Connecting Learners and Museums through Educational Metadata Initiatives
Darren Milligan, University of Leicester, USA, Melissa Wadman, Smithsonian Institution, USA, James Collins, Oakland University, USA
Personalized learning involves standardizing and harnessing data being created about specific student learning strengths and weaknesses, and connecting those needs with appropriate learning content. To achieve this, galleries, libraries, archives, and museums (GLAMs), as well as scientific institutions like zoos and aquariums, which already produce open educational content, need to improve the discoverability and retrieval of their digital resources. We must develop complete learning-appropriate descriptions of what we have and share this descriptive language with users in many settings. Two of the most promising programs to address this challenge are the schema.org metadata project, called the Learning Resource Metadata Initiative, and the Learning Registry, a federally created technology infrastructure for the distribution of such metadata and the consolidation of information about its usage. We discuss the history and impacts of both programs, share our methodology for implementing and evaluating a Smithsonian project in progress, and propose recommended next steps for GLAMs.
Keywords: metadata, education, OER, evaluation, learning
We take advantage of the search and retrieval capabilities of computer networks and the Internet every day, and most of the time we easily get the kind of results we expect. In education, however, the existing search and filtering of results of these systems are inadequate. Teachers need to find educational resources that precisely meet the distinct needs of their students (Association of Educational Publishers, Knovation, Educational Systemics, 2013).
In order to have an impact with educators and make their many resources fully available, galleries, libraries, archives, and museums (GLAMs), as well as scientific institutions like zoos and aquariums, need to ensure that they and their students can easily find the resources that they search. One of the most effective ways to enable discoverability and retrieval of learning resources is participation in broad, national initiatives such as the Learning Resource Metadata Initiative (LRMI) and the Learning Registry, which are in the process of addressing that need. The Smithsonian is embarking on a large-scale project to create and implement this metadata and enable GLAM educators and technologists to do the same. A case study in progress is described in this paper to illuminate the process for participation.
2. Digital resources and education
More than ever before, teachers use the Internet to inform, support, enhance, and even enable their teaching. In a recent survey of more than two thousand educators, 92 percent say the Internet has a “major impact” on their ability to access content, resources, and materials for their teaching (Purcell, 2013). The Department of Education as well, in its National Education Technology Plan, calls for the increased development of digital experiences to enable personalized learning.
The always-on nature of the Internet and mobile access devices provides our education system with the opportunity to create learning experiences that are available anytime and anywhere. When combined with design principles for personalized learning and Universal Design for Learning, these experiences also can be accessed by learners who have been marginalized in many educational settings: students from low-income communities and minorities, English language learners, students with disabilities, students who are gifted and talented, students from diverse cultures and linguistic backgrounds, and students in rural areas (National Education Technology Plan Technical Working Group, 2012).
Personalized learning involves standardizing and harnessing data being created about specific student learning strengths and weaknesses, and connecting those needs with appropriate learning content. Once states and districts make this data available, teachers can access digital tools that offer personalized content for each student’s needs. In order to connect these needs with appropriate learning resources, content producers and publishers need to describe in more complete ways what they have. Richard Culatta (2013), director of the Office of Educational Technology for the U.S. Department of Education, sees several problems in the structure of traditional education that technology is uniquely positioned to address. These problems include the treatment of all learners with similar pedagogy, homogenous schedules, and the latency of performance indicators. Digital technology specifically addresses these issues through pace adjustment, student agency in learning approach, real-time feedback, and radically improved access to learning resources. These resources are an essential component to enabling personalized learning experiences, and access to them can be improved through the strategies outlined in this document.
Educators and students do not always locate resources using the search provided on our institutions’ websites and those of our collaborators, but rather come directly from a search engine where they have keyword-searched for their topic of interest. In fact, Google, Bing, and the other top search engines continue to be the primary way that users arrive at most Smithsonian websites, for example. Search engines deliver as much as 50 percent of users arriving to Smithsonian Education (http://smithsonianeducation.org/), the central teacher portal at the Smithsonian. As well, the ever-expanding nature of digital content on the Internet is proving to be an increasing impediment to these users finding our learning content (Waters, 2013). Users must spend too much time sorting through the seemingly endless search results. A 2013 survey of educators demonstrates this need and frustration (Winter Group, 2013).
