Early View
RESEARCH ARTICLE
Open Access

One schema to rule them all: How Schema.org models the world of search

Andrew Iliadis

Corresponding Author

Andrew Iliadis

Klein College of Media and Communication, Department of Media Studies and Production, Temple University, Philadelphia, Pennsylvania, USA

Correspondence

Andrew Iliadis, Klein College of Media and Communication, Department of Media Studies and Production, Temple University, 2020 N. 13th Street, Room 225, Philadelphia, PA 19122, USA.

Email: [email protected]

Search for more papers by this author
Amelia Acker

Amelia Acker

School of Information, University of Texas at Austin, Austin, Texas, USA

Search for more papers by this author
Wesley Stevens

Wesley Stevens

College of Information and Communications, School of Journalism and Mass Communications, University of South Carolina, Columbia, South Carolina, USA

Search for more papers by this author
Sezgi Başak Kavakli

Sezgi Başak Kavakli

Klein College of Media and Communication, Media and Communication Doctoral Program, Temple University, Philadelphia, Pennsylvania, USA

Search for more papers by this author
First published: 24 February 2023

Abstract

Several industry-specific metadata initiatives have historically facilitated structured data modeling for the web in domains such as commerce, publishing, social media, and so forth. The metadata vocabularies produced by these initiatives allow developers to “wrap” information on the web to provide machine-readable signals for search engines, advertisers, and user-facing content on apps and websites, thus assisting with surfacing facts about people, places, and products. A universal iteration of such a project called Schema.org started in 2011, resulting from a partnership between Google, Microsoft, Yahoo, and Yandex to collaborate on a single structured data model across domains. Yet, few studies have explored the metadata vocabulary terms in this significant web resource. What terms are included, upon what subject domains do they focus, and how does Schema.org represent knowledge in its conceptual model? This article presents findings from our extraction and analysis of the documented release history and complete hierarchy on Schema.org's developer pages. We provide a semantic network visualization of Schema.org, including an analysis of its modularity and domains, and discuss its global significance concerning fact-checking and COVID-19. We end by theorizing Schema.org as a gatekeeper of data on the web that authors vocabulary that everyday web users encounter in their searches.

1 INTRODUCTION

Metadata are central to web and internet processes in that they help organize and structure data (Greenberg, 2005; Iliadis et al., 2021; Mayernik & Acker, 2018; Pomerantz, 2015). One can merely think about how search engines or online retailers structure information in results, including menus and knowledge panels. Several industry-specific metadata initiatives have historically facilitated structured data modeling for the web in domains such as commerce, publishing, social media, and so forth (Guns, 2013). The standardized metadata vocabularies produced by these initiatives allow web developers to “wrap” information on the web to provide machine-readable signals for advertisers and user-facing content on apps and websites. Thus, metadata assist with information search and surfacing specific facts about people, places, and products (Acker, 2021). Where such efforts were once fragmented across industries, today, this landscape is changing with globally coordinated efforts to create a universal structured data model across fields. The implications of such a model have far-reaching consequences concerning the monopolization, centralization, and exploitation of data resources.1

One such project, Schema.org, started in 2011 and is the result of a partnership between Google, Microsoft, Yahoo, and Yandex—organizations that are traditionally commercial competitors but decided to collaborate on a single structured data model for the web to benefit their search products (Guha et al., 2016). Today, millions of websites and applications use the Schema.org structured data model to assist with information search and retrieval, including news organizations involved in verifying misinformation and fact-checking with structured data (Adair, 2020). The schema also feeds data to virtual assistant products such as Alexa and Google Assistant which pull structured data from the web to answer queries (Kollar et al., 2018). Such practices show the benefits of using a single structured data model for the interoperability of web data across sites and applications. Yet, scholars have not adequately attended to the conceptual model of Schema.org's metadata vocabulary and its domain coverage (i.e., what topics it includes) relative to its central role in structuring experiences of the web. In its positioning as a global web schema, Schema.org acts as a conceptual model of the world for search engines. As such, this partnership between internet companies engenders values, goals, and aspirations that have far-reaching consequences beyond searching for pizza delivery or the latest news headline.

One way to interpret Schema.org is to accept the definition provided on its website. The Schema.org website states that the project was made “to make it easier for webmasters to provide us with data so that we may better direct users to their sites.” Schema.org is a self-described “collaborative, community activity with a mission to create, maintain, and promote schemas for structured data on the internet, on web pages, in email messages, and beyond.” The website makes several other distinctions, including that the metadata “hierarchy presented on this site is not intended to be a “global ontology” of the world…it is still the case that schema.org is not intended as a universal ontology.” The website also states that “applications from Google, Microsoft, Pinterest, Yandex and others already use these vocabularies to power rich, extensible experiences” (schema.org). Figure 1 shows the beginning of the Schema.org metadata vocabulary in plain English. Figure 2 represents a metadata vocabulary item using JavaScript Object Notation for Linked Data (JSON-LD) syntax.

Details are in the caption following the image
Beginning of the Schema.org metadata vocabulary (schema.org)
Details are in the caption following the image
Example of a metadata vocabulary item represented in JSON-LD (schema.org)

As infrastructures for facilitating information interoperability, projects such as Schema.org are a less visible aspect of the web when compared to platforms like Facebook, TikTok, and Google, but information science scholars have long emphasized the necessary and often invisible power of classification standards in information infrastructures (Bowker et al., 2010; Bowker & Star, 1999; Star, 1999). Recent advancements in structured data modeling, knowledge graphs, and the reach of the internet allow companies to leverage metadata vocabularies and crawl the machine-readable data on the web as if it were a single database of facts and assertions. Such processes allegedly help these companies represent “knowledge” in their products, like the knowledge panels in search engine results and the maps and menus of service applications (Noy, Gao et al., 2019). Structured data thus increasingly provide underlying semantics and context-control related to content for developers while also “intuitively” helping users find what they are looking for on apps and websites without relying on specific keywords in their searches. For developers, such structured data modeling allows for a faster, dynamic display of information for users. For users, this data modeling allows for dynamic webbing across platforms like Instagram, Amazon, and Netflix, tracking and anticipating their next search or action. Structured data are at once both hidden and increasingly an expectation of responsive personalization design of platforms (Acker & Donovan, 2019).

Projects like Schema.org are a kind of comprehensive “crosswalk” intended to avoid the convoluted intersecting of multiple schemas, which can overburden web administrators and create confusion and can even resolve the multiple schemas problem by harmonizing web information through a central structured data project. Yet, as a global structured data model, Schema.org is a gatekeeper of structured data on the web. The schema authorizes the metadata vocabulary that web administrators and everyday web users encounter in their platforms and applications. Up to now, relatively few critical studies have explored the development and release of the metadata vocabulary terms in this significant web resource. What terms are included, upon what subject domains do they focus, and how does Schema.org represent knowledge in its conceptual model? This article presents results from our extraction and analysis of the metadata vocabulary's release history and complete hierarchy located on Schema.org's developer pages. We provide a semantic network visualization of Schema.org, including an analysis of its modularity and domains, and discuss its global significance concerning fact-checking and coronavirus (COVID-19). We end by theorizing Schema.org as a gatekeeper of data on the web that authors vocabulary that everyday web users encounter in their searches.

2 DEFINING SCHEMA.ORG

Some terminological clarification is needed to understand Schema.org's role across the web, given that the term refers to several things at once. Schema.org is each of the following: (1) a universal metadata vocabulary, (2) an administrative body with people who work for large internet companies, volunteers, and a governance structure, and (3) a website with code examples, release history, and term hierarchy. A final (4) category might be how Schema.org is applied and operationalized by web administrators and developers “in the wild.” This process is often orthogonal to its creators’ originally intended aims and purposes. Designs such as the conceptual model offered by Schema.org's metadata vocabulary are not consistently implemented in the ways that producers anticipate. In what follows, our concern is not with the implementation of Schema.org across the web in the “real world” (definition 4) but instead with the metadata vocabulary itself (definition 1). We set aside questions of its organization and governance (definition 2) and website (definition 3) for our discussion and conclusion. We acknowledge that there is frequently a disconnect between design and deployment (Iliadis & Pedersen, 2018; Liao & Iliadis, 2021), especially in the world of metadata where people design models and tools that fail to reach a critical mass of adopters or differ in their implementation (Maron & Feinberg, 2018). Such disconnects have been theorized as instances of technological de-scription where there are different conceptualizations between creators and potential users (Akrich, 1992). We thus reserve some of our broader and speculative claims concerning Schema.org's role in gatekeeping for our discussion and conclusion, and suggest additional studies that can be carried out either quantitatively with usage data or qualitatively with user studies. Indeed, studies focusing on Schema.org's use, adoption, and validation are beginning to appear (Meusel et al., 2015; Paulheim, 2015; Şimşek et al., 2018). The present work strictly concerns the metadata vocabulary that Schema.org offers, and merely theorizes the project's more prominent position on the web as a type of gatekeeper.

