Difference between revisions of "LRMI/FAQ"

This is Creative Commons' Frequently Asked Questions on the Learning Resource Metadata Initiative, a project to develop a common education metadata vocabulary being co-led by CC.

General questions

Skip to specification specific questions, technology questions, or legal questions.

What is the Learning Resource Metadata Initiative (LRMI)?

The Learning Resource Metadata Initiative (LRMI) is a project led by Creative Commons (CC) and the Association of Educational Publishers (AEP) to establish a common vocabulary for describing learning resources. The vocabulary will be the first independently developed industry-specific framework designed to work with schema.org, the web metadata framework launched June 2, 2011 by Google, Bing, and Yahoo!, thereby improving the practical search and discovery of learning resources online. A common framework for tagging and organizing learning resources can enable further applications; thus, in order to maximize buy-in and the realization of future benefits for all learners, interoperability and transparency will be key criteria for the vocabulary and LRMI’s development process.

Goals and activities:

• Document an abstract vocabulary representing the most common descriptions of learning resources used by existing educational metadata standards (e.g., Learning Object Metadata), by online publishers of learning resources (whether a machine-readable vocabulary is used or not), and addressing the contemporary desire to link learning resources to learning outcomes (e.g., Achievement Standards Network).
• Create a concrete expression of the abstract vocabulary for use within the schema.org hierarchy. Given this deployment target and the motivation to increase discoverability, utility for enhancing search queries and results will be a desired property for each term in the abstract vocabulary.
• Create a concrete expression of the abstract vocabulary as RDF, for interoperability with other applications and existing vocabularies. This drives another desired property for abstract vocabulary terms — to mirror the semantics of existing education matadata vocabularies to the extent possible, so that explicit equivalences and refinements may be established, protecting existing investments in educational metadata made by publishers and curators of learning resources and by institutions to date.
• Liaise with search engines, learning resource publishers, communities, and repositories, and other potential distributors and consumers of education metadata (e.g., Learning Management Systems vendors, National Learning Registry) to promote adoption and impact of the vocabulary.
• Explain the impact, value, and use cases of a common education metadata vocabulary to the general public, decision-, and policy-makers.

Who are the key partners involved, and what is Creative Commons’ role?

CC is co-leading the LRMI with the Association of Educational Publishers, which includes publishers Houghton Mifflin Harcourt, McGraw-Hill Education, Scholastic, Inc., Pearson, and education technology companies SMART Technologies Inc. and Promethean. Open education organizations in addition to CC are also invested in the project, which launches with the support of the Institute for the Study of Knowledge Management in Education (ISMKE), Curriki.org, BetterLesson.org, and the Monterey Institute for Technology and Education (MITE).

CC is responsible for coordinating the work of developing the common learning resources vocabulary. This includes educating the public about the project, gathering a community of supporters, and convening a working group of content and metadata experts who will then go on to develop the common metadata vocabulary for learning resources, with initial mappings to schema.org and RDF. CC will manage the working group as a neutral party, with an eye towards developing a vocabulary that at once draws from the lessons of previous learning metadata efforts, is simple enough to represent a consensus and gain rapid adoption, is congruent with other vocabularies represented at schema.org, and is interoperable with other technologies.

Why is this happening now?

Demand for a common learning metadata vocabulary has existed for years, as past efforts demonstrate. The urgency of demand has increased due to major investments in digital learning resources, including the U.S. Department of Labor’s $2 billion TAACCCT program. Previously, CC expected to push for development and adoption of a common learning vocabulary over the next few years in order to leverage these opportunities. The recent launch of schema.org with support from the three largest search engines adds the necessary pull for a feedback loop leading to universal adoption of a common vocabulary by publishers of learning resources and a constellation of applications utilizing this rich metadata.

What are the benefits of a common learning resources framework? Who will benefit?

Creating a common metadata schema will accelerate movement toward personalized learning by publishers, content providers and learners, and help to unleash the tremendous potential of OER and online learning.

When a critical mass of learning resources are described in a machine-readable fashion with a common vocabulary, everyone benefits. Search engines can return more relevant, richer results. Learners can discover and compare learning materials pertinent to their immediate learning situation. Publishers of learning materials can have their materials surface above the mass of web pages with seemingly relevant keywords, but no useful learning materials. Education technology developers can build applications that further leverage a now well-described universe of learning materials, further increasing the value of said materials to all stakeholders, including learners, publishers, schools, governments, and the general public.

I am an OER or other educational publisher. What does LRMI mean for me?

When complete, the LRMI vocabulary can help increase the discoverability and value of your educational resources. You can also get involved in LRMI development. See questions above for more.

I am a prospective grantee of the U.S. Department of Labor Trade Adjustment Assistance Community College and Career Training (TAACCCT) program. What does the LRMI mean for me? Where can I find support to mark up my resources correctly?

We envisioned that an education vocabulary and metadata framework, and search engine leverage of these would occur during the TAACCCT grant period; this is an opportunity to make it happen more quickly. We have a funded (free for grantees) program to assist TAACCCT grantees as resources are developed and published, see http://creativecommons.org/taa-grant-program to get in touch and take advantage of this.