- Nearly half (43.7 percent) of educators said they search online for instructional resources at least several times a week, and nearly a third (30.5 percent) search daily
- Two thirds of educators (64.8 percent) said they get too many “irrelevant results”
- Nearly nine in ten educators (86.6 percent) said they would be more satisfied with Internet searches if they could filter results by standard instructional criteria such as grade level, subject area, media type, etc.
3. Establishment of educational metadata standards
How do content producers ensure that educational users can quickly locate and analyze potentially crucial digital resources? The solution to this problem was to develop a standard way of tagging (the process of creating metadata) online content. This standard, now described at schema.org, was developed by Google, Bing, and Yahoo! to provide a consistent “markup schema” for general Web content. The project encouraged specialized communities (like those producing learning resources) to extend the schema to meet their own needs. LRMI emerged from a partnership between the Association of Educational Publishers and the Creative Commons (funded by the Bill & Melinda Gates Foundation and the William and Flora Hewlett Foundation), as a response to this exact problem (Redd, 2013).
The initiative (LRMI) was established to first understand the needs of the learning community and then to develop the metadata specification that they would need to accurately describe their content. In May 2013, the LRMI specification (14 metadata properties) was accepted and published into schema.org (Association of Educational Publishers, 2013a). Now, once LRMI metadata is embedded within a resource page, search engines can index this descriptive language and “recognize” learning resources. This process enables learning content to be included in the Semantic Web, an international standards movement led by the World Wide Web Consortium (2013), whose aim is, as Tim Berners-Lee and others (2001) described in a now-famous Scientific American paper, to create “an extension of the current web in which information is given well-defined meaning, better enabling computers and people to work in cooperation.”
An extension to the LRMI specification was adopted by schema.org as recently as January 18, 2014. These four “accessibility metadata properties” (Table 1) describe the accessibility features and hazards that a given resource contains. This framework, developed by the Accessibility Metadata Project (led by Benetech), further extends the matching of appropriate content to learners with specific accessibility needs (Capiel, 2014).
https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.table1_-300x236.png 300w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.table1_-380x300.png 380w" sizes="(max-width: 584px) 100vw, 584px" />
Table 1: Complete LRMI Specification version 1.1 (http://www.lrmi.net/the-specification). Note that non-educational resource-specific properties, such as title or author, are covered in other areas of schema.org. The accessibility metadata properties (http://www.a11ymetadata.org/schema-org-accepts-our-proposal/) were adopted into schema.org on January 18, 2014, but are not technically part of the LRMI specification.
4. The growing importance of standards
The LRMI specification importantly connects each learning resource to specific learning standards (via the educationalAlignment property). A multitude of standards exist for various disciplines, but many resource developers and school districts coalesce around a shared set, called the Common Core State Standards (CCSS). The CCSS is a state-led program (currently adopted by 45 states) to use a shared set of educational standards for grades K–12 in English language arts and mathematics (they also define expectations for reading, writing, speaking and listening, and language). Standards in the areas of science, world languages, and the arts are currently being developed through a similar methodology to the adopted standards, including scholarly research, understanding of college and workforce preparedness, comparisons to states and countries with high-performing students, and existing assessment frameworks (NGA and CCSSO, 2014). The LRMI specification enables content producers to directly link their resources to a specific standard, enabling educators to quickly locate targeted content. Educators indicate that alignment to specific standards is essential to their search efforts, only less important than content/subject area and grade level (Winter Group, 2013).
https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig1_-300x132.png 300w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig1_-500x221.png 500w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig1_.png 2000w" sizes="(max-width: 584px) 100vw, 584px" />
Figure 1: Survey responses from educators indicating the relative importance of descriptive metadata for learning resources (Winter Group, 2013).
The power of a learning resource coded with LRMI metadata might be best illustrated by an example. Imagine a fourth-grade teacher who is working with several struggling students on this CCSS for informational texts:
CCSS.ELA-Literacy.RI.4.4—Determine the meaning of general academic and domain-specific words or phrases in a text relevant to a grade 4 topic or subject area (NGA and CCSSO, 2010).