3 LITERATURE REVIEW

3.1 Schema.org's role in semantic platformization

Metadata vocabularies like the kind offered by Schema.org are part of a long history of implementing machine-readable semantics on the internet to enable greater data integration across platforms. Recent studies into the social significance of data integration tools like application programming interfaces (APIs) have shown that such technologies facilitate platformization (Berry et al., 2015; Bucher, 2013). Platformization has been said to entail the “extension of social media platforms into the rest of the web and their drive to make external web data ‘platform ready’” (Helmond, 2015). Concurrently, research into digital information infrastructures (such as the structured data offered by Schema.org) has developed adjacent to these platform-oriented approaches (Bowker et al., 2010; Edwards et al., 2009) with a subset focusing on online knowledge infrastructures for sharing things like research data (Edwards et al., 2013; Karasti et al., 2016) and ensuring their long-term preservation (Acker, 2019; Acker & Kriesberg, 2020). More recently, there have been calls to combine these infrastructure and platform-based frameworks to examine information exchange mechanisms on the web through digital tools for knowledge sharing (Plantin et al., 2016).

Our paper builds and extends the critical work on these platform and infrastructure studies in what has been referred to as “knowledge as programmable object” (Plantin et al., 2016) by explicitly focusing on recent advancements in structured data modeling across web platforms. As Bucher (2012, 2013) and Helmond (2015) show data portability in the context of web platforms requires a certain level of semantic annotation. This semantic interoperability is perhaps one of several defining features of the web, or what has been referred to by computer scientists as the “semantic web” (Antoniou et al., 2012; Szeredi et al., 2014). Since its inception as a global project reaching back to the beginnings of the web in the early 1990s, semantic web practitioners have privileged the status of metadata for providing fine-grained levels of contextual expressivity needed for machine-readable web data. Now web semantics are commonplace in products like knowledge panels, virtual assistants like Siri and Alexa that rely on Wikidata, and other sources that engage markup in platformized versions of knowledge representation. Schema.org thus contributes to the increasing semantic platformization of the web by enabling data integration through applications of its metadata vocabulary.

3.2 The semantic web

The semantic web is a longstanding internet project focused on an interconnected group of technologies, standards, and methods for improving data interoperability across the web (Calaresu & Shiri, 2015). Schema.org is a modern outgrowth of the semantic web project, which was originally initiated by web inventor Sir Tim Berners-Lee through his World Wide Web Consortium (W3C) standards body. W3C produces web standards such as the Resource Description Framework (RDF) (a structured data model standard).2 The Schema.org website plainly states that the data model used therein is “derived from RDF Schema” which Berners-Lee's W3C previously published. Schema.org founders and leadership also have or had close ties to W3C. These people include Dan Brickley, who runs the daily operations for Schema.org and is on the steering group as Google's representative (Brickley previously developed RDF and semantic web technology at W3C). Ramanathan Guha, also from Google, initiated Schema.org and is one of its founders (Guha previously co-edited the RDF Schema specification with Brickley). The website further states that “since April 2015, the W3C Schema.org Community Group is the main forum for collaboration” and that current Chief Operating Officer of W3C Ralph Swick (along with Vint Cerf) “helped establish the relations between Schema.org and the W3C.” Though Schema.org has close ties to the W3C it is not an official part of the W3C nor is it considered one of their standards, indicating on their website that Schema.org is not “a standards body like the W3C.”

Previous studies have discussed the semantic web before the rise of Schema.org, focusing either on historical and philosophical conceptions of the semantic web, on notable standards bodies like the W3C, or public and institutional use of semantic web technologies such as in libraries, governments, and laboratories. Few studies (McCarthy, 2017; Patel-Schneider, 2014) have attempted to examine Schema.org from critical or sociological perspectives, and those that do have not commented on the metadata vocabulary of Schema.org, nor have they provided any substantial empirical evidence of Schema.org's domain coverage and conceptual model in this era of semantic platformization.

3.3 Critical histories of the semantic web

Several studies have critically examined the social history of the semantic web and its ideal of a universal standard semantics for the internet. Guns (2013) charts its linguistic and intellectual history and argues that as features present in early web standards, linking and metadata connect the semantic web to early artificial intelligence work involving semantic networks. Adding to this history, Hallo et al. (2016) describes the concept of “linked data” (primarily understood as the second “wave” of the semantic web) and examine libraries' attempts to gather and share data on the semantic web to highlight processes for data integration. Following Guns' theoretical concerns, Hallo et al. discuss the linguistic and conceptual ontologies that underscore linked data and consider the more significant societal and practical implications of global shifts in data storage and accessibility.

Other information researchers map the fascinating conceptual “prehistory” of knowledge representation technologies on the web that have shaped and given rise to the semantic web's ontological and individualized underpinnings. Wang (2019) maps a conceptual history of domain analysis and knowledge organization on the semantic web, tracing notable scholars in semantics research. Machado et al. (2018) provide a historical concept of the semantic web, beginning in the late 1990s, through an examination of the publications of Berners–Lee. The review draws concrete definitions of linked data and related terms, offering an operationalization of web semantics within the context of data and intelligent, automated information systems.

Similarly, Swanson (2020) charts the rise of information systems and their role as new social and economic infrastructures which govern daily life for individuals and institutions alike. Reflecting on historical developments in finance, retail automation, and digital commerce over the last several decades, Swanson provides a foundation and context to understand the semantic web and its embeddedness in the everyday lives of web users. Halford et al. (2013) also critically examine semantic web tools and technologies and how these both complicate and structure current iterations of knowledge production on the web. Each of the above studies offers portions of semantic web history, underscoring a critical dimension to web semantics that we further take up in this article.

3.4 Conceptualizing digital objects

Other studies have attempted to provide a formal definition for semantic “digital objects” in the context of the semantic web, often interrogating data models such as the triples represented in RDF's modeling of subjects, predicates, and objects. Conceptualizing “digital objects” as semantic data, Hui (2016) examines how modern-day semantic systems have a social and philosophical lineage. Campbell (2009) takes Foucault's historicization of medical classification techniques and practices and applies them to semantic web technologies. The triadic relationship that emerges is posited as helpful in theorizing the semantic web's taxonomic, descriptive, and analytic dimensions, further addressing semantic search systems in creating technological rationality. Greenberg (2005) likewise explores how metadata has been operationalized and offers a reconceptualization of metadata schemes and how the human component involved in shaping digital objects poses critical considerations related to how these technologies permeate and govern aspects of social life.

The work of Kallinikos (2010) examines digital technologies and the semantics of information objects and how “information is a subset of culture” (p. 4) including “the meaning of information tokens” (p. 10). Kallinikos et al. (2010, 2013) discern “digital objects” on the web and their function by further exploring how search engines ultimately shape the way a digital object is located, authenticated, and understood, arguing that these processes are embedded in a fundamentally social context and that, as such, semantic entities that make these processes possible must take some measure of responsibility for them. McCarthy (2017) offers one of the first sociological analyses of Schema.org and the digital objects which structure the modern semantic web, and, subsequently, how knowledge is represented through semantic triples such as those identified in the RDF data model. McCarthy thus examines the “instability” that accompanies the semantic web, which largely remains invisible through search engines, algorithms, and other associated tools.

3.5 Political economies of semantic infrastructures

DeNardis (2009, 2014) and Russell (2011, 2014) detail the governance of internet infrastructure by examining the roles that institutions such as W3C play where information is controlled, manipulated, and eventually commodified by platform companies. At a lower level of abstraction, Edwards et al. (2011) offer a reconceptualization of metadata not as an outcome of fixed product search mechanisms but as a fluid and temporal process of collaboration, arguing that the very nature of metadata renders it a signifier of the unfinished and in-progress. Researchers have foregrounded these explicitly political issues related to networked semantics on the web. Waller's (2016) and Poirier's (2017, 2019) ethnographic studies examine how infrastructures on the semantic web (e.g., taxonomies, schemas, and ontologies) are built and reveal the power dynamics embedded in their creation and use. Beghtol (2005) offers a specific set of global actionable solutions for making more ethical decisions when creating and maintaining knowledge representations in information systems, arguing for more thoughtful treatments of institutionally backed systems. Douglas et al. (2018) explore the extent to which linked data provides meaningful solutions for decolonizing archival work, arguing that linked data can better represent more complex connections that can be used to include marginalized perspectives in archives.

In line with these ethical and political insights, Ford and Graham (2016a, 2016b) examine linked data and how platforms (e.g., Wikipedia and Google, respectively) utilize linked data to depoliticize cultural constructions of place while dismissing alternative voices through digital tools. Srinivasan (2013) argues that research and cultural understandings of ontologies are primarily understood through a Eurocentric perspective, asserting this needs to be de-centered for a more culturally diverse set of codes for new media. Others have examined the political contexts where ontologies are used in military, security, and surveillance contexts (Iliadis, 2018; Iliadis, 2019; Iliadis & Acker, 2022) and in their service operations.

Schema.org presents itself as a community collaboration between nonprofits, standards organizations, and corporate partners in the service of web administrators and users who will benefit from the structured data model. Yet, as a seemingly general-use structured data model, Schema.org distances itself from the label of being a global or universal ontology. Schema.org thus represents a unique opportunity to study the emergence of a worldwide metadata vocabulary that can play a significant moderating role concerning how web data are shared and displayed on major web applications and platforms. As Schema.org grows and becomes a critical infrastructure in structured data modeling across the web, the terms and vocabularies expressed through its ontology also represent what the organization considers worthy of knowledge representation. With these conceptual models in mind, we ask the following two research questions concerning how Schema.org grows its metadata vocabulary:

RQ1: Since the beginning of Schema.org's release history, what kinds of terms have been included in the Schema.org metadata vocabulary?

