Crowd sourced Representation Information for Supporting Preservation (CRISP)

A new initiative to encourage collaboration in collecting representation information to support long term digital preservation 

The challenges

It has long been agreed that representation information is required for digital resources to remain accessible into the future. However, the collection of representation information is a significant challenge. There is no formal consensus on 'which' representation information is required and how that may differ for different digital objects. A number of representation information repositories have been established, but they are sparsely populated. Furthermore, experiences to date suggest that collecting RI is a highly resource intensive activity. How then can we address these issues? 

The internet remains one of the best sources of information for solving digital preservation challenges, but as we know, websites themselves are far from permanent. Vital information about preservation tools and file formats can be transitory. Finding the best information on the web can also be difficult. Community-created resources are typically spread around quite thinly, with much duplication (as for example with lists and registries of tools useful for digital preservation).

Collaboration and coordination across the wider preservation and curation communities is substantially lacking. Encouraging more coordination and awareness of existing intiatives would reduce duplication of resources and maximise effort in creating and maintaining the resources we need to make preservation effective.

The proposal

We are therefore proposing an online collaborative event that will help us encourage identification of representation information sources from the ground up. We will set the barrier for participation as low as possible, and encourage contributions from as many digital preservation practitioners as possible. Although the resulting collaboration will be at a superficial level, we hope this will be useful in kick starting further joined up activities and developing a greater appreciation across the community for resources that are already available.

We are launching an initiative that will ask individuals to contribute at least one URL of a website useful for preservation, along with tags to enable that website to be categorised. We're looking for websites on the following topics:

• Information about File formats 
• Other types of representation information (give examples)

We are not seeking to include specific software tools at present, though this may be incorporated into later stages of the project.

The resulting lists will be passed to participating web archives who will crawl the sites and preserve them for posterity. Direct access to the tagged lists will also allow users to browse or search for useful entries.

Practical matters

The crowd sourcing will be captured via two mechnanisms that will make it easy for interested parties to participate. The primary method of submitting information is via this form. The second, more experimental approach is via mentions of this @dpref Twitter account. We were hoping to use a social bookmarking system like Delicious or Diigo, but we found them to either be unreliable or have too high a barrier to submission. Both also failed to have suitable methods for exporting the curated dataset.

Timing and engagement

We are hoping to tie in the launch of the initiative with an appropriate digital preservation conference in order to give it an initial boost, but we are hoping primarily to engage with the community via online communication.

Who are "we" and how can you support CRISP?

This is a collaborative initiative between the digital preservation and web archiving teams at the British Library, alongside the JISC funded SPRUCE Project. SPRUCE is tasked with supporting digital preservation via community approaches. However, this initiative will take place in "neutral" locations (Twitter, Delicious) and will not be strongly branded (other than acknowledging funding sources). Our key aim is to foster greater collaboration and communication and we feel that this is best labelled as a community owned effort, assuming sufficient buy-in becomes possible.

However, we would like you to support this initiative and put your name to it. Please email: "p (dot) r (dot) wheatley (at) leeds (dot) ac (dot) uk" with your name and institution/project affiliation and we will add you to the list we use to publicise the event. Please let us know if you would like to assist with announcing, and disseminating the call for participation in CRISP.

We would also like existing web archives to support this initiative by crawling the list of URLs that we collect. Again, please get in touch if you would like to do this for us.

More?

If we're able to make CRISP a success we'd like to follow up with further online collaborative activities, that may take in one or more of the following topics: experiences of using preservation tools, file format magic, tool registry entries, and preservation requirements. We are of course open to suggestions from the community!