If the topic the students are studying is climate change, for example, a Web search for “climate change lesson plans” would yield more than two million results (as of January 30, 2014). Analyzing these results to find the right resources (at the right age level, time requirement, and relevance to standards, for example) would be beyond the capacity of most educators. Performing the same search, after LRMI has been adopted, would enable the teacher to quickly identify a handful of targeted resources that meet these highly specific needs (Gladney, 2012).
Several large publishers of learning resources, both for- and non-profit, have already begun developing LRMI metadata to describe their resources. These organizations include BrainPop, McGraw-Hill, and the Institute for the Study of Knowledge Management in Education, developers of the OER (Open Educational Resources) Commons (http://www.oercommons.org/), who have developed metadata for fifty thousand resources (Association of Educational Publishers, 2013b, 2014).
Little is published about GLAM organizations and their participation in this initiative. The Smithsonian project described below may be the first effort in this discipline.
5. Overview of the Learning Registry
LRMI provides a common language when talking about learning resources. By standardizing these terms, it becomes possible to leverage smart searches to help educators more easily find and use those resources. The Learning Registry provides the network infrastructure necessary to share these resources with each other.
The Learning Registry began in 2010 through the combined efforts of Smithsonian Secretary Wayne Clough, U.S. Department of Education Secretary Arne Duncan, and FCC Chairman Julius Genachowski. Under the current oversight of the Department of Education and Department of Defense, the Learning Registry exists as a technology infrastructure for the distribution of learning resource metadata and paradata (Duncan, 2011). The Learning Registry now hosts nearly four hundred thousand resources from a variety of public and private publishers (Learning Registry, 2014a).
At its core, the Learning Registry is a network of connected nodes. Each node holds a collection of learning resources tagged with both LRMI metadata and resource usage information, known as “paradata.” Paradata is a term for crowdsourced analytics and evaluation data. Each of these nodes hosts a piece of the total metadata and paradata available via the Learning Registry.
Nodes can perform a variety of different tasks. All nodes store learning resource data, but the node’s administrator can enable a number of other options as well. These functions include:
- Distribution Services to synchronize and copy data from other nodes.
- Publishing Services to allow nodes to send data to other nodes. This is commonly used to send data from a smaller node to the Learning Registry’s central node.
- Access Services (Obtain and Harvest) to enable users to download learning resource records from the node. These can be individual records, sets of records (called slices), or the node’s entire set of records.
- Management, Administration, and Discovery Services to support maintenance of the node.
- Broker Services (not implemented), to allow nodes to manipulate the data they are holding and receiving.
Most nodes provide all of these services and ultimately publish their data to the Learning Registry’s central node. This node, managed by the Learning Registry team, acts as a central access point for the Learning Registry’s crowdsourced data.
6. Implementing the Learning Registry
The Learning Registry does not provide a user-friendly interface for accessing resources. Websites like the California Department of Education’s Brokers of Expertise (BoE) (http://www.myboe.org/), along with tools such as Easy Publish (discussed in the publishing section below), are needed in order to interface with this data. The Learning Registry relies on public and private organizations’ efforts to build these interfaces. They hope to encourage innovation in the use of Learning Registry data and support the development of new sites and tools by keeping the Learning Registry codebase open source.
https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig2_-235x300.png 235w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig2_.png 1237w" sizes="(max-width: 584px) 100vw, 584px" />
Figure 2: A learning resource on the California Department of Education’s website, Brokers of Expertise (BoE). The page shows all LRMI metadata described in the previous section of this paper, including the contributor, source, title, description, topics, grades, and a screenshot. The data for this resource is currently stored in the BoE’s Learning Registry node, along with many other Learning Registry nodes across the nation.
The BoE site allows users to search for learning resources in the Learning Registry using a keyword search combined with filters based on LRMI metadata fields. Figure 2 shows a screenshot of a learning resource. This information is pulled directly from the BoE’s Learning Registry node, which is itself being copied by other nodes. If a change is made to the data in one node, all nodes, including the BoE’s node, will be updated. By connecting to the Learning Registry network, the BoE site will always show the most up-to-date information for a resource. This is the power of sharing resources over the Learning Registry network.