RQ2: What do the terms in Schema.org's metadata vocabulary tell us about Schema.org's domain coverage (i.e., the topic areas of focus)?

4 METHODS

We wanted to learn more about the terms in Schema.org's metadata vocabulary and the conceptual model that emerges from it; to answer our research questions, we obtained digital materials from the Schema.org website and developer pages. First, we scraped the documented release history of the Schema.org updates and manually created a summary of the kinds of vocabulary topics included for discussion in each release.3 We then processed the text of the complete release history using Wordlist Maker to extract unique terms from the corpus.4 We removed non-schema-related words and isolated all the schema terms contained in the release history (n = 970). The full text of the Schema.org release history, documented updates, and an alphabetical list of schema terms are available in Tables A1 and A2.

Once we obtained the release history and distilled the schema terms, we also wanted to represent the complete metadata vocabulary hierarchy in a data visualization to observe what domain clusters appear (i.e., which terms/topics are more central in the metadata vocabulary). To visualize Schema.org's ontology in a semantic network, we downloaded two CSV files (Types and Properties) containing the Schema.org ontology from the Schema.org developers page and processed them using Gephi (Bastian et al., 2009).5 Using Gephi's interaction techniques, we explored the Schema.org ontology and customized the visualization by expanding the hierarchy to display the nodes (representing vocabulary terms) and edges (representing subTypeOf relationships). After importing the CSV material to Gephi, the ForceAtlas2 layout algorithm was employed to expand the network for an improved visualization (Jacomy et al., 2014). The nodes were resized according to their degree ranking (min 100 to max 1,500). The network's modularity was calculated using Gephi's statistics tools to observe communities (i.e., clusters of topics) in the metadata vocabulary (Gephi uses the Louvain method for detecting communities in networks described in Blondel et al. [2008]). All data we collected can be accessed via the Schema.org links in the footnotes provided in this section (ours was collected as of 05/11/22). The data thus represent Schema.org's metadata vocabulary in a “snapshot” or moment in its growth rather than chronologically over time or by future projections.

5 ANALYSIS AND FINDINGS

As of this writing, the Schema.org release history contains 14 significant updates and 53 total updates ranging from release 0.91 (released April 21, 2012) to release 14.0 (released March 17, 2022). Schema.org's vocabulary currently consists of 797 Types, 1,453 Properties, 14 Datatypes, 86 Enumerations, and 462 Enumeration members. The semantic network contains 1,387 nodes and 1,344 edges in a directed graph, which we represent in Figure 3. The network's modularity is 0.925, showing that the network is highly clustered in nature, with communities numbering 73 and an average degree of 0.969. Figure 4 shows the size distribution of nodes among each modularity class. A summary of the types of terms included for discussion in each of the releases is in Table 1. The average weighted degree is 0.969 (average total of the weights of edges linked to a node) with a diameter of 5 (the maximum distance between nodes in the graph), an average path length of 2.319 (average number of steps from one random node to another), and a density of 0.001 (how many links between nodes exist compared to how many links between nodes are possible).

Details are in the caption following the image
Schema.org semantic network visualization
Details are in the caption following the image
Schema.org modularity size distribution
TABLE 1. Topics covered in the Schema.org releases
Release version (date) Topics addressed in vocabulary
14:14 (2022) Search and rescue; adult oriented goods and services; online businesses; funding
13:13 (2021) Biology and life science, biochemistry, molecular entities, and chemicals; merchant return policies, return shipping, damaged products, labels, policies; job applications and postings; fact-checking and misinformation including reviews of media and claims; casual opinion claims unrelated to news or factual records; positive and negative reviews; countries of processing and assembly
12:12 (2021) Ineligible regions; durations of media episodes; structured sizes and measurement; backorders and item availability; health aspects relating to vaccines, health access, allergies, safety, pregnancy; occupational experience and requirements
11:11.0, 11.01 (2020) Offline and online events; animated stories; prices, invoices, down payments, cleaning fees, billing start/end dates; seeking and actions related to video content; math actions and solutions; learning and educational resources; copyright and credit notices
10:10.0 (2020) Educational questions, learning resources, and courses; floorplans and layout images; energy efficiency, consumption
9:9.0 (2020) Learning resources, creative works, and quizzes; product groups and variations; boat reservations and boat trips
8:8.0 (2020) Non-profit status, organizations, and types; delivery times, shipping destinations, rates and details; citizenship, legal, and visa requirements for job postings
7:7.0, 7.01, 7.02, 7.03, 7.04 (2020) School districts; special announcements and locations; hospital reporting; hackathons; start and end times for schedules; COVID-19
6:6.0 (2020) Media collections such as images, videos, or media galleries; 3D modeling; financial aid and educational occupational programs; job posting features such as security clearance or physical abilities
5:5.0 (2019) Health aspects; features within web sites, including pages, sites, and site elements; property information related to accommodations and real estate listings such as bathrooms, floor level, and lease length
4:4.0 (2019) Real estate information, purchase offers and losing information; podcast information related to series, seasons, episodes; sport teams information, including gender of athlete; job opening, employment, and unemployment information
3:3.0, 3.1, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9 (2016–2019) Life sciences; lodging and rental information, amenities, check in times; social media posts; recipe information like cook time and preparation time; health warnings for breastfeeding, prescriptions, and health conditions; course information, enrollment, prerequisites; how to directions for structured steps of instructions; claim type and claim review features to support fact checks; legislation; tourism, tourist attractions
2:2.0, 2.1, 2.2 (2015) Automotive/car properties, engine specifications, fuel/safety features; exercise courses, exercise related diets, health and fitness actions; payment features such as due dates, actions, receipts; geographic place names for different locations (home, work, gym), including radiuses and midpoints between locations
1:1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.91, 1.92, 1.93 (2013–2015) Audiences, publishers, technical publishing; citations to scholarly sources, creative works, medical articles; actions, potential and in the past; civic services, government services; television and radio content, video games series, episodes, role names; not-for-profit offers, sales, and item conditions for used goods; reservations for meals, hotels, airlines, or trips; question and answer features for questions, answers, facts; work performance, music events, live event announcements; visual arts, editions, forms, materials, surface
0:0.91, 0.95, 0.96, 0.97, 0.98, 0.99 (2012) Family members; software applications, mobile applications, web applications; medical and health care entities; time and temporality features, timezones

The 970 unique terms extracted from the release history range across various fields, and some topics appear to be added out of necessity based on current events. There are terms related to content moderation (an evolving concern given the proliferation of material on social media sites that goes against their terms of service, as explained in Gillespie [2018] and Roberts [2019]), fact-checking and misinformation (another area of increasing importance given the recent increase in media manipulation and disinformation campaigns online, as explained in Marwick and Lewis [2017] and Benkler et al. [2018]), educational activities, and identifying media artifacts, among many others. Other than the topics that appear to be of increasing global concern, there does not seem to be consistent logic concerning the types and kinds of metadata terms discussed and included in the Schema.org model. Indeed, possible community extensions to Schema.org vocabulary are proposed and temporarily hosted on a pending ad hoc basis before they are included in the Schema.org core, as explained on the website.

These schema terms range from simple entities and relationships in which “regular” everyday web users might be interested (such as locating information about employment or education) to more complex instances where specialists and experts would likely be the beneficiaries of the terms. The range of complexity represented in the unique terms contained in the Schema.org release history indicates Schema.org's coverage and possible application areas, spanning several domains and industries. For example, designating actions on the web are described, such as acquiring license pages and applications, providing feedback policies, or giving opinions, all of which can be used by web administrators to surface such actions in search results. More specialized terms related to fact-checking and misinformation labeling are provided through “review of claim” and “review of media” terms which news organizations can use to mark up their news stories and surface warnings in the news feed. Educational programs, credentials, occupations, events, and requirements terms are included for facilitators to boost these data points in their profiles for potential students searching for information.

While the unique terms in the Schema.org release history point to areas that have recently been under development, Schema.org's current core vocabulary shows what might be considered the earliest and seemingly most important domains for the project. For example, our semantic network offers that the high-level representations in the Schema.org hierarchy include top-level domains for vocabulary items relating to the categories of Person, Organization, Place, CreativeWork, Intangible, Action, MedicalEntity, and Event. The core categories illustrate the datafication of life within several areas, from tech industries to creative work. These top-level domains are relatively abstract and less specific than the lower-level vocabulary items, clearly indicating that from the beginning, Schema.org was intended to organize its structured data as knowledge across a wide variety of domains.

Details in our semantic network show several features of the Schema.org core hierarchy, including which hierarchy terms are subclasses of higher-level domain terms and which terms belong to specific clusters of related terms. For example, the CreativeWork category in the network has a high degree of centrality and connects to media texts, production, and consumption entities. Here, creative works are understood as units or types of labor, illustrating a schema's role in shaping normative standards of cultural production and consumption (Craig & Cunningham, 2019; Napoli & Caplan, 2017). Similarly, the Place categorization privileges locations related to culture, tourism, and commerce (i.e., where people are likely to spend money and leisure time).

