Paper presented at a conference of the Korean
Association of Archives Management in Seoul, 3-4 July 2002
Much has been written and discussed about the
preservation of digital information, objects or records. There is a plethora of
projects and initiatives in this area trying to find solutions for the problem
that digital information is volatile, intangible and mutable. Existing
preservation methods as for paper do not longer suffice and new and innovative
approaches and methods have to be developed. It is experienced not only in
traditional memory organisations as archival institutions, libraries, and museums,
but also in broadcasting companies, institutions with audio-visual collections,
entertainment industry, business companies such as pharmaceutical or
petrochemical industry, research institutions with huge amounts of scientific
data and so on. Everybody that produces digital information and has an interest
in preserving it for a certain period has to face the issue how this can be
achieved and that becomes more difficult the longer this information has to be
preserved. In other words the issue of digital preservation is not limited to
one organisation, or a certain domain like memory organisations, or a country,
it is really a global issue that everybody in this field is trying to deal
with.
Although there exist many initiatives in this
area, real and enduring solutions are not yet available. There are some
interesting strategies being developed to date, which may provide a solution
for the short and perhaps mid-term. These strategies may include
developing standards for storage
formats or preservation metadata, legislation, prescribing standards, or
developing reference models for preservation functions or repositories, or
defining minimum sets of metadata that allow the ongoing maintenance and access
of these information resources.
The natural ally of preservation is
accessibility, the one cannot without the other. Therefore any requirements for
preservation have to be based on the requirements of accessibility and use of
digital information. It is this whole, complicated area, which is being addressed
in this paper. The objective is to provide an overview of the main issues and
requirements, followed by the main developments to date in the search for
solutions with a brief description of the achievements and the available
methods and tools. Additionally I will try to identify some initial steps that
could be taken in order to implement proper management of digital records and
related preservation strategies, starting with record creating agencies.
Finally the main focus will be on electronic records, although I will take
sometimes a broader view.
The requirements for preserving digital
records (or other information resources) are derived as indicated from the use
organisations or people want to make of them. Well known reasons are the business
process, accountability, research, law suits, and they include also cultural
heritage. In order to enable these purposes and activities information
resources need to be retrievable, authentic, reliable, understandable and
usable. I will explain these requirements briefly.
Accessibility can be seen as one part of what
can be called a binovular twin. The counterpart is preservation or longevity.
Without accessibility preservation of information does not make sense and
similarly the other way around without adequate preservation of information
there will be no accessibility.
Accessibility as such is no new requirement,
but in the case of digital information it can be considered a real challenge.
With paper records one can always walk to a shelf to search physically through
the records, digital records can be considered as lost if there is no
consistent naming convention for documents or better files, or if search
applications as windows explorer cannot be used. Digital information is invisible
and intangible even if it is stored somewhere on a hard disk.
Accessibility requires arrangement and
description of the records. Records can or better should be structured and
organised for instance by applying a classificationscheme. This will not only
contribute to the accessibility, but also establishes coherence or more
correctly expresses the interrelationships between the documents that are
created and received in carrying out a certain business process (filing of the
records).
At closer look one could say that
accessibility consists of a sequence of three sub-requirements:
1.
Retrievability. Records have to
be retrievable. Instruments to enable that are making them identifyable,
arranging or structuring them, and describing them (in a digital environment
mostly metadata). The arrangement and description methods that are chosen for
records can be articulated in a structuring scheme, e.g. a business
classification scheme. Search results based on these descriptions and
structures should be presented to and can subsequently be interpreted by the
user, who can make then choices for one or more records or documents.
2.
Readability. As soon as digital records are retrieved and
identified as relevant they have to be reproduced and presented on the screen. This supposes that the appropriate software
is available to get the file, to interpret the encoded data and to reproduce
and present the record or the information in the correct (=authentic) form and
structure.
3.
Interpretability. If both
previous steps have been properly achieved, then the presented record (or
document) has to be interpretable. That means the record has to be accompanied
with the necessary and explanatory metadata about provenance and documentary
context. The user should know why and how the document once has been created,
for what purposes it has been used and what the interrrelationships with other
documents are, even if this is not always necessary for answering his or her
specific question. With this information or metadata the user can assess the
value of the document and its content in relation to the research or use.