7. Publishers and paradata
Public initiatives such as the National Science Digital Library and the Federal Registry for Educational Excellence have contributed large volumes of metadata to the Learning Registry. Combined with contributions from individuals and tagging events, the Learning Registry has been able to amass nearly four hundred thousand learning resources, including metadata on lesson plans, videos, infographics, and interactives.
Because of the LRMI and its adoption by schema.org, developing metadata for resources for the Learning Registry is a well-developed process. No such standardization has yet occurred for paradata. The Learning Registry team has created a specification with guidelines for formatting paradata, but the specification allows for wide flexibility in terms of what data to collect and what to name it. Without uniform collection practices, many of the greatest potentials of the Learning Registry remain untapped.
The BoE site (Figure 2) displays paradata in orange below the resource title. The site captures views, full accesses, favorites, and bookmarks. The promise of the Learning Registry is that this paradata can be published to the central Learning Registry node along with the resource’s metadata. The paradata from BoE can then be combined with paradata from other sites. This would allow every site to capture and display up-to-date usage information regardless of what site the resource is from.
By standardizing this paradata, the Learning Registry can enable powerful functionality, such as smart recommendations and smarter searches. Publishers will also be able to track the performance of resources across any site tied into the Learning Registry. As future sites and tools continue to capture even more elaborate activities, such as ratings, reviews, and statistics about classroom use, the Learning Registry will support even more complex and powerful functionality.
8. Node considerations
The Learning Registry is built on nodes, but a node is not necessary to publish content. In fact, most organizations and virtually all individuals will only need to know one node: the central Learning Registry node. Publishing to the central node ensures that a resource will be replicated across the majority of the Learning Registry. Despite some common misconceptions, publishing to the central node does not dissociate the publisher of the metadata from the resource. Each record contains publisher metadata so that ownership of the resource’s metadata is recorded and propagated throughout the network along with all of the resource’s other metadata.
There are two special situations that may prompt an organization to construct their own node: to (1) act as a redundant copy of the central Learning Registry node or (2) build a closed network. Much of the work involved in configuring these two scenarios goes beyond the scope of this paper, but the Learning Registry’s developer Google group is a good place to start for those who have identified a need for their own node. Nodes typically run on an Amazon Machine Instance or on a virtual server running Ubuntu LTS or Windows 7 with the Learning Registry’s Python script, an NGINX server, and a CouchDB database providing the core functionality.
9. Publishing to the Learning Registry
Most organizations do not need to be familiar with nodes, only with how to publish to the Learning Registry. The official method for publishing to the Learning Registry is to use or adapt their Python script. This method requires some technical expertise. Jim Klo’s Easy Publish tool (http://jimklo.github.io/EasyPublish/) is a simple and popular alternative. The Learning Registry website contains step-by-step publishing tutorials for both of these methods, along with a list of additional third-party tools for publishers to explore (Learning Registry, 2014b, 2014c).
Publishing using the Learning Registry’s Python script makes it easy to publish metadata for up to one hundred resources at a time. This method is easiest to set up in a UNIX environment, although it will work in a Windows environment as well. The publisher must have a Python compiler installed before beginning.
To prepare for publishing, all of the resource metadata must first be converted into a format that matches the Learning Registry’s technical specification. The specification details the appropriate JSON (a lightweight data-interchange format) structure required to be compatible with the Learning Registry system. After formatting, the metadata must be signed using a GPG key pair belonging to the publisher. GPG is a free encryption/decryption suite of utilities that provide privacy and authentication for the resource’s metadata.
Once the metadata has been properly formatted and signed, it is considered to be in a digital envelope, which can then be published to the target node. Most nodes require users to submit a request for publishing credentials before resource submissions will be accepted. A more in-depth explanation of this process is available from the Learning Registry. In theory, this method can also be extended to support publishing metadata automatically.
Jim Klo’s Easy Publish is a simple browser-based form that can be used to upload metadata for individual resources or a spreadsheet (in CSV format) containing multiple rows of resource metadata. Again, the Learning Registry’s preferred limit is a maximum of 100 resources per upload. The source files for Easy Publish are available on GitHub (https://github.com/jimklo/EasyPublish) and may be modified or extended as appropriate.
