[-empyre-] Culturally specific archives

[Craig Dietrich]

Hi Jon, Vanina, et al,,

It's been great following the discussion and thank you for mentioning
my work with culturally specific archives.  A relevant project is the
Mukurtu Archive (http://www.mukurtuarchive.org), a content management
system that bases its access to material on embedded Aboriginal
cultural protocols.

Jon mentioned the goal of respecting local difference in archives and
standards.  A big part of Mukurtu project development has been keeping
flexible the means by which system administrators can define cultural
protocols including family, country, a scared status.   It's been a
challenge to model the software's database and logic in such a way
that overlapping cultural protocols can be managed; attributed to
individuals, groups, and media; and hopefully soon, shared between
archives to preserve protocols as content is exchanged.  In practice
we've found that the task is greater than defining standard fields
(for example adding "family name" to compliment Dublic Core or MARC
fields).  Instead, we need a way to represent the stories that are
present with each piece of content, and how stories reveal content to
viewers based on their profile.

As described by Warumungu community members in Southwest Australia, a
piece of content kept in a safe keeping place shouldn't open up
(become available) unless certain family, country, gender or sacred
affiliations are present in the viewers.  From a Digital Rights
Management (DRM) perspective, a user attempting to view a file can't
gain access unless they can prove certain digital requirements.
However, for implementing in the Mukurtu Archive, we see a
metaphorical difference.  As project director Kim Christen describes,
many Aboriginal communities we've worked with aren't concerned with
'restricting' files on an individual bases, instead wishing to
'reveal' to the proper audience a collection of files based on one's
user profile.  The software is therefor less concerned with
heavy-handed DRM that keep people out and more concerned with creating
a safe place for users to browse --  where content reveals itself --
to facilitate people coming in.

Describing protocols that reveal rather than restrict might seem
technologically trivial, but it's a difference between implementing
DRM (which puts all sharing power into the hands of the publisher) and
folksonomy (which empowers individuals to accept, reset, or add files
and sharing protocols).  Much like other folksonomy archives, in the
Mukurtu Archive each file is "tagged" with categories, metadata, and
protocols.  But writing the code that manages the display of content
based on the this information became more hierarchical than
folksonomic.  As an example, below are a set of comments in our PHP
code from the first version of the Mukurtu Archive (from 2007) that
demonstrates how it determines which files can be displayed to the
user,

//  add public items that are described by categories in the user's profile
…
//  add non-public items that are placed in categories that are in the
user's profile
…
//   add public items who's categories match the user's based on root categories
…
//  gender and sacred status
…
//  other user profile matches
…
//  family and country affiliation
…

As you can see from the comments, the code is specific.  The system is
therefore dependent on the code to transform data into meaningful
protocol-aware content.  This isn't uncommon in software, but to share
or preserve elsewhere data in this version of the archive would
require the code's logic to be represented when passed to the receiver
system.

We're working with new strategies to create a flexible system that can
more easily interact with other systems and archives.  I'm
particularly interested in semantic web technologies.  For example,
using RDF-XML (a core format of semantic web systems), one can store
metdata such as Dublic Core fields, and attach one or more classes to
each data node. Then, using a RDF schemas one can define properties of
each class that can in turn be attached to user interface elements or
mapped to ontologies that describe sharing protocols.  However, the
difficulties of using semantic technology in the field muddy its
promise.   Though, perhaps certain aspects can be gleaned over others.
 Ignoring inference (the predictive algorithms that can recommend you
buy a book based on your previous purchases) the structure of RDF and
its corresponding ontologies could be a useful standard format for
storing data in a way that safely maintains local nuances.

Thanks, and looking forward to the conversations ahead,
Craig

--
Craig Dietrich
Institute for Multimedia Literacy
School of Cinematic Arts
University of Southern California

On Sat, Oct 2, 2010 at 10:02 AM, Jon Ippolito <jon.ippolito at gmail.com> wrote:
> Hi Vanina  et al.,
> On Sep 29, 2010, at 7:49 AM, Vanina Hofman wrote:
>
> -Taking into account the scarce resources of media art conservation in
> literature, we have decided to make Taxonomedia in Spanish. We thought that
> producing and translating information in this language can be itself a
> contribution to the topic....We decided to focus in managing cultural
> activities centered around the Latin America context.
>
> It was great to read the blog post about Forging the Future in Spanish. I
> think well-meaning standards bodies have wasted too much time trying to
> pound differently shaped pegs into the same square holes, instead of
> devising software that respects local differences.
>
> Craig Dietrich of Still Water and USC has done some extraordinary work on
> culturally specific archives, so I'm hoping he'll chime into this
> discussion.
>
> I'm also curious what Mona learned in Ghana about the preservation practices
> of oral cultures. I believe re-telling and re-performance is a better
> paradigm for preservation in the 21st century than the storage paradigm that
> came of age in the 20th.
> Finally, I sent this query earlier but am not sure it made it to the list
> due to the vicissitudes of my listserve membership:
> On Sep 30, 2010, at 8:53 AM, FILE_Arquivo wrote:
>
> This made way to organize this amount of information; it’s facing the
> instabilities, errors and ephemeralilties as inherent part of the complex
> electronic/digital art archive ambiances....This is a philosophical point of
> view, which we are trying to put in practice, working hard on interface
> design and in the database structure.
>
> Intriguing, Gabriela. Can you give us any more of a glimpse--via a prototype
> or just textual description--of how this ephemerality-friendly interface and
> database might work?
> jon
> ______________________________
> Forging the Future:
> New tools for variable media preservation
> http://forging-the-future.net/