One of the big issues around digital
documents or records is authenticity. Apparently it seems to be a new
phenomenon that causes quite a lot of discussion. It is however not a real new
thing, because it was already implictly present with paper records. Paper
records are physical entities and it is mostly easy to recognise whether it is
an original document or authenticated copy. The reason why it has become an
issue now is that digital document or records by their very nature are
intangible and volatile, and it is therefore not easy to apply the requirement
for authenticity. So what does authenticity mean in a digital environment?
A lot of research has been done and described
by Luciana Duranti about the notion of authenticity in several projects as for
instance the Inter Pares project, using and going back to the old discipline of
diplomatics.[2] It does not
fit into the framework of my presentation to dive into this subject here, which
is worth a conference on its own. I will only mention briefly some of the most
relevant aspects and issues. Authenticity refers to our requirement to be able
to retrace documents or records to their creation (or origin), so we are able
to identify why, when, where, by whom and so on, they were created and used. We
need the answers to these questions not only to know the identity of a record,
but also to know whether the presented information is trustworthy or reliable.
In other words we need to be able to position a record in the time and context
from which it purports to originate. After all, if we know somebody or
something (or think we know) we are able to establish whether we can trust him,
her or it, or not. In order to enable this judgement or assessment we need
information that can answer those questions. That information might not always
be used in reality, but it should be there one way or another, so we can fall
back on it. Mostly, however, people just trust the information or records as
presented without going through the procedure of checking it. This also depends
on what organisation or individual is supplying the records. A trusted party as
an archival institution is mostly considered more reliable than for instance an
arbitrary person providing information on the Web.
In this respect it will be necessary to have
not only information about the context of creation (e.g. about organisation or
individual, activity or function, time, place, reason), but also information about
the relationship with other documents (e.g. in a case file), and information
about the management of the records from their capture until the moment of
(any) presentation. Evidence should be available to answer questions as e.g.
how has the record been maintained, or has the document been tampered with?
This contributes also to whether we can trust the presented record or not.
Apart from this reason the digital nature of
records requires that information has to be captured about structure and form
and sometimes even about the behaviour of a document or record at the moment it
was used in a business activity. The level of detail of this description will
differ for documents in different business processes, since the authenticity
requirements will differ in each of these business processes. More explicitly
stated, authenticity requirements are dependent on the nature of the business
process and the context in which it is carried out. Criteria are for instance
legislative, regulatory, business and juridical aspects. They determine how
strict authenticity has to be interpreted. Those that are able to identify best
what the essential characteristics of certain types of documents in a specific
business proces are, are the people that are responsible for and carrying out
that business process.[3]
An instrument to describe the form and
structure can be XML (eXtensible Markup
Language). The description of the structure can be done by using a document
type definition (DTD) or schema, that subsequently also can be used to check
whether the created documents comply with the intended structure or not. It is
obvious and more cost-effective that a DTD or schema should be made and
implemented before creation of documents or records. With the eXtensible Style Language (XSL) the form
of a record can be described (in a style sheet). In some recent versions of
word processing applications these tools are already implemented. Several
projects on digital preservation are using this approach, e.g. the NARA
project, and the Dutch testbedproject.
Basically the sequence of activities in the
process of searching and using information consists of retrieving,
interpreting, trusting and using it.
One aspect is essential in relation to long term
preservation of digital records, and that is the distinction that has to be
made between digital records (the conceptual aspect) and the computer files
(the technical aspect). I will try to clarify what is meant with this
distinction.
With paper records it is obvious that the
document and the carrier are one. They form one physical entity. Form and
structure are physically fixed by paper (or parchment) and do not have to be
described as such. With respect to digital records it is very much different. These
records are no longer physical entities. Information technology has fragmented
them into different components, e.g. software, datafile, medium and hardware.
The consequence is that we have to make a distinction between an intellectual
or conceptual and a technical presentation. That makes it, more than in a paper
world, necessary to identify and describe what needs to be preserved.
At the one hand there is the representation
on the computer screen. That is the record that human beings can read. On the
other hand there is the physical representation, being one or more
computerfiles with encoded data that are stored on a medium (hard disk or
CD-ROM or tape). We are dealing here with different things that both need to be
managed in order to preserve and to reproduce digital records or other objects.