To use Easy Publish, a publisher must first obtain their encoded login information. This login information is based on the publisher’s Mozilla Persona account and will be sent to the server to verify their identity. This login information is called the user’s OAuth signature and is available on a node’s OAuth management page.
After obtaining the encoded login information, the publisher simply fills out the Easy Publish form with the resource’s metadata or, alternately, uploads the relevant CSV file. The login information (OAuth signature) goes in the Credentials box along with the target node’s URL. For testing purposes, the Learning Registry node (http://sandbox.learningregistry.org/) is typically used. The production node’s main URL is http://node01.public.learningregistry.net. After submitting, the publisher will receive either a confirmation or an error report.
10. Case study (in progress): The Smithsonian’s approach to LRMI implementation
Located primarily in Washington, DC, the Smithsonian Institution is a network of nineteen museums and galleries, nine major research centers, and the National Zoo. Founded in 1846, the Institution was created using funds from the bequest of an English scientist who left his fortune to the United States to found, “at Washington, under the name of the Smithsonian Institution, an establishment for the increase and diffusion of knowledge” (Smithsonian Institution Archives, 2004). While the founding leadership of the Institution saw its role primarily as a research organization, its second secretary, Spencer Fullerton Baird, quickly took to transforming the emerging Institution into the national museum, placing greater emphasis on the diffusion aspect of its mission (Smithsonian Institution Libraries, 1998).
Within the Institution, the Smithsonian Center for Learning and Digital Access (SCLDA) is a center of excellence focused on establishing the Smithsonian as a learning laboratory for everyone. As new technologies make it possible for audiences to connect with Smithsonian researchers, collections, and educational programs like never before, SCLDA’s mission is to coordinate among the museums and offices within the Smithsonian, provide models and methods that enable learners to access everything the Smithsonian has to offer, and empower learners to explore their own interests and engage with others.
SCLDA has a long history of both developing and providing unified access to Smithsonian learning resources. Currently, on the central teacher website for the Institution, Smithsonian Education (http://smithsonianeducation.org), SCLDA provides a searchable database that catalogs more than two thousand learning resources (lesson plans, research databases, interactives, games, etc.) for the classroom, from more than thirty Smithsonian museums, research centers, and programs. This database search feature is used more than one hundred thousand times a year to locate the rich and diverse learning resources that educators from across the Smithsonian have produced.
For more than ten years, SCLDA staff have focused on making these resources discoverable and relevant through a variety of methods, such as writing descriptions and keywords, using subject categories developed by the U.S. Department of Education, and aligning resources to all individual state standards as well as CCSS. As discussed above, this descriptive metadata is structured information that describes, explains, or locates a specific item’s content (National Information Standards Organization, 2004). These efforts have made it possible to distribute Smithsonian learning resources not only through our own websites but also on those of nonprofits and other resource providers. Through cooperative agreements, SCLDA has shared metadata through the State of California’s Brokers of Expertise platform and the New York City Department of Education, as well as resource providers such as Learning.com and the Pearson Foundation.
SCLDA is embarking on an LRMI and Learning Registry initiative in collaboration with the National Museum of American History, with funding from the Smithsonian Office of the Assistant Secretary for Education and Access, as well as external funding and in-kind support from inBloom, an “independent, non-profit organization whose mission is to provide a valuable resource to teachers, students and families, to improve education” (inBloom, 2014).
The project will connect the Smithsonian to national efforts to make personalized learning a reality for every U.S. student. The project builds on several years of research and pilot projects that support the search and discovery of educational content in which SCLDA has represented the Smithsonian, including a 2012 in-depth research project on the needs of educators and a 2013/14 study on the needs of middle school students digitally interacting with Smithsonian resources (http://smithsonian-digital-learning.wikispaces.com). SCLDA is preparing to produce LRMI metadata for 2,500 Smithsonian learning resources, publish this new metadata into the Learning Registry, and build capacity for Smithsonian educators and content creators to develop metadata as they publish new digital learning resources.
