Disclaimers
Copyright warning
Copyright in all forms of the publications, both within the PANDORA Archive and on the publisher's sites, are held by the authors and/or publishers of the publications. For information about copyright ownership in each case, please refer to the publisher's copyright statement, usually found within the publication, or contact the author or publisher. Permission to copy any part of a publication must be permitted by the publisher/author or must be in accordance with the Copyright Act 1968. For example section 40 of this Act allows the copying of a reasonable portion of the publication for the purpose of research or study and section 41 allows the use of the publication for the purpose of criticism or review provided that sufficient acknowledgment is made.
Content disclaimer
Websites and web publications are selected and preserved in the Archive with the purpose of collecting a comprehensive range of works, views and opinions that document Australian society and culture as it is, so as to present a resource for current and future researchers that is accurate and is neither censored nor sanitised. Inevitably the content of some resources may offend or be distasteful to some people. Content that is known to breach the law will not be included and access will be removed in respect to content that is subsequently proven to be in contravention of the the law.
Inclusion of a website or web publication in the Archive does not represent any endorsement by the National Library of the content of the site either in regard to the veracity of the content nor its views. The Archive does not promote the content of the the archived websites or web documents other than as documents of potential research interest.
Preservation authenticity and persistence of PANDORA content
Authenticity of a digital archive is quite a complex matter and involves a number of factors. Are the individual items within the archive faithful copies of the original and does the archive as a whole have integrity? Autheniticity of a digital archive depends on effective, ongoing management strategies to ensure that the archive and individual items in it are not accidentally or deliberately changed.
Policies and procedures for authenticity at the title level
The National Library has a number of policies and procedures in place to keep the contents of the PANDORA Archive authentic, especially its intellectual content.
Substantial effort is invested in ensuring the authenticity and integrity of each individual title. In copying a publication or web site into the Archive, the policy is to maintain its 'look and feel', that is, its appearance and functionality, as well as its contents, to the fullest extent possible. In most cases this is achievable. In some cases, however, because of the technical set up of a site, or because of the limitations of current harvesting technology, it is not possible to reproduce a site exactly. Some functionality may be missing, for instance, the ability to conduct a search of an archive of back issues on the web site of an e-journal. Functions that involve interaction with publishers' sites have to be disabled.
Where it is not possible to replicate exactly the 'look and feel' and functionality of a title with current technology, the Library takes a pragmatic view and argues that it is better to compromise on these aspects for the benefit of being able to preserve the intellectual content. It is essential that the intellectual content is preserved exactly as produced by the publisher. All sites are quality checked and are accurate copies of the publishers' sites at time of archiving. Date of archiving of each instance is collected by the system and is clearly indicated on the title entry page for each title.
Preservation activity to keep titles accessible as hardware and software changes may also result in some changes to 'look and feel' in time. Once again, preservation of the intellectual content exactly as produced by the publisher will be of paramount concern.
Policies and practices for authenticity at the Archive level
At the Archive level, a number of policies and practices are in place to preserve authenticity.
At least three copies of each instance (regathered copy of a title) are kept.
- The preservation master, which is derived from the raw output of harvested files and associated work logs before human or machine intervention occurs. It is therefore an exact copy of what is downloaded from the publisher's site and will be kept in perpetuity.
- The access master, which is a set of harvested files that have undergone quality checking procedures (modification of links, addition of missing images, etc.) to make the instance suitable for public display. The access master will also be kept in perpetuity.
- The access copy, which is served to users of the Archive.
The preservation and access masters are stored separately from the access copy on the Library's secure Digital Object Storage System (DOSS). Backup tapes of both the DOSS and the PANDORA Archive server are maintained, including copies that are stored off-site.
These practices contribute to the authenticity of the Archive in the following ways:
- Should the Archive fail or be destroyed or corrupted, backup copies would enable the restoration of the Archive;
- Should a hacker succeed in tampering with one of the copies of an instance, it is highly unlikely that s/he would succeed in tampering with all copies on all backup tapes, and the instance could be restored from the intact copies;
- Should quality control or preservation activity change an instance to an unacceptable degree, or alter the intellectual content at all, the preservation master can be returned to for an exact replication of the title as it was on the publisher's site at the time of archiving.
Persistent identifiers
Another aspect of authenticity is persistence. Each item in the Archive, from title level right down through instances and parts of instances to component files, has a unique persistent identifier automatically assigned by the digital archiving system. This enables researchers and authors to cite works and parts of works, for instance, an article in an e-journal, using the appropriate persistent identifier. Readers can return to the cited item in the Archive again and again, confident that it will remain there persistently and that it will be the same.
