|
home · articles · preserving digital heritage for perpetuity… or at least for the next 25 years |
A key change at iCommons
If you're not part of the iCommons mailing list, take a look at the letter that Heather Ford, Executive Director of iCommons, sent to the list yesterday:
Dear friends,
At the 2 August iCommons Board Meeting, the board decided to make some difficult but necessary changes at iCommons. It has become clear over the past months that our vision for iCommons is different from the... more
|
| |
Preserving Digital Heritage for Perpetuity… or at Least for the Next 25 Years
Sarah Kansa · San Francisco, CA (United States) · May 24th, 2008 12:26 am · 30 votes · 3 comments
|
| |
Sharing heritage digitally is becoming more and more commonplace. In the past few years alone, a new awareness has emerged of the value of digital sharing. The National Endowment for the Humanities has initiated a new Digital Humanities Program, which awarded over $1 million in 2007. The Society for American Archaeology’s annual conference in Vancouver in March saw twice as many sessions on sharing and preserving digital archaeological content than the previous year (see here). Clearly, interest is rising and more and more projects seek a way to share digital content. How can we guarantee that the content shared will be archived and accessible for future generations to learn and build upon?
The exponentially increasing amount of digital humanities content available online raises concerns over longevity. Data loss is especially ironic in the field of archaeology, where information rests in the ground for sometimes tens of thousands of years, only to be excavated (a destructive process), recorded and then potentially lost.
At the conference in a forum “Digital Antiquity: Planning an Information Infrastructure for Archaeology”, the question was raised: How long is perpetuity? Julian Richards, of the UK’s Archaeology Data Service, informed the crowd that we now know the answer to this question: About 25 years. In another forum on “Converging Communities in Digital Heritage” an even less optimistic definition of “in perpetuity” was offered: Five to ten years.
Where are these definitions coming from and what can we do to help extend the concept of “forever”?
The challenge of “forever”
There are great challenges to storing digital content for the future. The best solution for paper storage—lots and lots of copies—may prove to be the best solution for digital longevity as well. The more digital content is used, the more copies of it are out there, protecting it from loss. Most content, whatever its medium, needs active preservation, unless you happen to write on clay tablets or stone. Digital files happen to need more active upkeep over shorter time spans (max. 25 years) than the more traditional medium of paper.
One of the key differences between digital and print, therefore, is in the instability of the medium. Paper has remained a stable medium for thousands of years, witness to countless changes in languages and scripts. Digital poses a more complicated scenario: not only can the formats changing rapidly (languages, coding systems, etc), but the media are also changing (floppy discs, hard drives, cds, dvds, etc). Digital content that is relevant and widely used will have a greater chance of being updated to new formats. Here is where the need for open access and Creative Commons licenses becomes paramount: Digital content that carries clear and transferable metadata that allow for its use and reuse will stand a far better chance of being copied, converted to new formats and media and saved into perpetuity.
Taking small steps to the future
Complicating matters is the extreme diversity of digital content produced worldwide. In a related article entitled “Collating online collections: A comparison of cultural heritage collections online” Francis Deblauwe illustrates this diversity in his survey of digital cultural heritage collections. Each shared collection will have unique goals and will want to highlight specific content. The diversity of content, institutional and organisational motivations for online access to cultural heritage collections, as well as the myriad of different cultural heritage taxonomic and classification systems typically mean a great deal of development of highly customised technology platforms for putting heritage collections online. This means little transferability of technology solutions developed for one specific collection. It is usually only the best-resourced and well-staffed organisations that have the capacity to put collections online. This problem also makes interoperability difficult, since linking these customised systems together represents a host of technical and semantic problems. We are now faced with the problem of an inundation of digital content that is stuck in silos, unable to be used and unable to communicate with other online content.
Content in such silos is harder to build upon and reuse, making it less relevant and valued by the larger community, Therefore, closed silos with little data interoperability represent a preservation risk. How do we free this threatened content from silos and assure that it is used and kept safe? My own organization, the Alexandria Archive Institute (AAI), is addressing this issue from a number of angles, the general premise being that content becomes safer the more it is used.
The more valuable information is, the more likely it is to survive in digital format. Openness can go a long way to increase the value of content. Open Context, a free, online content dissemination system developed by the AAI, aims to increase the value of digital content through open licensing and open standards. It makes all content available globally for use, reuse, and commentary. It implements “faceted browsing” to facilitate exploration and discovery of its holdings. Ease of use, discovery and interoperability, therefore, are paramount features for assuring that data can be accessed and their value appreciated. Open Context is generalised to work with lots of content, rather than being a specialised custom solution for one collection. Heritage collections typically contain a diversity of structured data and Open Context offers a way to share such data without requiring custom development for each new collection. It converts data to a common, non-proprietary XML format, and future developments will use web services to expose data in common XML formats (such as Atom to make Open Context easier to use and “mash up” with other systems.)