11. Methodology for tagging and publishing metadata
There are a number of methodologies for metadata generation by learning resource providers. One method is to hire an external consulting specialist that focuses on this process. These consultants work with content-experienced educators (often current and former classroom instructors) to review resources and create metadata. Another method is for the learning resource provider to generate the metadata itself. This process is best undertaken by using one of the freely available, do-it-yourself tagging tools, such as the inBloom Tagger (https://tagger.inbloom.org) or the Illinois Shared Learning Environment (ISLE) tool (http://ioer.ilsharedlearning.org/Publish.aspx), both shown in Figure 3. The Smithsonian project will hire consultants to generate the initial large batch of LRMI metadata but later utilize these tagger tools for emergent resource metadata creation.
https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig3_-300x279.png 300w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig3_-322x300.png 322w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig3_.png 1108w" sizes="(max-width: 584px) 100vw, 584px" />
Figure 3: Examples of two LRMI metadata tagger applications showing a step-by-step interface for complete metadata generation: ISLE Tagger demonstrating selection and designation of CCSS (educationalAlignment), and the inBloom Tagger demonstrating selection of educational use for the resource (educationalUse). See Table 1 for the complete LRMI specification.
Once created, the LRMI metadata will be inserted into the HTML on the websites where the learning resources are housed. For the purposes of evaluation, the LRMI metadata will be implemented on the Smithsonian Education (http://smithsonianeducation.org/) and National Museum of American History’s History Explorer websites (http://historyexplorer.si.edu), as well as provided openly for reuse by others through publishing into the Learning Registry. In this way, we hope to demonstrate that a broader audience of educators will be able to quickly find and use Smithsonian resources that address specific needs and map to curriculum standards within their school districts.
https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig4_-300x289.png 300w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig4_-311x300.png 311w, https://mw2014.museumsandtheweb.com/wp-content/uploads/2014/01/milligan.fig4_.png 2000w" sizes="(max-width: 584px) 100vw, 584px" />
Figure 4: Smithsonian process for tagging and publishing learning resource metadata using the LRMI specification (based on Association of Educational Publishers, inBloom, and Educational Systemics, 2013).
12. Measuring impact
SCLDA’s LRMI project includes a robust evaluation to add to the limited knowledge base on GLAMs and the use of digital learning resources by educators and learners (Milligan et al., 2012). The base assumption of the evaluation is that by implementing LRMI, educators’ ability to discover and analyze relevant Smithsonian learning resources will be enhanced. This in turn will enable educators to better customize their students’ learning experiences. Since LRMI is a new initiative, there is little data on actual impact on enhanced metadata related to learning resources. There is data, however, that demonstrates the correlation between increased sales of publications tagged with rich metadata (Breedt & Walter, 2012). With that example in mind, the research question the Smithsonian project is focusing on is: Does LRMI enable educators to better discover, analyze, and use Smithsonian resources to create personalized learning resources for their students?
The evaluation design will include both quantitative Web analytics data, as well as qualitative data captured with educators searching for and using digital learning resources. The project’s evaluation study will investigate this question by:
- Working with a group of school educators, such as the National Museum of American History Teacher Advisory Committee, before and after LRMI implementation to document discoverability, analysis, and use of Smithsonian digital resources
- Analyzing quantitative data from Smithsonian Education and History Explorer to identify changes in access patterns of those resources with implemented LRMI metadata
- Following up with a subset of educators to understand how LRMI impacted the use of Smithsonian resources with their students
As this project addresses new goals related to digital learning resources, especially in the context of open educational resources developed by GLAMs, the evaluation will add much-needed advancement in the understanding of the usage of GLAM learning resources in the field of informal/formal education.
13. Encouraging the creation of metadata throughout the Smithsonian and beyond
While it is crucial that existing Smithsonian resources be coded and highly discoverable, the project also aims to build capacity at the Institution for educators to develop this metadata on their own for new resources. Working with collaborators inBloom, the project will provide a series of professional development opportunities for Smithsonian educators and those that work with them (Web professionals, local educators, and Smithsonian Affiliate museum educators across the country) on developing metadata for new learning resources.
