How to Become an OLAC Data Provider
The Open Language Archives Community seeks to embrace all members of the language resources community, from well-established institutional archives to individuals who want to share research results. Would-be participants thus operate in a wide variety of circumstances. In order to ensure that the barrier to participation should be as low as possible, OLAC provides three approaches to becoming an OLAC data provider. The following table summarizes the three approaches and the circumstances each is designed for. The most widely used option at the present time is the second (Static Repositories).
|Conventional Repositories||The data provider implements a software interface to an existing catalog database that permits metadata to be harvested via an HTTP-based protocol.||
|Static Repositories||The Static Repository Gateway is an OLAC service that provides the harvesting protocol for metadata repositories that are expressed as XML documents.||
|ORE||The OLAC Repository Editor is a form-based editor on the OLAC Web site that allows a user to construct a small metadata repository that will be served by the static repository gateway.||
The sections below describe each of the three approaches in more detail. A final section discusses migrating from one approach to another.
If your archive has the metadata for its collection in an existing database and if your institution has a Web server, then the best approach is to implement a software interface that gives access to the metadata records through the OLAC Protocol for Metadata Harvesting. The advantage of this approach is that you need not maintain a second version of your metadata in OLAC format; rather, the software you implement will query the existing database to get the metadata information and then dynamically return it in the format required by the OLAC standard. In order to take full advantage of the OLAC standards and services it may be desirable to add new information to the existing database (such as a way to identify languages according the standard identifiers used by OLAC), but it should not be necessary to maintain a duplicate version of any information in a different format.
In order to pursue this approach, you will need to talk with the system administrator for your institution's web server. Refer your system administrator to the OLAC Tools page for a list of existing implementations in various programming languages.
When your metadata harvesting interface has been implemented, follow the OAI registration procedure to ensure that you have correctly implemented the harvesting protocol. Once you are successfully registered as an OAI data provider, proceed to the OLAC registration procedure.
The OLAC Static Repository Gateway offers an OLAC-compliant metadata harvesting interface to a metadata repository that is expressed as an XML document. When your archive does not already have a metadata catalog in a database, then you can use an XML editor to construct such a catalog. On the other hand, if your archive does already have a metadata catalog in a database, you may want to export this data to a single XML file for harvesting by OLAC. Once you have created the required static repository file, simply post it on a publicly accessible web site.
Once you repository file is web accessible, the OLAC Static Repository Gateway is trivial to use. Construct the baseURL of the metadata harvester by appending the URL of the XML file to the URL of the gateway as follows:
http://www.language-archives.org/sr/yoursite.domain/path/yourfile.xml \_______________ _________________/ \_______________ _______________/ V V URL of static location of repository gateway your repository
For instance, appending "?verb=Identify" to the above URL will cause the gateway to execute the Identify request against your static repository file. Full details about static repositories and examples of existing static repositories are available on the OLAC Tools page. Several existing OLAC static repositories can be browsed using the OLAC Archives page. For information about available XML editors, please see the XML Software Guide, the XML Cover Pages, or free XML editors.
Once your repository file is posted at a publicly accessible URL, proceed to the OLAC registration page.
Questions about static repositories should be addressed to email@example.com.
A centralized forms-based service permits people to create and manage XML repositories. This service is provided for those who do not have the capacity to use an XML editor or to post an XML file to a local web site. For further information please see the ORE page on the LINGUIST site at: http://www.linguistlist.org/ore/. During the editing process, the metadata repository is stored in a relational database on the LINGUIST site. At any point, the repository can be exported as an XML file, and registered with OLAC to make it available for harvesting.
Questions about ORE should be addressed to firstname.lastname@example.org.
As the circumstances of your archive change over time, it is easy to migrate your implementation from one approach to another. The only constraint on migration is that you should not change the unique identifier of your repository. What will change as you migrate is the base URL for your data provider. When you have a new implementation ready to go at a different base URL, all you have to do to effect the migration is to re-register at OAI with the new base URL, and then re-register at OLAC.