As Deblauwe’s article demonstrates, not all digital heritage projects take a clear stand regarding access and reuse. Recognising that online content with restrictive or unavailable copyright is less likely to be used, Open Context makes crucial permissions explicit with Creative Commons licenses. Clear attribution and citation information are available for every item in the system. Clear information about how content can be used increases its impact and also assures that authorship is archived along with original content. This is important for encouraging scholars, who are often the custodians of cultural heritage content, to participate in openness.
In addition to examples of ways to increase the impact and exposure of data, creators of digital content would benefit from general guidelines for understanding user experience with digital humanities collections (that is, what are people using and why are they using it?). The AAI is initiating an assessment of community needs around digital cultural heritage resources, specifically with regard to Open Context. This study will identify the diverse needs of the various communities served by cultural heritage data dissemination and produce a set of guidelines to enable projects and individuals to broadly and intelligently distribute their content, make it useful to others, and assure that it retains its integrity for long-term future use.
There is no lack of digital content out there. Each community, institution or individual creating and sharing it needs to also take responsibility for preserving it. Currently, content isolated in silos stands the least chance of survival because of its inaccessibility and the lack of portability and re-usability of content. An open access (and open licensing and open standards) approach will go a long way towards preserving our digital cultural heritage in perpetuity, albeit a few years at a time.
tags: san-francisco united states education local-context-global-commons digital-archives cultural-heritage open-context
|
|
| |
comments  |
 add a comment |
|
| |
 |
Thanks for this article Sarah, you bring up many challenging points about preservation of digital cultural heritage on many fronts. You suggest that "Clear information about how content can be used increases its impact and also assures that authorship is archived along with original content. This is important for encouraging scholars, who are often the custodians of cultural heritage content, to participate in openness."
I agree that one important step with any of these types of projects is clearly addressing how and if content can be "shared". Creative Commons licenses help with this in some cases. I wonder, though, with the focus on "scholars" as the custodians, if we aren't leaving out another group--that is the indigenous custodians of the materials. Certainly it's legally the case that much cultural heritage is in the control of scholars, museums etc., but I wonder if there is a way to reach out and extend the reach of the notion of custodianship here? Certainly it would, most likely, then force scholars to grapple with the bounds and reach of "openness" in other contexts and stemming from multiple types of histories (of the objects and data as well as the collectors). I am involved with two projects (one in Australia and one in the US) looking at how to involve both Indigenous communities as well as scholars who have collected and deposited collections at various institutions. The goal is to decide the parameters for the "sharing" of this content (as well as the metadata) in dialogue with the Indigenous communities--recognizing that there is no one size fits all notion of "openness" that might translate. A similar project is underway at the UBC Museum of Anthropology. The technology involved with Open Context seems a great resource and model for cultural heritage, but does Open Context grapple with any of these issues of authorship and openness in relation to communities other than "scholars"?
Does Open Context
Kimberly Christen · Pullman (United States) · May 24th, 2008 1:48 am
2 out of 2 people believe this is useful
your take:   |
 |
Hi Kimberly and Sarah,
I think Sarah has a very good point that openness can be a great tool for preservation.
I also agree with Kimberly's point about different ideas about custodianship. Valuing data preservation has its own specific cultural context. The Society for American Archivists recently published some protocols to guide working with indigenous communities with regard to archival materials. Issues of preservation, access / restrictions, repatriation, and even curated destruction of materials are discussed, since each of these may be seen as appropriate depending on the specifics of the communities, history, and circumstances involved.
As far as Open Context goes, Kimberly raises an interesting point. Right now, "scholars" have received the main attention in Open Context for linking content with people and roles. This is because most metadata standards (especially Dublin Core) and the infrastructure built off of these standards (such as the Open Archives Initiative) all have a pretty similar world-view and model of authorship behind them. For example, the citation tool Zotero expects metadata of a certain type and format, and that's what Open Context delivers in order to work well with other research tools.
Nevertheless, in principle, Open Context's network-graph data model can support many different kinds of relationships between people, groups, and content (for example, it can show some people as "custodian", "keeper", etc.). You can also come up with rules for who can view or edit material based on such custom roles. However, these won't easily fit into many common existing Internet wide-metadata standards. Semantic Web technologies should similarly work with any culturally specific notions of authorship/custodianship, but these technologies are pretty arcane and therefore costly...
At any rate, some of these issues are just fundamentally hard. Creative Commons licenses are widely adopted standards. You get lots of benefits in terms of reduced costs and readily used infrastructure by adopting them. However, they contain their own cultural biases and are not appropriate for every indigenous community application. Same thing with technical standards.
All of this really shows how "standards" aren't politically and cultural neutral. If your world-view becomes the basis of a standard, other "nonstandard" world-views become disproportionately more costly to maintain. The ethics of all this is very important, given the role standards play in law, technology, and business in an ever more interconnected world.
Eric Kansa · Berkeley (United States) · May 28th, 2008 2:57 am
1 out of 1 person believes this is useful
your take: |
|
|
|
| |
add a comment: you must be logged on in order to comment. please log in or register at iCommons.org and and your comments right after.
|
|
| |
|
|
|
|
|
| |
|
|
send
printer-friendly version
invite to vote