Professional development opportunities will include:
- A series of training workshops to explain the purpose of LRMI and facilitate a mini-tagging workshop for Smithsonian educators and their collaborators (local educators, web professionals, etc.)
- A training video for GLAM educators to explain the value of LRMI and the process for creating and implementing LRMI metadata for their resources
- Two online webinars to explain the purpose of LRMI and facilitate a tagging workshop for Smithsonian educators, Smithsonian Affiliate educators across the country, and any GLAM professional interested in LRMI
- A written and illustrated guide to LRMI developed for GLAMs and other similar organizations that details the process for LRMI creation for open educational resources
The products developed and the evaluation findings from this project will be shared with the GLAM community and beyond with the hopes that many will be able to benefit from the Smithsonian’s efforts to implement LRMI (project wiki available at http://smithsonian-edu-metadata.wikispaces.com/).
As demonstrated in this paper, in order to enable increased discoverability and sustained usage of their digital learning resources by educators and learners, GLAMs, as well as scientific institutions, like zoos and aquariums, should become involved in the LRMI and the Learning Registry. To achieve this, different roles within each institution can focus on complementary aspects of the project.
GLAM educators can:
- Create LRMI metadata for their existing resources
- Work with Web staff to embed LRMI metadata on resource HTML pages
- Develop LRMI for emerging resources as they are developed
GLAM technologists can:
- Assist GLAM educators in utilizing tagger tools
- Work with GLAM education staff to embed LRMI metadata on resource HTML pages
- Publish LRMI metadata to the Learning Registry
- Play an active role in the LRMI and the Learning Registry communities to ensure that the needs of their institutions are well represented as these initiatives continue to develop and mature
GLAMs participating in these initiatives will help not only to ensure that their resources remain highly discoverable by their intended audiences, but also to play an important role in the development and implementation of personalized learning opportunities.
The authors wish to thank Stephanie Norby, Director, and Michelle Smith and Pino Monaco of the Smithsonian Center for Learning and Digital Access, and Carrie Kotcho of the National Museum of American History, for their support and assistance in producing this paper and in the development of this project. Special thanks is extended to Brian Ausland and Joe Hobson of Navigation North for their assistance is understanding the Learning Registry technical infrastructure. The authors also express their gratitude to Claudine Brown, the Smithsonian assistant secretary for Education and Access, as well as Karen Garrett, the Youth Access Grant program, and inBloom, for financial and in-kind support for the Smithsonian LRMI project.
Association of Educational Publishers. (2013a). LRMI is now part of Schema.org. Consulted January 30, 2014. http://www.lrmi.net/lrmi-is-now-part-of-schema-org
Association of Educational Publishers. (2013b). LRMI webinar draws enthusiastic audience. Consulted January 30, 2014. http://www.lrmi.net/lrmi-webinar-draws-enthusiastic-audience
Association of Educational Publishers. (2014). Who’s using the LRMI specification? Consulted January 30, 2014. http://www.lrmi.net/whos-using-the-lrmi-specification
Association of Educational Publishers, inBloom, & Educational Systemics. (2013). The content developers guide to the Learning Resource Metadata Initiative and Learning Registry. March. Consulted July 18, 2013. http://www.lrmi.net/wp-content/uploads/2013/03/lrmi_lr_guide.pdf
Association of Educational Publishers, Knovation, & Educational Systemics. (2013). The smart publisher’s guide to LRMI tagging. Association of Educational Publishers. March. Consulted January 30, 2014. http://www.lrmi.net/wp-content/uploads/2013/03/LRMI_tagging_Guide.pdf
Berners-Lee, T., J. Hendler, & O. Lassila. (2001). “The semantic Web.” Scientific American Magazine. May 17.
