The Section for Algorithmic Cultures
specialized term for archivists delimiting valuable records. After
the computer revolution "the file" became a key term in the
architecture of the machine. With the internet it became all the
things we share, so to speak. In the old days, the file was an
archival record. It designated a group of documents related by
use or topic, typically housed in a folder. A computer file is much
more than a record. It contains various layers of data including
instructions on how to connect to other files actively within a
network. It has become an artificial life form or part of what the
industry calls artificial intelligence.
Let us take it step by step: A computer file contains data. Indeed,
it could resemble an old fashion file; it could be an image or
a word document, the kind archivists are used to. However, a
computer-file, for example an image (a TIFF-file), contains
metadata of a magnitude never addressed by a normal archive.
The media lab group Constant in Brussels, have tried to activate
these digital metadata in an interactive archive related to
Erkki Kurenniemi's life archive, and the possibilities turned out to
be legio, as I show in "FileLife". Furthermore, a computer file
does not need to contain a document or an image; it can also be
a program, a set of codes. A file can in other words store coded
instructions for future tasks and operations. Software applications
or programs may arrive in file formats. The file which used to
be an expression of the actions of a governmental agency has
become one of the core elements of the action and command
structure of computation. Indeed, the file has become the
archive within the archive, or even, the archive of archives.
How should archives relate to this?
Any archive with a digital holding contains millions of files with
images and documents, and it can do this without any regard for
algorithms and programs (as long as you are not part of the
technical staff). However, there are special digital archives out
there, on the internet, some of them called GitHubs (perhaps
31 million repositories around the world), that specialize in taking
care of algorithms and codes. Many of these are developed by
hackers. They love to share codes and algorithms so they can
build things and solve problems. The question is to what extent
should "normal" archives, state archives, city archives, National
Libraries and National Museums, also take care of codes and
algorithms?
Archives and National Libraries should look into establishing
a Section for Algorithmic Cultures which could overlook the
institutions' handling of digitization, born digital material,
websites, video games, software and possibly influential
algorithms as well as investigations issues of cultural heritage
as big data and questions concerning Digital Humanities. In their
new book Big Data: A Revolution that will Transform How We
Live, Work, and Think (2015), Victor Mayer-Schonberger and
Kenneth Cukier talks about the need for a new profession:
the algorithmists. They write: "They would evaluate the selection
of data sources, the choice of analytical and predictive tools,
including algorithms and models, and the interpretation of
results. In the event of a dispute, they would have access to the
algorithms, statistical approaches, and datasets that produced a
given decision".
The mission of state archives is of special interest in this case,
because one of their key tasks is to enable historians to track the
record or working process of a governmental agency, even
Publiek reageert op Eivind Rossaak
tijdens KVAN lezing in Den Haag, 11 november
Foto: Angeline Swinkels, Den Bosch