Records in a digital environment: from object to hyperobject
When we think about authenticity and authentication it is inevitable that questions
must be addressed on what is to be authenticated. In this essay we focus on digital
records. In this paragraph I will describe my position. That position is built on a line
of reasoning that is derived from a probably unlikely combination of the work of
scholars from different disciplines. At the end of the paragraph I will introduce the
idea of digital records being a part of a hyperobject containing digital information.
What is a record?
There is a huge amount of definitions of record, and the definitions have shifted over
time. Geoffrey Yeo (2008) has written:
it now seems appropriate to characterize records as persistent
representations of activities or other occurrents, created by participants or
observers of those occurrents or by their proxies; or sets of such
representations representing particular occurrents" (p. 136).
Yeo's definition is used here because it does not give any restrictions on the form in
which the representation takes place, on who authenticates the record, on who
decides to create and manage the record and on whether the record should serve as
evidence or should serve other purposes. Furthermore, the definition is inclusive
regarding both strong and weak authenticity.
The definition gives a clear framework for understanding records. However, it does
not include a statement on what the representation consists of. In my opinion the
representation consists of information, as understood by Luciano Floridi. The
General Definition of Information states that an instance of information consists of
well formed, meaningful data, "well formed" meaning that the data are "rightly put
together according to the rules (syntax) that govern the chosen system" (Floridi,
2010, p. 20-21). This definition is very appropriate for records, since it does not
impose any limitations on the form of the representation as well. It is applicable for
medieval charters, for oral records from indigenous communities and for relational
databases.
In my opinion this addition should be made to Yeo's definition of a record as only
then it is possible to make general statements on questions like: when is a record, and
where is a record?
When is a record?
When we reason according to the definition of records given above, we might imply
that records are the result of data processing. The persistent representation that
results out of this data processing can take on any form as long as the result is
acceptable for the user. This user can be a human being, a machine or a network
consisting of connected machines and human beings. In this way it becomes hard to
talk about a record in the traditional sense.
Records, being the result of processing, can be compared to installation art. They
consist of elements (data) that together express what the creator intended to
express. For example, take a video installation from the 80-ies of the last century,
comprising a VHS-videorecorder, an old TV and some fluorescent tubes. The parts
of this installation will break down relatively quickly, especially when they should
function every day during the opening hours of a museum. The TV will break down,
the fluorescent tubes will fail and of course the VHS-recorder will come to its end
after some years. So, to maintain the authenticity, parts will have to be replaced.
And then the curator will have to have luck on his side: maybe the fluorescent tubes
are not in stock anymore, VHS recorders have disappeared from the earth and of
course the TV is so unique that it should deserve its own place in a technology
museum. So, after some years what will be left of the authenticity of the installation?
It seems that the authenticity of the concept can only be maintained by
undermining the authenticity of the original material. Maybe the only good
measure is to document and to account for all the preservation actions the curator
has undertaken. The curator can never "freeze" the installation in such a way that it
will always stay the same. Van Saaze therefore considers authenticity the result of
activities concerned with conserving the material and by documenting these
activities and to give a framework to assess the quality of these activities. (Van Saaze,
2013, p. 80-83)
Just like installation art, digital records can never be preserved in such a way that
they "always stays the same". They would vanish within a few years. An archivist
should make sure that the information will be available and accessible for as many
years as possible. The data must be migrated and converted. Data must be added.
Formats must be substituted. All these processes must be documented. To keep the
record authentic, it might even be necessary to violate the authenticity of its
components: the data.
A record could therefore be considered as a dynamic result of dataprocessing.
Acker has written:
"I would argue that the project of defining a record must be abandoned in
the age of networked record in the digital age the ontological purity
and drive for the true nature of the record is over. It cannot be exclusively or
definitely located because it is not in a single place, it never ends, and is
'always in a process of becoming' in the continuum" (Acker, 2016, p. 316).
However, Acker's position is debatable: whereas processing and re-processing records
can be described as "a process of becoming", the digital or analog record, once
created and stored, should stay the same in its persistent representation of activities.
Where is a record?
If records are the results of data processing, is it still possible to speak about records
as being objects that can be located? It is possible in the classic situation where
records are fixed on information carriers, stored in physical repositories. However, it
is obvious that it is very difficult to locate digital born records in this way.
When records are considered the result of data processing, it might be argued that a
digital informational object does not exist at all, at least in the way physical objects
are defined within the classic view on paper records. Of course, digital data are
always stored somewhere. In that sense digital information is as material as analog
information. However, the place of storage of all the components that make up for
digital information (data and software) can vary endlessly.
archives in liquid times
252
frans smit records, hyperobjects and authenticity
253