The Intangible category hierarchizes purchasing behaviors and commodities, a common theme among all the second-level entities (beyond the top-level entity Thing). Moreover, some items appear far less tangible than others (e.g., the difference between making a reservation, purchasing a ticket, or enumeration). This category includes sub-level items such as Brand, MerchantReturnPolicy, and MediaSubscription. What is notable here is not necessarily how well these items fit within the assigned category but the extent to which each class is defined by purchasing behaviors, cultural commodities, and creative labor. The privileging of creative work and focus on cultural labor made possible through the metadata vocabulary provided illustrates the complex webs of meaning-making within these spheres and the role that Schema.org plays in legitimizing and making those meanings accessible to the people who use them. We note here that developers themselves use their labor to mark up their content with the vocabulary and thus are also involved in work by giving their labeling labor to search companies. Scholars have researched such uneven labor practices in the context of data workers. Irani (2013) has examined unfair labor practices in the context of microworkers involved with Amazon's Mechanical Turk crowdsourcing website, and Posada (2022) has similarly examined unjust labor in the context of data annotators.

Several other interesting entities appear in the Schema.org hierarchy, including ClaimReview, which enables developers to markup news stories with fact-checks, so they appear as such in search results, and terms related to COVID-19, including CovidTestingFacility, SpecialAnnouncement, and eventAttendanceMode.

5.1 Fact-checks and misinformation

Since 2016, Schema.org appears to have released vocabulary to combat conspiracy theories and misinformation on the internet. One term, ClaimReview, enables developers to markup news stories with fact-checks to appear as such in search results. A detail of the cluster to which ClaimReview belongs in our semantic network is represented in Figure 5. Google announced the display of this feature on October 13, 2016. The use of this tool grew in connection to the fake news debates surrounding the 2016 US presidential election (Bing introduced support for fact-checks using ClaimReview on September 14, 2017). ClaimReview positions Schema.org and Google as critical arbiters of what is credible, leveraging the perception of their widespread legitimacy as a source of information.

Details are in the caption following the image
Detail of ClaimReview cluster in Schema.org semantic network

Alexios Mantzarlis, the former News and Information Credibility Lead at Google News Lab, writes that “fact-checking initiatives around the world mark up their work with ClaimReview, which allows fact-checkers to signal—and anyone online to automatically detect—that a webpage contains a fact check” (Mantzarlis, 2019). Google's Fact Check Explorer is a search feature to discover fact-checks, including those that use the ClaimReview markup. Academic researchers are also interested in ClaimReview and are involved in spreading literacy about the feature. The Duke Reporters’ Lab runs the ClaimReview Project initiative that shares information about the fact-check function (Lim, 2019). A similar project, Full Fact, is a registered charity and consists of “a team of independent fact-checkers and campaigners who find, expose and counter the harm” (see fullfact.org). Facebook and Google fund Full Fact; the charity works for these companies, including Facebook's Third-Party Fact Checking Program and Google's work on COVID-19 fact-checking.

A recent Full Fact guest blog post at Schema.org about ClaimReview discusses the link to fact-checking (Dudfield & Dodds, 2021). The post is titled “Enriching ClaimReview for Fact Checkers” and is written by Andrew Dudfield (Head of Automated Fact Checking at Full Fact) and Leigh Dodds (an open data expert). The post states that ClaimReview is one of the “hidden jewels” of the schema hierarchy because of the way it enrolls human fact-checkers. The authors also note that the Full Fact organization has been “exploring ways to revise and extend the claim review metadata to provide more detail that might enable further reuse and labeling of content, and further insights into the fact checking process.” These areas include consolidating the recommendations and examples for implementing the feature Google and Bing have included in their structured data markup documentation pages. Others describe expanding ClaimReview to include further term identifiers and linking, enriching the markup to include claims, information about errors and corrections, and citing evidence.

5.2 COVID-19 and data commons

Schema.org's vocabulary about COVID-19 is another example of a rapidly growing topic, thanks to recent current world events. In the first 2 months of the pandemic, Schema.org fast-tracked the approval of terms in the schema hierarchy to assist with information search and retrieval related to the pandemic. The first such release occurred with version 7.0 on March 17, 2020 and included the fast-track release of terms such as CovidTestingFacility, SpecialAnnouncement, and eventAttendanceMode. Material marked up with these terms can appear on Google's Fact Check page for Coronavirus, in Google and Bing knowledge panels, and is available for retrieval by apps and platforms. These schema features populate some of the rich search results that users can see, allowing them to look up information and possibly take action concerning COVID-19. Schema.org is also helping a National Science Foundation-funded study for a semantic integration platform to fight COVID-19 (Zverina, 2020). A detail of the cluster to which CovidTestingFacility belongs in our semantic network is represented in Figure 6.

Details are in the caption following the image
Detail of CovidTestingFacility cluster in Schema.org semantic network

Another initiative tied to Schema.org relevant to accessing structured COVID-19 data is Data Commons, “an open knowledge database of statistical data started in collaboration with the U.S. Census, Bureau of Labor Statistics, World Bank, and many others” (Raghavan, 2020). This project assists with finding statistics and presenting them in the knowledge panels when searchers look for something like “Covid-19 cases in Philadelphia.” Its website is “an open knowledge repository that combines data from public datasets using mapped common entities. It includes tools to easily explore and analyze data across different datasets without data cleaning or joining.” Schema.org is one way these data can map to these familiar entities in emissions, health, water, energy, education, employment, income, and other areas. Specifically, the website states that “DataCommons.org builds upon on the vocabularies defined by Schema.org.” Google officially announced that it was making Data Commons available directly in Google Search and knowledge panels as a new layer of their Knowledge Graph as of 2020. Since Google announced Dataset Search in 2020, it has indexed almost 25 million datasets; it is “a dataset-discovery tool that provides search capabilities over potentially all datasets published on the Web” and draws heavily on Schema.org (Noy, Burgess et al., 2019).

6 DISCUSSION: A GLOBAL ONTOLOGY FOR WHOM?

As seen in our semantic network, Schema.org contains several clusters that address emerging topics of key international importance, including fact-checks and COVID-19. Yet, other terms and clusters concern commerce-related topics, such as shopping, vacationing, and cultural work. However, the domain coverage is not limited to these terms and is very broad, which suggests that Schema.org is indeed a global ontology, or at least aspires to be. Which leads us to ask the following question: a global ontology for whom?

On the one hand, Schema.org might not be as “global” in its data interoperability as its creators envisioned. For example, ClaimReview might exacerbate problems related to misinformation or provide a false sense of credibility. Things can be labeled “credible” without regard to the validity of the claim; thus, it would be possible for bad actors to fact-check a claim simply by using the metadata vocabulary. Another thing to consider is that many medical terms seem applicable to privatized US healthcare systems, indicating a type of platform nationalism (Jin, 2015). On the other hand, Schema.org does seem to be approximating something like a global ontology. In answer to our research questions, we find that the terms included since the beginning of the Schema.org ontology span a wide variety of domains and that Schema.org is not a domain-specific ontology. Instead, it aims to crosswalk several domains specific to human actions and published structured data. The top-level domains are reserved for abstract entities under which more specific domain information can be located. The ontology terms, along with the information in the release history and on Schema.org's website, tell us that Schema.org does aspire to become a global ontology (if it is not already). For example, smaller domain-specific structured data initiatives have been “absorbed” since the rise of Schema.org, as their vocabularies became adopted and instantiated in Schema.org and thus were made redundant. Existing projects developing Schema.org extensions to connect their work will likely be eclipsed and outdated.

Some benefits come with using structured data like Schema.org. Certain features are directed at using structured data for social good, including vocabulary items relating to providing information about the COVID-19 pandemic. Yet, Schema.org is a double bind for web developers who are encouraged to use Schema.org to increase the findability of their content while simultaneously being victims to how search engines present the information. Google's knowledge panels which show structured data are a way of keeping traffic on their search engine. Such practices illustrate the company's attempt to monopolize information as a semantic middle layer that discourages users from visiting primary sources. Knowledge panels that encourage implicit reliance on auto-populated facts (instead of visiting “external” sites) are a significant part of the internet for users today. For example, one study showed that “62 percent of mobile searches in June 2019 were no-click” and that “people ages 13 to 21” are “twice as likely as respondents over 50 to consider their search complete” once they have seen a “knowledge panel” (Kelley, 2019).

Indeed, Schema.org's structured data model has made searching for facts and known sources easier and faster for people using search engines (in the English language). But the model also has embedded sociocultural biases towards privatized and for-profit contexts. Search engines benefit from closed platforms, cookies and user tracking, data collection from searchers, and targeted advertising in results. The result of retrieving sources and facts in a privatized space presents a kind of capitalist realism (Fisher, 2009) where every search and click has potential value to platforms, data brokers, and corporations underwritten by advertising mechanics. This reality can be laid bare if one examines the political economy of Schema.org and the influences of its schema governance, particularly how new terms and features are added (and retired) from the structured data model. Most terms in Schema.org are directed at financially oriented topics relating to logistics, purchasing, traveling, and so forth. Fewer terms focus on areas that might benefit socioeconomically underprivileged groups. For example, as far as we can tell currently, there is no clear information relating to pro bono legal or immigration services (though there are terms for general legal services), shelter information for humans (rather than pets), food drives (instead of fast food which is included), retirement homes (there is information about hotels). There are many other areas where the hierarchy is aimed at those already privileged (tourism, shopping, etc.). One of the larger subclasses is for different types of stores (including outlets, liquor, and pawn).