Sometimes there might be a one to one relationship between a document or record
and a computerfile, but mostly more than one file will be necessary to
reproduce the conceptual record. Or the other way around sometimes one
computerfile may even contain data of one or more records, so the relation can
be 1:n, n:1 or even n:m. Especially in the case of multimedia and more complex
digital documents or objects the latter will be the case. To give a simple,
perhaps a little bit technical example, in presenting a MS Word document the
wordprocessing application uses not only a file with bytes that contain the
text characters in the document, but also a so-called dll-file (dynamic link
libraries-file) that is used by the system software (Windows) to present the
characters in the font or fonts used by the writer. Though these dll-files are
not part of the user created document, they are necessary to reproduce and
represent the document properly and in the way the writer intended it. This
goes the same with other types of content such as for instance graphics.
So each time a digital record has to be
presented on the screen, it has to be reproduced or reconstituted by software
using the digital components as stored on the storage medium. As soon as a
document on the screen is ‘closed’ or saved, it will be stored as encoded data
that are only interpretable by the (application) software. In other words that
what is stored on the storage medium is not the same as what is presented on
the screen. To know whether the document that is presented on the screen is
similar to the document that was once received or created in a business
process, we need a description of this (conceptual) document. The components
that have to be described are content, context, form, structure and in some
cases behaviour (e.g. a spreadsheet that contains formulas). To what extent
depends on the authenticity requirements.
The awareness of the distinction between
these conceptual and technical aspects of digital records is essential to
manage them (see diagram). In the Netherlands a Ministerial Regulation has
recently been issued that addresses both and establishes requirements and
standards for archival records to be implemented in procedures, systems, and software applications used to create, manage
and preserve records in government organisations. These applications could be
dedicated records management software applications, but also records management
functionality integrated in business applications.
Figure
1 One strategy for
two different objects
In the previous sections the main requirements
and fundamental issues have been identified. The next step is to develop
frameworks, strategies, infrastructures, and methods that enable to deal with
them, so proper records management over time can be implemented and achieved.
To summarise briefly the issues posed by digital information:
-
the increase of control necessary to ensure authenticity and integrity
of the digital records
-
the awareness of dealing with two types of objects: the intellectual or
conceptual objects (=records) and the technical objects (=computerfiles)
-
the functionality or processes necessary to achieve that the (archival)
requirements are implemented properly
-
the organisation of metadata structures required to enable management
and preservation, and to ensure accessibility, interpretability, reliability,
integrity and authenticity of the records. One of the major issues in this
respect is interoperability.
-
although out of scope of this paper I could add also the development of
user interfaces that enable searching, presenting, and using records and other
kinds of digital information. Searching should be enabled across organisations
and across domains.
The required strategies and methods have not
to be invented by organisations from scratch though, because there are already several
strategies, methods and tools available that could help to build a safe and
proper environment for preserving and managing digital records.
I will list the following initiatives and
projects in this respect and try to position them when discussing implementation:
-
First of all there is the recently launched ISO Records Management
Standard (15489:2001) upon which I will dwell a little bit more in the next
section. Closely related and a good and useful example of guidance for
implementing the records management standard is the Australian ‘Manual for
Designing and Implementing Record Keeping Systems’ (DIRKS).
-
The results of several projects such as the Inter Pares project (with
requirements for preserving authentic records), the European NEDLIB project (about
long term preservation of digital publications), the NARA project (on
preserving digital information using XML), the Dutch Testbed project (on
possible preservation strategies and practical guidelines) offer other rich
sources for implementation.
-
Reference models for a preservation function, such as the Open Archival
Information System (OAIS) and the Inter Pares preservation function model for
archival records, provide useful overall frameworks with all aspects that have
to be taken into account.
-
Collaborative initiatives that enable information exchange, discussion
of key issues, dissemination of reasearch results, new research projects,
trials and so on, such as ERPANET and PADIFORUM.
-
Metadata standards, such as the resource discovery metadata set of
Dublin Core, and archival description standards as ISAD/G and ISAAR, the
preservation metadata report of CEDARS and the recent report of the
preservation metadata working group of RLG.[4]
In this area much activity is going on, including defining record keeping
metadata guidelines based on the ISO records management standard.
-
The draft report of RLG/OCLC ‘Attributes of a trusted digital
repository’, that provides an overview of all kinds of aspects relevant for the
proper, safe and reliable management and preservation of digital objects.[5]
With this report organisations are able to identify to what extent they comply
with all the requirements for preserving digital objects in a safe and reliable
manner.
