Platform strategy

Mission

• Build a metadata server
• as an open platform
• that provides transparent access and useful tools
• so that developers can create more value from library, archive and museum metadata,
• and so that this metadata can be integrated into the Internet ecosystem.

Goal

• Provide computer-based, programmatic access to information about the DPLA's distributed, contributed collections of content and associated metadata.
• Provide the access, dev tools, examples, and community tools that will encourage developers to build apps that use the platform, and that will ease their path to success.

Who are the platform's clients?

1. Developers who want to build apps on top of the platform
2. Other sites/services that want to use its metadata
3. Libraries that want to adopt the software for their own local use
4. Indirectly, the end users of 1-3.

Dev principles

1. Collaborative
2. Open everywhere
3. Rapidly iterate without shame

Benchmarks of success

• The ability to ingest and make available key types of data and metadata
• A reasonable and useable integration of content across collections and media type
• Lively development community
• Community creation of innovative apps that are useful to libraries, librarians, knowledge seekers, and the public
• Integration into the Web ecology broadly and via key knowledge hubs
• The platform's software (open source) is deployed by others

What the DPLAtform is not

• It is not a front-end for end-users
• It is not itself a repository of content. (Exception: it will provide repository services for any content the DPLA may decide to *generate on its own, e.g., through digitization projects.)
• It is not itself a source of authority, recommendations, etc., although it will have metadata that will enable others to make claims of authority, recommendations, etc.
• It is not primarily a search engine

Core functionality provided to developers using the platform

1. Search metadata for items, collections, events, etc. that the platform knows about
   1. Filter on facets
   2. Support complex queries across collections
2. Maintain links out to source content
3. Add collection metadata, with tools for harmonizing/normalizing the it
4. Unify items across specific domain IDs (e.g., OCLC, OpenLib, isbn, etc.)
5. Host content created by DPLA (e.g., results of DPLA digitization projects)
6. Provide usage data for items
7. Bulk downloads of data
8. Support multiple output formats (RDF, JSON, etc.)

The platform as open source

• The software that is the DPLAtform will be available as open source to be used and altered for other projects
• The initial goal is to enable other libraries and groups to create metadata middleware for their own use.
• This will entail providing some functionality useful to local instances but not useful to the dpla instance, e.g., availability data

Types of metadata to be managed

1. Item (multi-media)
2. Item event data (circulation++)
3. Collections
4. Library holdings data from participating libraries et al.
5. User generated data
   1. ratings, reviews, comments, favorites, lists...
   2. user profile?
   3. social graph?
6. Licenses
7. Site usage
8. Pointers to content
9. Repository for DPLA-specific content

Potential demo apps

• Library analytics
• Book shelf visualizer
• Search engine
• Recommender
• Wikipedia book page integration
• Reed-Elsevier integration
• Embeddable widget that displays DPLA usage data
• Longtail browser