Schema.org's governance, according to our review of its website and email list, is primarily controlled by Google, which might explain the focus on commodities, services, and purchasing (Andrews, 2020a, 2020b). According to the project's website, it is also unclear the extent to which Microsoft, Yahoo, and Yandex still have personnel affiliated with the project. Schema.org could be more transparent and forthcoming with its governance structure and describe who benefits from its structured data. Given that Schema.org seemingly appears like a Google-run property (due to the decreasing participation of other platform companies), what, then, can web developers and administrators do so that they are not merely giving their labor to a giant platform company that will essentially “take” information from their pages to represent in semantic search results? One option might be to initiate similar or adjacent projects focusing on justice and equity, such as multiple languages or social services.

As we argued above, developers must use the schema for their content to appear in search results or risk becoming less visible. Yet, the very nature of the schema means that content from their sites can be displayed in Google's rich search results, including knowledge panels. These practices result in fewer people visiting their sites once people receive the information they need, which is already extracted and presented by Google. In this way, Schema.org may contribute to Google searches being dominated by Google information and keeping people on its product, thus substantiating claims of Google's knowledge monopoly. Mager (2012) describes how capitalist society shapes search engines through the politics of privatization, and we view such practices occuring with Google's influence over the governance of Schema.org. By engineering software that analyses structured metadata that it controls and uses to provide ontological descriptions of the world, companies like Google have designed technological infrastructure that privileges capitalistic conceptualizations of work, labor, news, life, and technology that favor its platforms, while limiting or ignoring alternatives. Through its ability to surface information sources in retrieval, Schema.org acts as a regulator and gatekeeper controlled by search engines such as Google. Researchers might also set for themselves the task of locating examples of limitations. For example, researchers may surface semantic errors in information retrieval on products like Google's knowledge panels and Alexa's answers which rely on the markup that Schema.org provides or may conduct studies to see what content areas are omitted. As we have argued elsewhere, researchers concerned with the power of metadata might also examine the implications of standards engineering in private, closed spaces and how that may influence what internet users can see, access, find, and know (Acker, 2019; Acker, 2021; Iliadis & Acker, 2021).

Lastly, Schema.org connects to projects like Wikipedia given its ability to present information alongside Wikipedia infoboxes in search results like knowledge panels and virtual assistant responses (Minhaj & Urs, 2021). Google has specified its preference for Schema.org markup and has recommended a preference for JSON-LD over other serializations (such as Microdata and RDFa). The emergence of Schema.org as the de facto metadata vocabulary on the web and JSON-LD as the preferred serialization offer opportunities for researchers to study these new advancements in the industry. For example, how and why were these models and serializations chosen, how have practitioners implemented them across the web, and what opportunities are there for studying their impact in current search and knowledge discovery paradigms?

7 CONCLUSION

Semantic infrastructures such as Schema.org assert what is possible, standard, and appropriate in web search while consolidating the datafication of information in ways that benefit companies and corporations. The Schema.org partnership represents a critical juncture where tech companies have evolved to govern how representations of labor, information, cultural commodities, and media products as programmable knowledge are understood and accessed on the web through search. In doing so, platform companies have created a hierarchy of what they think people want to know while underscoring these processes with their marketing and cultural logic. Google's close relationship to Schema.org and its use of structured data in search results also has implications for growing Google's monopoly in the information ecosystem. Further information studies/science research into structured data will make the organizational role of platform companies more transparent, specifically in knowledge management and new media. We encourage scholars to take up a renewed interest in critical search engine studies (Haider & Sundin, 2019; Rieder, 2022) to see how new gatekeepers and sociocultural biases are created in the semantic “knowledge” that is constructed and imparted to users by large platform companies.

Endnotes

  • 1 In this article, we use the terms “metadata vocabulary” and “structured data” somewhat interchangeably for the sake of readability, with the key distinction that we view all metadata vocabularies as a form of structured data but not all structured data as a form of metadata.
  • 2 There is not enough room in this article to explain the complicated history of the semantic web project; interested readers may see Allemang et al., 2020; Berners-Lee et al., 2001; Calaresu et al., 2015; Guns, 2013; Hitzler, 2021; McCarthy, 2017; Szeredi et al., 2014.
  • 3 https://schema.org/docs/releases.html.
  • 4 https://design215.com/toolbox/wordlist.php.
  • 5 https://schema.org/docs/developers.html.
  • APPENDIX

    TABLE A1. Full Schema.org release history as of May 11, 2022
    Release Overview Description
    14.0 (March 17, 2022) Version 14.0—this release includes improvements proposed via the pending section, including additions around audience restrictions for eCommerce.

    Vocabulary

    Changes in Core section:

    • Issue #3026: Fixed incorrect longitude value for empire state building in example eg-0005.
    • Issue #3059: Added gameEdition for VideoGame by analogy with bookEdition.

    Changes in pending section:

    • Issue #3052: Added a new subtype of organization to distinguish organizations that are search and rescue organizations: SearchRescueOrganization.
    • Issue #3053: Updated keywords property to clarify that comma-separated usage only applies to textual keywords. Also updated to expect this property in several other high-level types.
    • Issue #3057: Added DigitalPlatformEnumeration as a convenience enumeration for use with actionPlatform.
    • Issue #3056: Added funding as a convenient inverse of fundedItem, with the associated types for both properties alligned. Added more examples around Grant.
    • Issue #3058: Added a new subtype of ConsumeAction called PlayGameAction.
    • Issue #2989: Added enumeration AdultOrientedEnumeration for use as value for new property hasAdultConsideration. Added hasAdultConsideration for use on offer and product. Also added isFamilyFriendly for use on offer and product in addition to CreativeWork.
    • Issue #3001: Added new property returnPolicyCountry for use on MerchantReturnPolicy.
    • Issue #2915: Added new property iso6523Code for use on organization.
    • Issue #3028: Added OnlineBusiness as a subtype of organization and OnlineStore as a subtype of OnlineBusiness.
    • Issue #3025: Added CategoryCode as an expected value for category.
    • Issue #2927: Added provider to action type.

    Site improvements

    PR #2990: Update supporting python version and libraries—RDFLib 6.1, python 3.10.

    13.0 (July 7, 2021) Version 13.0—see planning pages. This release includes improvements proposed via the pending section, in particular a number of terms proposed by the bioschemas project, alongside other additions around e-commerce returns policy markup, job postings, and MediaReview markup.

    Vocabulary

    Changes in Core section:

    Changes in pending section:

    • Issue #2892: Clarifications around blog, BlogPosting and LiveBlogPosting, including better cross-referencing, and acknowledgment of related terminology (“weblog,” and “blog” in the sense of posting). For clarity, the example showing LiveBlogPosting is no longer tagged to show up as an example of blog too.
    • Issue #2862: Additions proposed by the the bioschemas community. Adds several types for basic life science concepts. A type taxon, with properties: childTaxon parentTaxon taxonRank hasDefinedTerm. A general type BioChemEntity: associatedDisease, bioChemInteraction, bioChemSimilarity, hasBioChemEntityPart, hasRepresentation, isEncodedByBioChemEntity, encodesBioChemEntity, isPartOfBioChemEntity, taxonomicRange, biologicalRole, hasMolecularFunction, isInvolvedInBiologicalProcess, isLocatedInSubcellularLocation. The following types are subtypes of this convenience type. A type MolecularEntity, with properties inChI, inChIKey, iupacName, molecularFormula, molecularWeight, monoisotopicMolecularWeight, smiles. A type ChemicalSubstance with properties chemicalComposition, chemicalRole, potentialUse. A type Gene with additional property hasBioPolymerSequence. A type protein with additional property hasBioPolymerSequence.
    • Issue #2880: Additions to merchant return policies. Implemented in #2381. Adds ReturnLabelSourceEnumeration, ReturnMethodEnumeration and several enumerated values for these. Also adds several new enumerated values to existing ReturnFeesEnumeration. Adds properties returnMethod, itemDefectReturnFees, customerRemorseReturnFees, returnShippingFeesAmount, itemDefectReturnShippingFeesAmount, customerRemorseReturnShippingFeesAmount, returnLabelSource, itemDefectReturnLabelSource, customerRemorseReturnLabelSource, restockingFee, returnPolicyCountry. A dds new type MerchantReturnPolicySeasonalOverride and new property returnPolicySeasonalOverride. Adds existing additionalProperty and itemCondition to MerchantReturnPolicy and hasMerchantReturnPolicy to offer
    • Issue #2907: Added directApply property to indicate on a JobPosting whether direct applications are supported.
    • Issue #2844: Additional detail for MediaReview description. To cross-reference (via link or embedded description) between ClaimReview and MediaReview, two new properties: associatedClaimReview, associatedMediaReview. These are defined as specializations of a common superproperty: associatedReview. To describe exact, specific, “frozen” snapshots of a potentially changing MediaObject, three new types: ImageObjectSnapshot, AudioObjectSnapshot, VideoObjectSnapshot. A new property, sha256, can be used to identify the specific file by a hash of its contents. To group together multiple elements for reference by itemReviewed, a new type: MediaReviewItem. Its property, mediaItemAppearance, can use the “snapshot” types to bundle several different appearances of the same basic media item. For the case of an image with embedded text, a new property: embeddedTextCaption (a subproperty of caption). To characterize the original context of a potentially decontextualized media item, a property originalMediaContextDescription, alongside originalMediaLink, embeddedTextCaption, interpretedAsClaim, claimInterpreter, and archivedAt to provide additional supporting information useful for fact checking and misinformation-related workflows.
    • Issue #2912: Add a statement type, parallel to claim, for simple statements such as “fun or interesting” facts, to distinguish them from more formal (e.g., fact-checkable) claims.
    • Issue #2882: Added organization as an expected value for buyer.
    • Issue #2895: Add actor to PodcastSeries.
    • Issue #991: Adding countryOfLastProcessing and countryOfAssembly, to improve compatability with GS1.
    • Issue #2911: Added startOffset, which was accidentally omitted from the SeekToAction definitions previously.
    • Issue #2825 (also #2823): Added location, startTime and endTime to InteractionCounter.
    • Issue #2832: Added positiveNotes and negativeNotes properties for review, providing a simple dedicated syntax to support pro/con lists summarizing reviews.

