem
- e.g. social solution to technical problem

Paul David on "productivity paradox"

New technologies depress productivity until they are naturalized and workflows accommodate, adapt them
e.g. "Computer is a bad typewriter" (153)

Bowker on "infrastructural inversion"

"Historical changes frequently ascribed to some spectacular product of an age are frequently more a feature of an infrastructure permitting the development of that product" (153)

Studying infrastructure problematizes the relationship between background and foreground (153)

How infrastructure happens

Essential to the development of working infrastructure:

Standardization, including communication protocols
- Some automated (MIME), others bureaucratic: journal submission
- Not only bytes being standardized but work habits, flows (154)
Classification

Standardization

With many possible standards in circulation, no guarantee that the "best" set of standards prevails

"In an infrastructure you are never alone: no node is an island" (155)
Positive network externalities

Interchangeable infrastructures (155)

Physically connecting 2 machines
Connecting 2 machines over a network
Carrying a disk between two machines

Two extremes of standardization:

Colonial: Government fiat, monopoly enforcing a single standard (Windows)
Democratic: Interoperable APIs, gateways, protocols (HTTP, TCP/IP)

"The internet is only virtually stable - through the mediation of a set of relatively stable protocols: for this reason, Edwards (1998) suggests that we call it an internetwork technology rather than a Hughesian (Hughes, 1983) network technology" (156)

"Empirical studies have show time and again that people will not see it as a good use of their time to preserve information about their data beyond what is necessary to guarantee its immediate usefulness" (157)

Classification

As in the development and maintenance of a database

Access and usability issues

Relational aspects of infrastructure most apparent in barriers to access (158)

On class, training/education, ability, hardware, etc.

Conclusion: Design implications

Reference Stewart Brand, "How buildings learn: What happens after they are built" (159)

Looking only at an architect, you miss the modifications/uses that emerge over time
Compare with infrastructure (159)

Good information infrastructures should be:

Stable enough to allow information to persist over time
But modifiable - at both individual and social levels to serve new/changing/emergent needs (159)

Re: FOSS, spec ESRs assertion that "complex, interoperating systems ... can be built in a distributed fashion, providing you have a good set of protocols in place for receiving and vetting changes" (159)

similar to Scandinavian "participatory design"

Metadata ('data about data') Michener et al 1997 Rayward, 1975 - Otlet on the universal library - why isn't this part of the Nelson / Bush / Engelbart myth? Abbott 1988, on microfiche Yates 1989 orlikowski 1993 hurley keller 1999 Bud frierman 94 clark fuijmura fischer 92 friedlander 95 rosenberg 82

Off topic

Stewardship

Ideally, big business wants to have the cake + eat it too
Be a platform for public discourse but not be responsible for protecting it

"Patronage" (Burgess & Green, 2008)

'patron' of collectivity creativity