Metadata is critical for organizing archival and museum collections. It not only provides fundamental facts for researchers like the dates, dimensions, and creators of objects, but can also make resources more discoverable by providing search terms or multiple ways to sort and browse through items in an online database. The role of metadata as part of an interpretive exhibit is more ambiguous, however. While detailed metadata about exhibit items should be made available, incorporating it prominently into an exhibit can create large, crowded labels that distract from the exhibit narrative. Furthermore, whereas historical exhibits often select and juxtapose objects to reveal a shared theme or context, museum metadata consists of information about artifacts as solitary pieces, potentially disrupting an exhibit’s effort to emphasize context and relationships.

The online exhibits at Recollection Wisconsin achieve an ideal balance between making rich metadata available for those who want it without interrupting their exhibits’ narrative flow. The Exploring Effigy Mounds exhibit, for example, showcases photographs and documents about mounds from collections around Wisconsin while embedding them within a succinct narrative about the origins, destruction, and protection of the monumental earthworks built by indigenous people. A caption below each object gives its creator, date, and current owner in a concise format that does not distract from the primary exhibit text. Viewers can still access more detailed metadata, however, because each object is hyperlinked to its corresponding database page at the repository that owns the item. The database page provides full catalog metadata including dimensions, physical description, rights, and other information not included in the exhibit captions. Researchers can also search these databases individually or all together via Recollection Wisconsin’s unified search engine to find additional related objects not shown in the exhibit itself.

A potential downside to the exhibits at Recollection Wisconsin is that because the databases holding metadata are provided by many different repositories, they could conceivably provide data in inconsistent formats or with varied levels of description. In the Exploring Effigy Mounds exhibit, this is not really a problem, because each participating repository uses the same database system, ContentDM, and a mostly common set of metadata fields. An exhibit that showcased items from competing databases with multiple metadata standards like Dublin Core, CDWA, EAD, and MARC, however, might make it more difficult for researchers to compare specific details about the exhibit items. Another potential issue with linking to the metadata at another repository is that if that institution changes or updates its database system, it could potentially change the URL of the item description and break the exhibit’s hyperlinks, requiring vigilant maintenance by the exhibit host.

The Wisconsin Historical Museum’s online exhibits take a different approach to metadata that is adequate, if a bit less successful. Although the Wisconsin Historical Society hosts several collection databases with rich metadata — including many of the databases linked from Recollection Wisconsin — the Wisconsin Historical Museum does not link all of its own online exhibits to the society’s databases. Instead, in exhibits like From Shell to Symbol: Art of the Ethnic Easter Egg, basic metadata such as the date, creator, and object number is displayed only in captions. As at “Exploring Effigy Mounds,” these captions are an unobtrusive way to label items with their most important metadata. The “From Shell to Symbol” exhibit even succeeds at incorporating some of these basic facts in its exhibition narrative in a seamless, contextualized way. The drawback, however, is that no more detailed information is immediately available to researchers. Facts like dimensions, materials, and donors are unavailable unless they happen to be mentioned in the interpretive labels. Furthermore, the display of basic metadata in captions is not standardized — some pages of the Ethnic Easter Egg exhibit include the creator and date of objects in interpretive text but not individual captions, while others show this information in captions only after an image is clicked. These are minor issues, and the exhibit still provides visitors the most crucial metadata in one place or another, but linking the exhibit to database entries with fuller metadata for each item would provide a more consistent gateway for research — albeit perhaps at the expense of more staff time in entering data.

For my group’s project in Public History New Media at Loyola this semester, we’ll be using Omeka to build an online collection and exhibit. Omeka provides tools to create individual item pages with detailed metadata as well as narrative exhibits that showcase selected items together with interpretive text. As a result, we’ll be able to create exhibits that that link to items with detailed, consistent metadata hosted on our own site — allowing viewers to either follow the flow of the exhibit or click away to find detailed information and search for objects according to their own interests.

In Digital History, Daniel Cohen and Roy Rosenzweig wrote that the growth of “the history web” has been driven as much by “grassroots historians” as by formal history institutions or credentialed professionals. Archive Team serves as an organizational website for some of this grassroots history, exposing both its promises and flaws. Describing itself as “a loose collective of rogue archivists, programmers, writers and loudmouths dedicated to saving our digital heritage,” the site provides advice, software tools, and coordination for crowd-based campaigns to archive born-digital history such as websites and social media communities that are threatened with deletion.

Archive Team was founded in January 2009 by Jason Scott in response to the closure of a number of commercial online communities like AOL Hometown that held irreplaceable user-contributed material. Scott later described creating the project “out of anger and a feeling of powerlessness, this feeling that we were letting companies decide for us what was going to survive and what was going to die.” Rather than host digital archives itself, Archive Team offers, in Scott’s words, “a really profane, rough and tumble version of a library sciences convention” comprising a wiki where volunteers can keep track of websites threatened with closure and coordinate efforts to download backups of these sites and their content before they disappear. The resulting backups are generally uploaded to the Internet Archive or other third-party repositories that make them available to the public.

Soon after Archive Team launched, Yahoo! announced in 2009 that it would be closing GeoCities and deleting the 38 million websites its users had created there since 1994. In response, Archive Team’s volunteers set out to download as many Geocities sites as possible and released them publicly on bittorrent, gaining significant publicity for the project. Since then, the growing team has also preserved parts of other shuttered or threatened sites including Friendster, Tabblo, Posterous, Google Reader, Genealogy.com, and many others.

Although Archive Team’s early projects preserved data with little standardization and demanded a fair amount of technical expertise and communication between volunteers, it has made major improvements on both counts. Volunteers today can contribute their internet bandwidth to preserve threatened websites just by downloading VirtualBox and the Archive Team Warrior, which takes care of automatically coordinating tasks with other users and downloading and processing data for Archive Team’s ongoing preservation projects. Archive Team also now uses the WARC format for its archival files, ensuring that metadata and provenance are recorded along with preserved digital files.

Despite its improved usability and data-standardization, Archive Team remains a visibly grassroots project driven by what Scott called its “three virtues: rage, paranoia, and kleptomania” toward commercial web providers who have sometimes been irresponsible stewards of their user-contributed content. The Archive Team wiki’s irreverent tone and somewhat disorganized structure may not be inviting to all volunteers. Furthermore, the group’s aggressive stance towards downloading copyrighted or private material without permission have sometimes caused backlash. On the other hand, the site’s willingness to work outside institutional norms have given it the agility to preserve material quickly in the rapidly shifting online landscape. The group’s success in preserving digital memories has garnered awards and praise from web users who would have otherwise lost their own family photos and online scrapbooks to website closures. As Scott explained, “this is why we do it: because these people had their history taken away.” Hopefully as the Archive Team matures it can better address issues of privacy and civility while continuing to empower communities to preserve their online heritage for current and future generations.