    Site improvements

    • Issue #2790: Schema markup validator: validator.schema.org is available for review. See documentation for details.
    • Issue #2891: Published draft Schema.org feeds specification for implementor discussion and feedback.
    • Issue #2842: Fixed many examples after review and update of ruby CI tests.

    12.0 (March 8, 2021) Version 12.0—see planning pages. This release consists primarily of additions to the pending section of schema.org, alongside some bugfixes and improvements to core terms.

    Vocabulary

    Changes in Core section:

    • Issue #1417: Fixed typo in isLiveBroadcast.
    • Issue #2786: Updated JSON-LD example in mainEntityOfPage.
    • Issue #2803: Add ineligibleRegion as expected on the MediaObject type.
    • Issue #2821: Add duration as expected on the episode type.
    • Issue #2840: Clarified definition of schemaVersion.
    • Issue #2803: (implemented in PR #2804): Added property ineligibleRegion to MediaObject type.
    • See also pending for cases where new terms required minor adjustments to core vocabulary.

    Changes in pending section:

    • Issue #2844: Updates to MediaReview reflecting community work towards clarifying a list of codes for MediaManipulationRatingEnumeration.
    • Issue #2810: Fixed bug: marginOfError should have values of type QuantitativeValue.
    • Issue #2811: Additions to support structured sizes and measurements. Implemented in #2819. Adds SizeSystemEnumeration, WearableSizeSystemEnumeration, SizeGroupEnumeration, WearableSizeGroupEnumeration, MeasurementTypeEnumeration, WearableMeasurementTypeEnumeration, BodyMeasurementTypeEnumeration, SizeSpecification and several initial enumerated values for these. Also add sizeGroup, sizeSystem, hasMeasurement, suggestedMeasurement, suggestedAge. Updated sizeproperty to include the new SizeSpecification type as a potential value. Additional expected values for valueReference: MeasurementTypeEnumeration, text and DefinedTerm.
    • Issue #2651: Add BackOrder as an enumeration member of ItemAvailability.
    • Issue #2799: Additional enumerated aspects for HealthAspectEnumeration, and clarified HowOrWhereHealthAspect for the case of vaccines. Additions: GettingAccessHealthAspect, AllergiesHealthAspect, SafetyHealthAspect, EffectivenessHealthAspect, PregnancyHealthAspect, IngredientsHealthAspect, HowItWorksHealthAspect.
    • Issue #2681: Added vocabulary around JobPosting and occupation so that we can explicitly indicate when degree-level qualification is not required. Added a boolean experienceInPlaceOfEducation property for JobPosting. Also added a new OccupationalExperienceRequirements type as an additional possible value for the existing experienceRequirements property. Added a number-valued property monthsOfExperience for use on OccupationalExperienceRequirements.

    Site improvements

    • PR #2706: Replaced TravisCI with GitHub actions for continuous integration testing of changes to the schemaorg repository.
    • Issue #2822: Updated punctuation used in breadcrumb path displayed at the top of property pages to reflect that a property is not a subtype of the property type.
    • PR #2789: Identified and corrected 80+ broken links within site. Created a localized link checking script (brokenlinkcheck.Py) for future testing possibilities.
    • PR #2800: Reorganized directory structure of repository and development environment to separate out software and test elements (into a software subdirectory) from vocabulary definition elements.
    • PR #2814: Completed process of moving Schema.org to https. This step included the move to https of vocabulary term definitions and consequent change to https of the canonical URI displayed under the term pages [more…] tag. The web site will continue to respond to both http and https URLs. Download files will continue to support both protocols.
    • PR #2815: Fixed links in release page that return a 404 response.

    11.01 (December 8, 2020) Version 11.01—see planning pages. This interim release consists primarily of bug fixes and site improvements. Site improvements
    • Fix invalid link to home page on attic home page.
    • Reinstated home pages for auto, bib, health-lifesci and meta sections.
    • Fix invalid links to section home pages on schemas page.
    • Update wording on pages for terms that are superseded, pending or retired.
    • Reinstated link to section home page for non-core terms.
    11.0 (November 30, 2020) Version 11.0—see planning pages. This release consists primarily of additions to the pending section of schema.org, alongside some bugfixes and improvements to core terms.

    Vocabulary

    Changes in Core section:

    • Issue #2733: Fixed typo in description of ReceiveAction.
    • Issue #2746: Fixed invalid JSON-LD in example #eg-0449 on VideoGame.
    • Removed redundant property definitions for upvoteCount & downvoteCount.
    • Issue #2776: Fix invalid term links identified in examples files.

    Changes in pending section:

    • Issue #1842 implemented in PR #2744: Added examples for online and mixed online & offline events. Attached to event, eventAttendanceMode, and VirtualLocation descriptions.
    • PR #2737: Removed subproperty relationship between hasCategoryCode and hasPart.
    • Issue #2482 implemented in PR #2736: Added examples for sameAs, titleEIDR, and editEIDR properties.
    • Issue #2646 implemented in PR #2719: Added new AmpStory type—“A creative work with a visual storytelling format intended to be viewed online, particularly on mobile devices.” This is intended in particular, for works designed for AMP story readers. A more general supertype could be added if there is implementation interest.
    • Issue #2712 implemented in PR #2716: Added new PriceTypeEnumeration enumeration type with values InvoicePrice, ListPrice, MSRP, MinimumAdvertisedPrice, SRP, SalePrice. Added to range of priceType property on the UnitPriceSpecification and CompoundPriceSpecification types.
    • Issue #2689 implemented in PR #2715 & PR #2758: Added new PriceComponentTypeEnumeration enumeration type with values installment, subscription, Downpayment, ActivationFee, CleaningFee, DistanceFee. Added to range of new priceComponentType property on type UnitPriceSpecification. Added new billingStart and billingDuration properties on type UnitPriceSpecification.
    • Issue #2766: Added new HyperToc, HyperTocEntry types, tocEntry, tocContinuation and utterances properties to support navigation and annotation within MediaObject content.
    • Issue #2722 implemented in PR #2732: Added SeekToAction type to be used as a value for the potentialAction property on a VideoObject type.
    • Issue #2740 implemented in PR #2752: Added SolveMathAction, for sites than can analyze a mathematical expression, potentially solving or simplifying it. Also added a related type, MathSolver.
    • Issue 1401 implemented in PR #2745: Adjusted definition of competencyRequired so that it can be used to described the skills, knowledge, ability required in order to understand a learning resource.
    • Issue #2764 implemented in #2765: Added DefinedTerm to learningResourceType and educationalUse.
    • Issue 2,659: Added copyrightNotice and creditText, to help improve interoperability between schema.org and IPTC metadata.
    • Added missing supertype (intangible) to EnergyConsumptionDetails.

    Site improvements

    • PR #2706: Improvements to underlying site build architecture including separation of vocabulary definition and display logic; move to python 3; and simplified local build and test processes. Full type hierarchy page is now more usable and accessible (collapsible navigation). Copy example to clipboard button added to examples on term definition pages.
    • PR #2760: Added structured representation of JSON-LD examples tab to examples display on term definition pages.
    • PR #2726: Removed previous RDFa term description from within html of term definition pages, replace with equivalent JSON-LD script element.
    • Issue #2747 implemented in PR #2748: Fixed invalid comment syntax in the JSON-LD element of 47 examples.

    10.0 (September 7, 2020) Version 10.0—see planning pages. This release consists primarily of additions to the pending section of schema.org, alongside some bugfixes and improvements to core terms.