Will this initiative improve search and discovery for open educational resources only?

No. It will improve the discoverability and increase the value of any educational resource appropriately tagged with the vocabulary, whether open or proprietary, institutional or community-oriented, traditional or otherwise. More information, in the form of wide use of a common vocabulary, makes the market work better, allowing institutions, governments, learners, parents, and teachers to make the best, most effective choice of learning material for their particular context.

I have licensed work under CC in the past. How does this affect me?

You do not need to change your license or any metadata published with your licensed work. If you publish learning materials, you may benefit from adding metadata to your licensed works once the LRMI vocabulary is finalized. Publishers of other works may also benefit in the future as this work is applied more generally (after all, most learning resources are books, documents, media objects, and other staples of CC licensing in other domains). Early adopters are of course encouraged to participate in developing and testing the LRMI vocabulary.

How can I contribute?

Keep up-to-date and contribute to the broader conversation by following http://creativecommons.org/tag/lrmi and using the tag #lrmi on social media. If you want to get involved, join the LRMI list at http://groups.google.com/group/lrmi and introduce yourself.

What is the rough timeline on this project?

See LRMI/Timeline. Keep in mind that schema.org is a brand new initiative, while the challenges of developing a common vocabulary are well known. We will navigate the knowns and unknowns with a balance between urgency and quality. We count on your involvement and feedback to keep us on track!

Technology questions

How does LRMI relate to other education metadata initiatives?

LRMI aims to establish a common metadata schema to identify learning resources that will complement learning standards, for example those encoded in the Achievement Standards Network, including Common Core State Standards for K12 (US), as well as all other online learning vehicles. Interoperability is a key precept of LRMI. While simplicity is necessary for mass adoption and search engine implementation, mixing with and mapping to other vocabularies will be possible -- for example by mirroring the semantics of existing education matadata vocabularies (e.g., Learning Object Metadata) to the extent possible, so that explicit equivalences and refinements may be established, protecting existing investments in educational metadata made by publishers and curators of learning resources and by institutions to date.

Additionally, LRMI will begin by examining lessons from previous initiatives and real online descriptions of educational resources, whether machine-readable or not. In this, we aim to utilize the technology-agnostic aspects of the microformats process, described at http://microformats.org/wiki/process.

CC currently uses and recommends RDFa to describe its licenses and public domain tools and to express license and other information about works released under CC licenses. Doesn’t schema.org utilize microdata instead of RDFa?

In addition to using and recommending RDFa, CC was a significant contributor to the development of RDFa. We think RDFa is clearly the best technology for adding structured data to the web. Additionally, RDFa 1.1 is arguably just as simple and concise as microdata, while benefiting from years of open development, testing, and deployment. We wish schema.org had chosen to use RDFa 1.1.

However, without use by tools, metadata is nearly useless. Search engines are the fundamental tools of the web, and the three largest search engines have agreed to collaborate via schema.org on metadata vocabularies that some or all will utilize to provide enhanced search results. This ought to prove a tremendous win for structured data on the web, which CC has always envisioned as necessary for making openly licensed works maximally discoverable -- using computers to facilitate sharing and collaboration rather than attempting to suppress the same. We are incredibly excited about this potential inflection in the use of structured data on the web.

What is microdata?

Microdata is a relatively new format for adding structured data to web pages. To the casual observer, it looks very much like RDFa, and it is. The main difference, from a high level, is that microdata has no direct heritage from RDF and the Semantic Web activity. This difference brings a set of technical and political trade-offs that are too nuanced to attempt to describe here.

What is schema.org?

Schema.org is a collaboration among the largest search engines to curate a collection of vocabularies that can be used to add structured data to web pages and enhance web search results. Schema.org uses microdata as the format to embed data using these vocabularies, but in theory other formats could be used.

What about interoperability?

RDF is easily extracted from microdata and schema.org vocabularies can be expressed as RDF, so any software utilizing RDF for aggregation and data integration ought to be able to continue to do so.

Most importantly, the LRMI will develop its vocabulary as an abstract model, with schema.org and RDF representations; utilization by other metadata technologies will be eminently feasible. This would be crucial even if the situation with structured data and HTML was settled, as non-HTML and non-web applications will also realize value from a common education vocabulary.

How does microdata relate to XMP, which CC has recommended for metadata embedding in images and other file formats?

Microdata (like RDFa) is for adding metadata to web pages (technically RDFa has been used in other applications, and in theory microdata could be, but web pages are the overwhelming use case). XMP is complementary, as its use case is embedding metadata in media files. XMP uses a subset of RDF. Microdata can be mapped to RDF. The educational vocabulary LRMI develops will be usable in microdata, XMP, and many other metadata schemes.

How will schema.org play out? What does it mean for existing users of RDFa?

We can’t predict the future, of course! Schema.org seems poised to greatly increase the utility, and thus use, of structured data on the web, and with LRMI, CC will do its utmost to leverage this unique opportunity.