It is clear that one is not alone in this
area. The point is how to find your way through the results of all these
projects and how to assess and to identify what is relevant and useful. The
above mentioned reference models offer an initial and useful framework, but
they still are not easy to understand and more practical guidance is needed for
organisations to develop their strategy, to start a program and to implement
methods and infrastructures. Especially the issue of costs and risk analysis is
still unclear. There is yet not much experience with preservation of current
types of digital records. Most experience has been gathered with preservation
of relatively simple databases, not with the kinds of records or objects
records managers and archivists or other disciplines are confronted with now.
In the following section I will try to discuss these implementation issues a
little more in depth.
Most people consider requirements,
frameworks, and the results of most of the projects as rather theoretical and
want more practical guidance. In other words to know what the requirements are
and what they really mean is one thing, the implementation of them another. At
least two perspectives can be distinguished, that of the record creating
organisations and that of the archival institutions, but more important
implementation has different aspects that have to be taken care of, e.g
organisational, technical, and archival. The latter has been discussed above in
the section on requirements, but the other two need further explanation.
One key principle is that preservation does
not start at the moment records are transferred to an archival institution, or
even at the moment when records will be created, but already in the design
stage of information systems. It requires already in that stage, before any
record has been created or received, an analysis of the relevance of the
business process as such and in relation to other business processes, of the
regulatory context, and of the records that will or need to be received and/or
created . Based on this first assessment the specific requirements for records
and records management can be identified and it will also be clear whether
records with archival value will be produced and subsequently whether long term
preservation measures have to be taken. This approach will be the most
cost-effective.
From the perspective of archival institutions
the position taken will be mostly as follows:
1.
The responsibility for the accessibility and preservation of archival
records lies with the record creating organisations as long as they are not
transferred
2.
The archival requirements are not something on their own. They need to
be embedded in the whole of records management measures, procedures and so on
(including the records that do not have archival value), especially in the case
of digital records. These records are subject to the rapid developments in IT
(as regards to storage media, storage formats, software versions, hardware
components and so on) and the subsequent threat of obsolescence. Five years is
already a long period. Many records have to be retained and preserved for
reasons of carrying out activities or to be accountable during a certain
period, although they do not have archival value. That requires for reasons
just mentioned that proper attention has to be paid to their management (e.g.
procedures, adequate systems, and access control) and technical maintenance,
such as refreshing, conversion and even migration of digital files.
3.
The current developments in government to realise digital service
delivery to citizens (in e-government programs) and in business companies are a
good starting point or trigger for implementing and integrating the necessary
requirements and record keeping functionality including preservation. For these
new online services whole new infrastructures have to be created and developed.
4.
The archivist may help and support in these developments with his
expertise. This supposes though that that expertise and knowlegde is available
within archival institutions and that is not always the case, to say the least.
This is all true, but archival institutions
also play a key role in detemining what records have archival value and as such
they are involved, as also identified in the ICA ‘Guide for managing electronic
records from an archival perspective’.[6]
The other point is that no government organisation (at least in the
Netherlands) really has organised its digital records management appropriately
if at all. They are still very much in the area of experimenting and pilot
projects. In most cases still paper records are being archived, though they may
have been created digitally with word processing applications. That goes
particularly for certain types of digital records such as e-mail messages that
are being printed. Nonetheless theincrease of digital documents (including
databases) that should be considered records, but are not managed as such, is
huge. Take only the large amount of e-mail messages that is being sent and
received, and the many databases and documents made with word processing
applications, they all fall under the same category. Records management
applications are not yet in place, but even if they are, preservation is not
their strength. That touches another issue, the lack of adequate IT-tools.
Suppliers or vendors offer a range of software tools, that in their opinion are
compliant with all records management requirements, but there are no sufficient
assessment tools to certify that, and worse there is not one set of
requirements. The DoD 5015.2 standard for instance is only one of many and it
has a certification process connected to it. Nonetheless these emerging sets of
requirements provide a first basis for developing adequate software tools. In
the following section I will briefly descibe the main developments that are
taking place to date, starting with the ISO records management standard, a
first common and broadly accepted framework.
The international Records Management Standaard, ISO 15489-1:2001, is one of the
recent major achievements in the area of managing a specific group of
information resources, called records. [7] It provides an excellent starting point for
implementing an appropriate records management regime, that also addresses
preservation, though only at a high level.