Breedt, A. & D. Walter. (2012). The link between metadata and sales. Nielsen Book Services Ltd. January 25, 2012. Consulted July 30, 2013. http://www.isbn.nielsenbook.co.uk/uploads/3971_Nielsen_Metadata_white_paper_A4(3).pdf
Capiel, G. (2014). Schema.org accepts our proposal. Accessibility Metadata Project. Consulted January 31, 2014. http://www.a11ymetadata.org/schema-org-accepts-our-proposal/
Culatta, R. (2013). Reimagining learning: Richard Culatta at TEDxBeaconStreet. Online video. TEDxTalks. January 10. Consulted January 20, 2014. http://www.youtube.com/watch?v=Z0uAuonMXrg
Duncan, A. (2011). “Launching the Learning Registry.” State Education Technology Directors Association leadership summit. Also available at http://www.ed.gov/news/speeches/launching-learning-registry
Gladney, D. (2012). “The LRMI: A common sense method for finding Common Core resources.” Edtech Digest. December 14. Consulted July 18, 2013. http://edtechdigest.wordpress.com/2012/12/14/the-lrmi-a-common-sense-method-for-finding-common-core-resources/
inBloom. (2014). About inBloom. Consulted January 29, 2014. http://www.inbloom.org/about-inbloom
Learning Registry (2014a). Home page. Consulted January 31, 2014. http://learningregistry.org
Learning Registry. (2014b). Publishers. Consulted January 31, 2014. http://learningregistry.org/publishers
Learning Registry. (2014c). Tools & apps. Consulted January 31, 2014. http://learningregistry.org/toolsapps
Milligan, D. et al. (2012). Digital Learning Resources Project, Volume I, Review of literature. Smithsonian Center for Education and Museum Studies. October. Consulted January 30, 2014. http://smithsonian-digital-learning.wikispaces.com/file/view/DLRP_Volume-I_Literature-Review.pdf
National Education Technology Plan Technical Working Group. (2012). National Education Technology Plan 2010. U.S. Department of Education. November. Consulted July 18, 2013. http://www.ed.gov/technology/netp-2010
National Governors Association Center for Best Practices (NGA) & Council of Chief State School Officers (CCSSO). (2010). Common Core State Standards (ELA-Literacy). NGA Center for Best Practices, CCSSO, Washington D.C. Consulted January 29, 2014. http://www.corestandards.org/ela-literacy/ri/4/4
National Governors Association Center for Best Practices (NGA) & Council of Chief State School Officers (CCSSO). (2014). Common Core State Standards Initiative frequently asked questions. Consulted January 29, 2014. http://www.corestandards.org/resources/frequently-asked-questions
National Information Standards Organization. (2004). Understanding metadata. Consulted July 18, 2013. http://www.niso.org/publications/press/UnderstandingMetadata.pdf
Purcell, K. (2013). “How teachers are using technology at home and in their classrooms.” Pew Internet & American Life Project. February 28. Consulted July 18, 2013. http://pewinternet.org/Reports/2013/Teachers-and-technology.aspx
Redd, B. (2013). “Introduction.” The content developer’s guide to the Learning Resource Metadata Initiative and Learning Registry. Association of Educational Publishers. March. Consulted January 30, 2014. http://www.lrmi.net/wp-content/uploads/2013/03/lrmi_lr_guide.pdf
Smithsonian Institution Archives. (2004). James Smithson & the founding of the Smithsonian. Consulted January 30, 2014. http://siarchives.si.edu/history/main_founding.html
Smithsonian Institution Libraries. (1998). Spencer Fullerton Baird, Smithsonian secretary, 1878–1887. Consulted January 20, 2014. http://www.sil.si.edu/Exhibitions/Smithson-to-Smithsonian/collection_01.html
Waters, J. K. (2013). “Sifting the data.” T.H.E. Journal. January 2013. Consulted July 18, 2013. http://online.qmags.com/TJL0113?sessionID=48518412BEE8C4B0D0C81C218&cid=2664012&eid=17935#pg15&mode1
Winter Group. (2013). LRMI Survey report August 2013 update, Ease and discoverability: Educators and publishers on the search for educational content. Association of Educational Publishers. August. Consulted January 30, 2014. http://www.lrmi.net/wp-content/uploads/2013/08/LRMI-Survey-Report-August-2013-Update.pdf
World Wide Web Consortium. (2013). Semantic Web. Consulted January 29, 2014. http://www.w3.org/standards/semanticweb/
. "Connecting Learners and Museums through Educational Metadata Initiatives." MW2014: Museums and the Web 2014. Published January 31, 2014. Consulted .