However, it is too early to tell how the technical format (microdata) or the primary vocabulary aspects of schema.org will impact the web. The impact could be relatively narrow, in a few domains where it turns out structured data is a big win for search engines, searchers, and publishers, and that the domain vocabularies curated at schema.org drive this. The impact could be very broad if it turns out that schema.org is a scalable model for curating common vocabularies across many valuable domains.

Given all of this, it is too early to tell (yes, this bears repeating) what schema.org means for existing users of RDFa. It is entirely possible that with adequate demand, schema.org will support RDFa as well as microdata, as Google’s pioneering Rich Snippets feature (which can be seen as a forerunner of schema.org) does.

CC will mitigate these risks first by developing the LRMI vocabulary as an abstract model applicable to microdata, RDF (hence RDFa), and other metadata technologies, and second by developing the LRMI vocabulary in an open and transparent manner that builds on knowledge from previous initiatives (see question about other initiatives above).

My website is currently marked up with RDFa from the CC license chooser or my platform generates RDFa for CC licensed works. What will I have to change?

Nothing at this point. CC deeds, which currently consume RDFa to provide copy/paste attribution and license notice markup, will continue to consume RDFa; support for microdata will be an addition.

As schema.org, LRMI, and subsequent initiatives bear fruit and best practices develop, you may wish to update your website or platform to take advantage of new capabilities.

Where can I find technical support for this change? How can I help?

For questions about and contributions specific to LRMI, please join http://groups.google.com/group/lrmi.

For questions and contributions not specific to the LRMI education vocabulary project, please join http://lists.ibiblio.org/mailman/listinfo/cc-devel.

Your questions and comments will be very valuable in guiding us in these early stages. Please introduce yourself and do not be shy. We will especially need assistance in documenting other initiatives and use cases, and in later stages, testing vocabulary drafts in real applications. We’re looking forward to your participation!

Specification specific questions

There does not appear to be a "Subject" field that takes a value such as "Social Studies" or "Algebra."

The 'about' propery that is a part of Schema.org already adequately accomplishes this goal. LRMI only adds properties that are not already covered.

There is a 'typicalAgeRange' but no term for grade level.

This is partly true. To ensure that the LRMI terms that are included with Shema.org are as internationalizable as possible it was decided to use a two prong approach.

1. Use typicalAgeRange when that information is known and available. This can even be used when the only text visible on a webpage is the grade level, as long as there is a known and relatively agreed upon mapping between grade level and age range.

An example would look like:

<p>This resource was written for an audience of <span itemprop="typicalAgeRange" content="12-13">6th graders</span></p>

2. Use of the educationalAlignment term, in combination with the alignmentObject to model the grade level using some applicable standard in your jurisdiction/country/state/etc.

To do this use the "alignmentType" term (which is a part of alignmentObject) and provide a value of "educationLevel". Then, with either using the targetUrl (if there exists a standard URL/URI for the grade level in question) or targetName to reference the grade level.

An example would look like:

<p>This resource was written for an audience of <span Itemprop="educationalAlignment" itemscope itemtype="http://schema.org/AlignmentObject"><meta itemprop="alignmentType" content="educationalLevel"><meta itemprop="targetUrl" content="http://www.example.com/standards/grades/6/" /><span itemprop="targetDescription">6th graders</span></span></p>

Accessibility is very important in education; where are the terms that deal with accessibility in LRMI?

That is true that accessibility is very important in education. It is also greatly important in other areas of online content. As such, the LRMI Technical Working Group did not address that in favor of accessibility experts proposing an extension to Schema.org.

Do all publishers have to go through their entire catalog of educational materials and align them to multiple educational standards now (eg: Common Core State Standards)?

No. There is no requirement to use any part of the LRMI specification. However, if a publisher or platform chooses to align content to an education standard (for whatever reason) they now have a way to describe that alignment faithfully using LRMI

Do all Open Educational Resource publishers and platforms have to go through all of the content to align them to multiple educational standards now (eg: Common Core State Standards)?

No. There is no requirement to use any part of the LRMI specification. However, if a publisher or platform chooses to align content to an education standard (for whatever reason) they now have a way to describe that alignment faithfully using LRMI

My publisher/school/community/platform finds the use of another property to be immensely valuable when searching for educational content, however that term is not available in LRMI or Schema.org. How can we use this property?

LRMI was started to extended Schema.org to better describe educational resources and is itself an extension, in some sense, of Schema.org. If the combination of Schema.org/LRMI does not address all of your needs when publishing your educational material online then please join the LRMI mailing list and make the case for your new term/property. If the term/property is indeed novel and of great interest in the educational community the discussion should shift to the Schema.org vocabulary discussion mailing list.

Legal questions

How will the LMRI vocabulary be licensed?

All documentation, including the vocabulary itself, will be developed and published under a CC BY license. Documentation published on schema.org is published under CC BY-SA, so it is likely the schema.org version of the vocabulary will be published on schema.org under that license as well. Any software code developed by CC in support of LMRI will be released under CC0, or an existing open source software license if required by dependencies.

In addition to using CC BY-SA for copyright, schema.org's terms include a brief patent policy. Along with the rest of the community we are evaluating this and will provide feedback as needed.