One important and leading principle in
organising records management, including long term preservation for an
organisation is that it should be the business process that determines and
influences the way how to implement it, and not the point of view of records
management itself. That view is taken by the ISO records management standard
and has also been translated in a ministerial regulation on accessibility and
arrangement of archival records in the Netherlands.
It is the organisation that has to carry out
activities and that has to identify the risks if records management does not
meet all requirements. That seems to be obvious, but in practice both have
grown apart and are not always tuned in an optimal way to say the least. The
ISO standard articulates this principle by offering a framework that identifies
the needs of an organisation to have evidence of its activities, to enable
accountability and to manage its information resources for as long as needed.
The standard also includes a methodology that
supports the integration of records management into business processes. This
methodology consists roughly of the following steps:
-
A preliminary investigation that identifies the
role and objectives of an organisation and her regulatory framework
-
An analysis of business activities that should lead to a classification
scheme and an overview of the different business processes and transactions.
-
Identification of the records requirements in each business activity
-
Assessment of the existing records systems based on the requirements
evaluation
-
Identification of strategies in order to comply with the records
requirements
-
Design of a records system
-
Implementation of a records system
-
Post-implementation review of the performance of the records system and
if necessary adaptation.
This methodology provides a firm basis for
achieving proper records management and implementation of a records system.
What a records system should look like is another question. That depends on the
requirements derived from the activities of an organisation, its regulatory
framework and the risk analysis. The answers to these questions are provided by
using this methodology, which is
further described in the Technical Report accompanying the ISO standard. It is
derived from the Australian methodology called Designing and Implementing Recordkeeping Systems (DIRKS). [8] Using this methodology requires a big
investment. The standard and its technical report with guidelines offer a good
overview of all the issues involved in achieving proper records management and
the records processes of which its consist, including preservation. As such it
can also be a model for implementing preservation strategies.
With the archival requirements and the
framework of the records management standard the following set of initial steps
for implementing proper preservation strategies could be carried out:
1.
Identify the ambitions of the organisation in relation to e-government
or electronic service delivery
2.
Assess the current situation as regards to records management, records
systems and use the ISO records management standard as a check list. Determine
based on the results of both the approach, objectives and phases or steps for
change.
3.
Involve senior management by asking them to make policy statements
about the value of information as source for creating knowledge, the quality of
information management, the use of standards, the necessity of proper
accessibility, and about acquiring the necessary skills and competencies, and
so forth.
4.
Follow the steps for implementation as indicated in the ISO standard.
In this phase the above mentioned reference models for preservation can be
used.
5.
Monitor and review regularly the performance of the implemented systems
and procedures.
This is only an example and can be adapted if
needed, based on the circumstances in an organisation, and extended. In
addition to this framework for getting proper records management in place,
reference models for preservation as the OAIS-model or the more
records-specific Inter Pares preservation function model can be used.[9]
They offer a more in depth insight of the necessary functionality of
maintaining, preserving, and reproducing digital objects. Though the models
focus on long term preservation the functionality is not fundamentally
different for managing current records. The same can be said of the archival
requirements that govern the preservation system.
Since almost no organisation will start from
scratch one of the issues to be dealt with is that of the already existing
information resources (e.g. in legacy systems). First of all has to be
identified whether they have archival value or not. Another issue is what
relation they have with paper files. Since most organisations are still in a
transitory state towards working completely digitally that relationship will
exist in most cases. Much of the archiving is still taking place by using paper
files. What is being done electronically is communication by using e-mail and
storage by using databases of all sorts, as for instance for geo-spatial data.
However in most of these cases there has been or will be no proper electronic
records management. That means that there is a huge backlog in trying to make
these information resources preservable and usable for future generations. Much effort and resources will be needed to
take care of those archival records that are created in this situation. It is a
separate issue, but no less important and it requires a separate approach.
Organisations need to stop this hybrid situation as soon as possible and take
the necessary measures that ensure that appropriate management of digital
records is in place. In the case of legacy systems each of them has to be
analysed to see whether they contain valuable records and if so, what steps
have to be taken to preserve them properly.
Useful for building a program for (longterm)
preservation are gateways as for instance the PADI website and very soon also the
ERPANET website that intends to provide reviewed information on preservation
projects, websites and publications.[10]
They both are a solid starting point for getting an overview of the field of
digital preservation and the latest state of the art.