    Vocabulary

    Changes in Core section:

    • Issue #737: Improve support for controlled taxonomies (e.g., SKOS) by supporting URLs into taxonomies for the keywords field: Added URL and DefinedTerm as expected values.
    • Issue #2647: SkiResort should be a subclass of resort (in addition to the existing SportsActivityLocation supertype).
    • Issue #2664: a parentItem of a comment should be a comment, not a question.
    • Issue #2662: The supertype of question should be comment, by analogy with answer.
    • Issue #881: Changed description to reviewBody in a review example.

    Changes in pending section:

    • Issue #2663: Add a property eduQuestionType for questions that act as practice problems.
    • Issue #2678: Indicated that learningResourceType is expected on LearningResource.
    • Issue #2483: Add hasCourse property to EducationalOccupationalProgram to specify which courses are components of a program.
    • Issue #2690: Add a layoutImage property to FloorPlan to clearly indicate an image that depicts the floorplan.
    • Issue #2670: Add vocabulary to express the energy efficiency of products. Adds a hasEnergyConsumptionDetails property for use on product to indicate an EnergyConsumptionDetails item which in turn has a hasEnergyEfficiencyCategory property whose values are drawn from enumerations. An enumeration class EnergyEfficiencyEnumeration with two initial subtypes, for EUEnergyEfficiencyEnumeration and EnergyStarEnergyEfficiencyEnumeration. The elements in the EU scheme are ordered, allowing energyEfficiencyScaleMax and energyEfficiencyScaleMin properties to specify the portion of the scale applicable to the relevant product category.

    Site improvements

    • Issue #2663: Updated all examples to use https://schema.org instead of http://schema.org.
    • Style guide: Added a style guide for schema authoring to the site, giving conventions for term naming and capitalization.

    9.0 (July 21, 2020) Version 9.0 see planning pages. This release consists primarily of additions to the pending section of schema.org, alongside minimal edits to core terms to integrate the new vocabulary into the larger system.

    Vocabulary

    Changes in Core section:

    • Issue #2604: Improvements around enumerations. Some code improvements to present subtyped enumerations more usefully, and addition of a StatusEnumeration type to group numerous enumerations from various kinds of status-related situations.
    • Issue #2512: Fixed definition of LockerDelivery and ParcelService to correctly define them as enumeration members of DeliveryMethod.
    • Issue #2268: Defined expected type of several properties in ExercisePlan to be QuantitativeValue instead of the incorrect value of QualitativeValue.

    Changes in health-lifesci section:

    • Issue #2268: Defined expected type of several properties in ExercisePlan to be QuantitativeValue instead of the incorrect value of QualitativeValue.

    Changes in pending section:

    • Issue #1401: Add a type LearningResource to indicate CreativeWorks that have a particular and explicit orientation towards learning, education, skill acquisition and other educational purposes.
    • Issue #2611: Add a quiz type, as a kind of LearningResource, defined as a “test of knowledge, skills and abilities.”
    • Issue #2587: Add two properties size and pattern for type product to support common product variant dimensions.
    • Issue #1797: Add a ProductGroup type to allow modeling of a group of products (variants) that vary along explicitly defined dimensions. Add properties variesBy and hasVariant for type ProductGroup to explicitly define the variant-identifying properties and the list of variants belonging to the group. Adds existing property isVariantOf to type product to indicate the ProductGroup it is a variant of. Adds property productGroupID as explicit identifier for type ProductGroup. Also adds inverse property inProductGroupWithID for type product to identify the product group it is a variant of.
    • Issue #2605: Clarify the definition of collection to make clear that it covers a broad range of kind of “collection.” Add a ProductCollection type to allow modeling of a collection of products that are commonly bundled together. Extend the existing property includesObject for use with the type ProductCollection to indicate the (type and number of) products that are member of the collection.
    • Issue #2620: Added several new examples.
    • Issue #2599: Add byMonthWeek to schedule type.
    • Issue #1755: Add BoatReservation, BoatTrip (and arrivalBoatTerminal, departureBoatTerminal properties), BoatTerminal types.

    Site improvements

    • Issue #2304: (implemented in PR #2654):Rationalization of dump files. Machine readable vocabulary definition files, available from the developers area, have been rationalized to include definitions from the core plus all sections of the vocabulary. Now download file names have been used to reflect this change. Also introduced two versions of each file, one containing http URIs the other https URIs.

    8.0 (May 1, 2020) Version 8.0 see planning pages. This release consists of additions to the pending section of schema.org.

    Vocabulary

    Changes in Core section:

    • N/A

    Changes in pending section:

    • Issue #2543: Add a nonprofitStatus property to indicate non-profit organizations, alongside an initial enumeration of non-profit organization types, NonprofitType.
    • Issue #2506: Add vocabulary to help online vendors publish information about their shipping-related policies, associatted with an offer using the shippingDetails property. Adds OfferShippingDetails and supporting properties (including deliveryTime, doesNotShip, shippingDestination, shippingLabel, shippingRate, shippingSettingsLink, transitTimeLabel). Also added supporting types ShippingDeliveryTime, DefinedRegion, PostalCodeRangeSpecification, OfferShippingDetails, ShippingRateSettings.
    • Issue #2427: Simplifies education-related markup for describing what a learning resource teaches or assesses, and its educationalLevel. Allows these properties to be used for EducationEvents.
    • Issue #2384: Add property to JobPosting for legal requirements relating to eligibility to work such as citizenship, visa or other documentation.
    • Issue #2469: Add two properties to represent EIDR identifiers: titleEIDR and editEIDR.

    7.04 (April 16, 2020) Version 7.04. See planning pages. This was a minor release, with improvements relating to the COVID-19 response effort.

    Vocabulary

    Changes in Core section:

    • Issue #2540: EducationalOrganization has an additional supertype of CivicStructure, clarifying that it can be used with announcementLocation.

    Changes in pending section:

    • Issue #2534: Improved vocabulary for describing government benefits in the context of a SpecialAnnouncement. A new governmentBenefitsInfo property can now reference a GovernmentService. The existing serviceType property can now optionally use enumerated values from GovernmentBenefitsType. This enumerates several kinds of government benefits. While the list is general purpose, our priority is to support the COVID-19 response. We anticipate adding more structure here in future to better capture per-country systems and new kinds of benefit. As part of this work we also introduce a jurisdiction property as a superproperty of legislationJurisdiction.
    • Issue #2536: Fixed a bug in the amended definition for category.
    • Issue #2626: Add a hackathon type.
    • Issue #2493: Add startTime/endTime to schedule.

    7.03 (April 2, 2020) Version 7.03. See planning pages. This was a minor release, with improvements relating to the COVID-19 response effort.

    Vocabulary

    Changes in pending section:

    • Issue #2521: Adds a document and schema definitions capturing a 1:1 representation of a CDC CSV format for hospital reporting.
    • Issue #2519: Clarifies that webFeed can be expected on the SpecialAnnouncement type.
    • Issue #2518: Updated pending draft of 3DModel to place it under MediaObject rather than clutter CreativeWork.

    7.02 (March 31, 2020) Version 7.02. See planning pages. This was a minor release, with improvements relating to the COVID-19 response effort.

    Vocabulary

    Changes in pending section:

    • Issue #2514: Added an announcementLocation property, clarifying how a SpecialAnnouncement can reference both regions and specific places.

    7.01 (March 22, 2020) Version 7.01. See planning pages. This was a minor release, with improvements relating to the COVID-19 response effort.

    Vocabulary

    Core changes

    • N/A

    Changes in pending section:

    • Issue #2500: Added a SchoolDistrict type.
    • Issue #2497: Added datePosted to SpecialAnnouncement type.

    Site improvements

    • Issue #2495: Added an example for the geographic region of a SpecialAnnouncement.

    7.0 (March 17, 2020) Version 7.0. See planning pages.

    Vocabulary

    Core changes

    • Issue #2448: Removed several largely un-used medical health properties whose names were inappropriately general: Action, background, cause, cost, function, indication, origin, outcome, overview, phase, population, purpose, source, subtype. Note that we do not remove terms casually, but in the current case the usability consequences of keeping them in the system outweighed the benefits of retaining them, even if flagged as archived/superseded.
    • Issue #1842: Introduced a VirtualLocation type, to support description of events whose location is online. Added eventAttendanceMode to clarify the current attendance mode of an event (online, offline, or a mix). Also added a new enumeration value for eventStatus for event: EventMovedOnline.

    Changes in pending section:

    • Issue #2490: Added (fast track) a SpecialAnnouncement type with several properties and supporting documentation, as a contribution to the global response to the COVID-19 coronavirus pandemic. We expect to iterate on this design after implementator and publisher feedback.
    • Issue #2460: Added properties to EducationalOccupationalProgram to support cases where it is not clear the distinction between classroom-based and work-based programs.
    • Issue #2373: Added petsAllowed for use with ApartmentComplex, and petsAllowed, amenityFeature and floorSize to FloorPlan. Created new properties for numberOfBedrooms and a new tourBookingPage.
    • Issue #2450: Added draft type MediaReview, and associated MediaManipulationRatingEnumeration with two example values AuthenticContent and MissingContext. Note that this is editorial work in progress and not a complete design. See also NiemanLab background article on this work.
    • Issue #2454: Added usageInfo, and acquireLicensePage to CreativeWork.

