NDSA:MetaArchive Use Cases

From DLF Wiki
Revision as of 15:18, 11 February 2016 by Dlfadm (talk | contribs) (2 revisions imported: Migrate NDSA content from Library of Congress)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

MetaArchive Use Cases

The following use case is intended to spotlight a couple of examples of applying the Reference Model and comparing/contrasting its concepts to the MetaArchive Cooperative.

Open Pull vs. Push on Ingest

Use Case

Preparing content for ingest in the MetaArchive stands somewhat in distinction from other digital archives where the final preparation of content for preservation is almost entirely in the hands of Archival Management (an archive and its staff). In such cases, Producers (content providers) typically supply half-formed Submission Information Packages (SIPs), and the Archival Management stages it prior to preservation to round out the SIP as an Archival Information Package (AIP) conforming to the repository’s requirements.

Because of our participatory and decentralized model, and our use of the LOCKSS peer-to-peer software, the roles and responsibilities of ingest and formation of SIPs (and even AIPs) rest primarily with the Producer (member institution), who has to very closely align the organization of data with defining the on-going and open ingest mechanism for that data. This in turn largely determines the successful formation of an AIP. Again, this is a little different than most archives where the final preparation of content, moving it from a SIP to an AIP, is often entirely in the hands of Archival Management (which is an archive and its internal staff).

This may be characteristic (but not at all intrinsic) of other repositories that rely heavily upon on-going and open web crawl mechanisms for ingest – making use of a “pull” vs. “push” metaphor.

We would invite further use cases.

Comments

The nature of SIPs and AIPs and their formation is not really at issue. It is perhaps more so an issue of clarifying or modeling definitions of Producers and Archival Management. The Reference Model is often keen to point out that one Archive may be a Producer or Consumer of another Archive. In such arrangements the functional elements of Archival Management remain encapsulated within an Archive. For participatory and distributed models such as MetaArchive it might serve to affirm that various responsibilities of Archival Management (completing an AIP) may be accomplished in some cases by those entities also serving otherwise as Producers. Matt.schultz

Questions

  1. What additional safeguards should an Archive implement to ensure proper AIP formation without subsuming Producer participation & responsibility? Matt.schultz


Dark Archiving, Full Preservation & Designated Communities

Use Case

Because the MetaArchive is engaging in dark archiving (i.e., not providing on-going end-user access) and doing so in a cooperative fashion, it has, to date, largely shifted responsibility for end-user (Access) concerns closer to the Producer (who also occupies the role of the Consumer – through their extended scholarly communities).

This has meant that issues related to ensuring “Full Preservation” (such as vigorously monitoring representation information, requiring normalization, carrying out format migration, readability via access interfaces, etc.) get considered in a very deliberative fashion – balancing preservation benefits versus impacts of cost and effort, and properly allocating responsibilities.

Through on-going communication with and by our members and through routine reporting, monitoring, and recovery testing, the MetaArchive members gauge their needs in tandem and experiment with issues related to long-term renderability of contributed content as resources and interest permit. The member buy-in ensures graded approaches and accountability around these issues over time.

Clearly this arrangement blends together many characteristics of Consumers, Producers, and Archival Management and makes for a less delineated notion of an Archive preserving on behalf of a Designated Community (more typically defined as an external requesting archive and/or end-user community of AIPs and DIPs; i.e., Consumers).

Comments

The Reference Model acknowledges that Archives may occasionally restrict access in a dark archive fashion – dark archiving itself is not at issue. However its participatory and distributed implementation in the MetaArchive means that the Consumer for all intents and purposes is also the Producer, and as mentioned earlier in some ways the Archival Management.

There is an intersection of both concern and desirability for ease of decision-making in regards to prioritizing the long-term renderability of contributed content. Each member Consumer/Producer, to date, has agreed to manage Full Preservation on an individual basis as opposed to positioning it as a universal requirement for Archival Management. The Archive places no technical barriers to formats, format migration, or the management of representation information nor does it impose it.

We think this a completely valid and acceptable approach given the decentralized accountability.

Questions

  1. Should the Reference Model better account for such participatory models, overlapping roles, and responsibilities through statements of exception when assigning the activities of Producers, Consumers for things like Content Information & Preservation Description Information on behalf of content for Designated Communities? Matt.schultz
  1. What would best practice recommendations be for member institutions and an archive in such a participatory model to effectively communicate to their Designated Communities their rationale for a measured and differentiated approach to servicing "Full Preservation"? Matt.schultz


<< Back to Other Gap Analyses & Use Cases