Some institutions recently started to build their own repositories as for
instance the Dutch Royal Library, one of the leading organisations in this
area, and the British Library.[11]
In the USA the NARA is building a program to develop persistent strategies for preservation
and the Library of Congress is leading a so-called ‘national digital information infrastructure preservation program’
that should provide a national planning report or schedule for dealing with
long-term preservation.[12]
More practical experience is gathered in
projects as CAMiLEON and the Dutch testbedproject for longterm preservation.
The first tries to identify the applicability of emulation and migration for
different kinds of digital objects and analyses the significant properties of
these objects. The Testbed project carries out all kinds of experiments in
order to identify what preservation strategies can be best applied to what
kinds of records.[13] It are
these practical guidelines that people help to understand the issues and enable
them to implement the proper methods for preservation of valuable information
resources.
The challenge to preserve digital objects in
all their forms is huge, just because of their nature and triggers everybody
who has an interest in it and what is more, there is a growing tendency to
co-ordinate efforts in developing solutions, strategies and approaches.
Examples of collaboration are, apart from the ISO standardisation work, the
recent Digital Preservation Coalition in the UK, the international Inter Pares
project, the Research Libraries Group (RLG) though that existed already but has
taken an important role in it, the already mentioned European ERPANET-project
and last but not least the standardisation work around the reference model of
the Open Archival Information System (OAIS).
Partners can and should be found within and
across domains as is shown for example in the OAIS working group with
librarians, archivists, and scientists, and in the Inter Pares project, with
archivists, computer scientists, and representatives from the music community
and the scientific community.
It is also important to realise that in order
to carry digital records through time and deliver them to current and future
users, a close relationship is necessary between creators and preservers on the
one hand and preservers and users on the other. In the records domain for
instance that concerns the record creating organisations, in the library domain
that may be the collaboration between publishers and librarians. Sometimes
these roles may be united within one organisation, but as time goes by that
will hardly be the case. Creators produce digital records for carrying out
their business and preservers have the responsibility to maintain them over time
and keep them accessible and
understandable for other reasons that have a broader, societal scope, e.g.
cultural heritage. Understanding
(current and future) users and the way they use records or information
resources and what is of value to them is another issue preservers have to deal
with. It will influence preservation strategies and the management of records.
In practice however this is an area that needs much more attention than is is
given today, especially in the archival community.
Figure 2
The digital object and relevant actors
In diagram 2 the connection between the
different players is shown. Two other types of players can be added here. First
there is the IT-industry that provides the technology to produce and maintain
digital objects. They should be more involved in developing solutions in
longterm preservation and provide tools. As such they should have a connection
with the preservation infrastructure and standardised methods. Finally there is
a group that plays a role in controlling the quality of the preservation
systems and procedures. Quality assurance will be an important factor in
longterm preservation.
Preserving digital objects poses
unprecedented challenges to all organisations that have a responsibility in
this area and many of them are trying to find solutions both within their own
domain and together with other partners. The awareness of the issues on digital
preservation is now at a point that this can be better achieved and that more
progress can be made. In many government agencies that awareness is still
insufficient though.
The increasing effort that is going into this
area of research is producing more and more valuable results that can be used
in building preservation programs and strategies, but also more practical
implementation. The situation has reached the stage where theory is being moved
into practice. The first digital repositories are being built and projects with
experiments and testbeds collect useful practical experience, guidelines and
facts. Still, a lot of further work and research has to be done, both in theory
and in practice. Preservation technologies as migration and emulation need to
be investigated more in depth in relation to different types of records or
objects, before they can be applied more succesfully. The increasing complexity
of types of electronic records is making that even more necessary. The growing
opinion is that not one strategy will be sufficient, but that a combination of
strategies will be necessary to be succesful in preservation.
In the records domain measures with respect
to longterm preservation have to be taken already in the design stage of new
systems in record creating agencies. They have to be integrated in records
systems or business systems depending on where the records will be managed. It
will be too costly if they will be taken later in the life cycle.
In this respect record creating organisations
play an essential role and it will be crucial that senior management expresses
its interest and commitment in achieving and maintaining good records
management including longterm preservation and make that position clear to all
employees in an organisation. Regulations alone will not be sufficient. If
people do not see the relevance of these or that they can benefit they will not
be obeyed, certainly not if there are no sanctions.
An important trigger in establishing the
appropriate environment for managing and preserving electronic records can and
should be the current developments around e-government and electronic service
delivery. This will require that organisations review their existing
infrastructures, both technically and organisationally, and rebuild them
completely. This will only be succesful if the information and records
management is integrated properly.
There is also an increasing tendency to
building partnerships and networks of interested organisations, not only within
certain domains, as library or archival communities, but also across domains.
Not only the openness of the world wide web invites or perhaps forces
organisations to this, but there are more triggers as the awareness of similar
research going on in other domains, and in relation to that the need of
interdisciplinary research in order to achieve solutions, the objective to
avoid redundancy, the willingness to share results of projects or research and
to discuss key issues and so on. In this dynamic field the archival community
in general however still has to catch up with the developments going on in
otherdomains. Collaborative efforts as the European funded project ERPANET,
which is not focusing on records or archives in particular, may help in this
respect and will also provide useful platforms for information exchange and for
building research agendas.
Some
references:
Inter Pares project on authenticity and
integrity of records and long term preservation: www.interpares.org
Reference Model for an Open Archival Information
System, ccsds 650.0-R-1.2 (draft) July 2001:
http://ssdoo.gsfc.nasa.gov/nost/isoas/ref_model.html
ISO Records Management Standard 15489:2001 : www.iso.ch (English)
International standard for Archival Description
(ISAD/G) and International Standard for archival authority files (ISAAR): www.ica.org/cgi-bin
Dublin Core, a standard for information
resource discovery: http://dublincore.org
NARA project on digital preservation: www.sdsc.edu/nara/publications.html
XML and other related standards: www.w3c.org
NEDLIB project, a collaborative project of
several national libraries in Europe: www.kb.nl/coop/nedlib
RLG report ‘Attributes of a trusted digital
repository’ : www.rlg.org/pr/pr2002-repositories.html
PADI, main gateway to information about
digital preservation: www.nla.gov.au/padi
[1] Hans
Hofman is working as a senior advisor at the National Archives of the Netherlands
and involved in several (national and international) projects in the area of
digital preservation, such as the Inter Pares project and the ISO TC46/SC11
records management committee, in particular as chair of the Working Group on
records management metadata.
[2] More information can be found
on the website of Luciana Duranti (www.slais.ubc.ca/users/duranti)
and of the Inter Pares project (www.interpares.org).
[3] In the CAMiLEON project
research is being done to establish the requirements (future) users will have
as regards to authenticity of digital objects. The context of the user is the
important factor. What is the reason for using a record or object, but it is
equally relevant to be aware of the intention of the creator in creating the
record or object.
[4] The archival description
standards are available on www.ica.org/cgi-bin.
The CEDARS report on www.leeds.ac.uk/cedars. The RLG metadata report can be found on www.rlg.org or www.oclc.org/research/pmwg.
[5] The final report can be found
on www.rlg.org/pr/pr2002-repositories.html
or www.rlg.org/longterm/repositories.pdf. .
[6] Published in 1997. Website: www.ica.org/eng/mb/com/cer/index.htm
[7] Available on: www.iso.ch
[8] The manual of DIRKS can be
found on http://www.naa.gov.au/recordkeeping/dirks/dirksman/dirks.html.
[9] See for the Inter Pares
preservation function model: www.interpares.org/documents.
The model itself is not finished yet, but there is a high level description in
the final Inter Pares 1 project report.
[10] See www.nla.gov.au/padi and www.erpanet.org.
[11] Information on the joint
IBM/KB study on longterm preservation can be found on: www.kb.nl/kb/resources/frameset_kenniscentrum.html
[12] Amy Friedlander, ‘The
National Digital Information Infrastructure Preservation Program’, D-Lib Magazine Vol.8 No. 4, April 2002 (www.dlib.org/dlib/april02/friedlander/04friedlander.html).
Information on the NARA project can be found on the website of NARA (www.nara.gov) and of the San Diego Super
Computer Centre (www.sdsc.edu/nara/publications).
[13] Information about CAMiLEON can be found on www.si.umich.edu/CAMILEON/about.htm; information about the Dutch testbed project: www.digitaleduurzaamheid.nl/index.cfm?paginakeuze=181&categorie=2