    Site improvements

    • Reinstated 301 redirect to https for http requests.
    • Issue #2412: Implemented in sdopythonapp submodule issue #6: Enhancements to CORS support, providing appropriate responses to HTTP OPTIONS request.

    6.0 (January 21, 2020) Version 6.0. See planning pages.

    Vocabulary

    Core changes

    Other changes in Core section:

    • Issue #637: Made itemOffered and offers fully inverseOf each other.
    • Issue #2397: Reworded datePosted to avoid tying this property to overly-specific usecase (i.e., it originally assumed usage only with JobPosting).
    • Issue #2433: Shortened definition of PreOrder from ItemAvailability enumeration.
    • Issue #2375: Added note to descriptions for offer and AggregateOffer indicating default businessFunction.
    • Issue #2309: (implemented in PR #2380): Extended rangeIncludes of datePublished to include DateTime.
    • Issue #2420: Added MusicRecording as an expected value of audio.
    • Issue #2421: Added person and organization to the domain for interactionStatistic.
    • Issue #2139: Added MediaGallery as a super-type of ImageGallery and VideoGallery, to allow for a mixed-media collection page type.
    • Issue #2382: Added BroadcastService to the domain for inLanguage.
    • Issue #1951: Added SportsEvent to the domain for sport.

    Changes in pending section:

    • Issue #2372: Noted that other kinds of thing have callSign beyond radio broadcasts, in particular vehicles/boats and non-radio broadcasts. Added vehicle, person and BroadcastService (the latter also covers the initial type, RadioBroadcastService) to callSign accordingly.
    • Issue #2384: Adds three properties to JobPosting for requirements relating to physical abilities, sensory abilities and security clearance.
    • Issue #2396: Improvements to provide more information about the employer in JobPosting. Adds properties for applicationContact, employerOverview, and allows DefinedTerm to be used with the industry property.
    • Issue #2373: Improvements around real estate description including the addition of a property accommodationFloorPlan and type FloorPlan for use when a number of accomodation entries share a common layout. Added numberOfFullBathrooms and yearBuilt. Also noted that leaseLength is expected on offer and that datePosted can be applied to RealEstateListing.
    • Issue #2389: Changes to the pending schedule type: Adds the duration property, extends the range of byDay to allow for iCal-style rule syntax and adds the scheduleTimezone property.
    • Issue #2418: Added financialAidEligible property for the types of financial aid available for an EducationalOccupationalProgram.
    • Issue #2429: Added a number of properties to EducationalOccupationalProgram to help represent information for determining how long a program typically takes, when terms typically run, and when applications are accepted. Specifically: applicationDeadline, applicationStartDate, educationalProgramMode, numberOfCredits, maximumEnrollment, termDuration, termsPerYear, timeOfDay, typicalCreditsPerTerm.
    • Issue #2405: Add guide and recommendation as new types.
    • Issue #2394: Added isResizable to 3DModel.
    • Issue #2381: Renamed ProductReturnPolicy (and related) to MerchantReturnPolicy.
    • Issue #2108: (implemented in PR #2352): Introduced new pending type PronounceableText enabling phonetic markup of text values.

    Changes in bib section:

    • Issue #2366: (implemented in PR #2367): Added example for audiobook indicating a relationship with a printed counterpart.

    Site improvements

    • Issue #2025: Corrected xml encoding value in schemaorg.Owl file to be “UTF-8”
    • Issue #1986: (implemented in PR #2378): Added test to check for duplicate entries in jsonldcontext files.
    • Issue #2277: (implemented in PR #2388): Fixed error in definition of eventSchedule property.

    5.0 (November 1, 2019) Version 5.0. See planning pages.

    Vocabulary

    Changes in pending section:

    • Issue #2372: Extended and redescribed callSign property to cover other common situations where it is applicable.
    • Issue #2358: Introduced a type WebContent to allow descriptions to abstract away from the differences between url-addressable content in form of WebSite, WebPage or WebPageElement.
    • Issue #2374: Introduced HealthTopicContent and an enumeration of “aspects” of the different topical aspects found in such content, as well as a property hasHealthAspect associating WebContent instances with those identified aspects.
    • Issue #2373: Added properties accommodationCategory, numberOfBathroomsTotal, numberOfFullBathrooms, floorLevel, and leaseLength to accommodation for detailed real-estate description. leaseLength also added to RealEstateListing
    • Issue #2322: Linked skills description to formal definitions of competencies.

    Site improvements

    • PR #2365: Split codebase that underpins the Schema.org site into a separate repository, to simplify interaction with vocabulary definition and documentation files in the schemaorg repository area.

    4.0 (October 15, 2019) Version 4.0. See planning pages.

    Vocabulary

    Changes in pending section:

    • Issue #2348: Added draft types for RealEstateListing, OfferForPurchase and OfferForLease.
    • Issue #2289: (implemented in PR #2290): Added types EducationalOccupationalProgram, WorkBasedProgram and associated properties.
    • Issue #373: (implemented in PR #2007): Added types PodcastSeries, PodcastSeason, PodcastEpisode and associated properties.
    • Issue #2110: (implemented in PR #2142): Added property subtitleLanguage to BroadcastEvent type.
    • Issue #2341: (implemented in PR #2344): Added property ineligibleRegion to ActionAccessSpecification type.
    • Issue #2242: (implemented in PR #2345): Added property gender to SportsTeam type.
    • Issue #2329: (implemented in PR #2330): Added property totalJobOpenings to JobPosting type.
    • Issue #2296: (implemented in PR #2297): Added property employmentUnit to JobPosting type.

    Changes in bib section:

    • PR #2232: Simplified example for audiobook.

    Site improvements

    • Issue #2331: (implemented in PR #2332): Fixed mobile usability issues of site—identified by Google search console.
    • PR #2312: Fixes typo for rdfs:Label URI.

    3.9 (August 1, 2019) Version 3.9. See planning pages.

    Vocabulary

    Core changes

    • Issue #2294: (implemented in PR #2295): Adding example for JobPosting.

    Changes in pending section:

    • Issue #2291: Added StatisticalPopulation, observation and supporting properties to enable statements (e.g., aggregate statistics) about sets of entities that have particular properties.
    • Issue #2300: Added ratingExplanation property to rating (expect type text): “A short explanation (e.g., one to two sentences) providing background context and other information that led to the conclusion expressed in the rating. This is particularly applicable to ratings associated with ‘fact check’ markup using ClaimReview.”
    • Issue #2289 (implemented in PR #2290): Added types for educational programs: EducationalOccupationalProgram and WorkBasedProgram plus supporting properties: hasCredential (for organization and person types), occupationalCredentialAwarded, programPrerequisites, timeToComplete, trainingSalary, and salaryUponCompletion.’

    Site improvements

    • PR #2292: Fixed typo in Jane Doe's name in example for person.
    • PR #2285: Corrections to examples identified as errors by Travis CI scripts.
    • PR #2199: Fix to grammar in description of knowsAbout.

    3.8 (July 1, 2019) Version 3.8. See planning pages.

    Vocabulary

    Core changes

    • Issue #2294: (implemented in PR #2295): Adding example for JobPosting.

    Changes in pending section:

    • Issue #2291: Added StatisticalPopulation, observation and supporting properties to enable statements (e.g., aggregate statistics) about sets of entities that have particular properties.
    • Issue #2300: Added ratingExplanation property to rating (expect type text): A short explanation (e.g., one to two sentences) providing background context and other information that led to the conclusion expressed in the rating. This is particularly applicable to ratings associated with ‘fact check’ markup using ClaimReview.
    • Issue #2289 (implemented in PR #2290): Added types for educational programs: EducationalOccupationalProgram and WorkBasedProgram plus supporting properties: hasCredential (for organization and person types), occupationalCredentialAwarded, programPrerequisites, timeToComplete, trainingSalary, and salaryUponCompletion.

    Site improvements

    • PR #2292: Fixed typo in Jane Doe's name in example for person.
    • PR #2285: Corrections to examples identified as errors by Travis CI scripts.
    • PR #2199: Fix to grammar in description of knowsAbout.

    3.7 (June 1, 2019) Version 3.7. See planning pages.

    Vocabulary

    Core changes

    • Issue #2192: Added CategoryCode to rangeIncludes for property occupationalCategory and DefinedTerm to rangeIncludes for property jobTitle.
    • Issue #2251: Extended rangeIncludes for property estimatedSalary to include MonetaryAmount and number.
    • Issue #1848: Updated definitions of addressRegion and addressLocality that a street address is in a locality and a locality is in a region.
    • Issue #2167: Fixed invalid JSON-LD in examples showing educationRequirements for person and occupation.
    • Issue #1991: Fixed incorrect links (to webschemas.org) in docs/releases.Html.
    • Issue #1993: Updated definition and provided examples for isBasedOn property.
    • Issue #2136: Fixing some typos and missing commas in example for schedule.

    Site improvements

    • Issue #2132: Added some hierarchy to the documentation page.

    3.6 (May 1, 2019) Version 3.6. See planning pages.

    Vocabulary

    Core changes

    Moved from pending to Core:

    Other changes in Core section:

    • Issue #2202: Restored estimatedSalary as a property expected on the JobPosting type.
    • Issue #1469: Corrected format of example durations in timeRequired property description.
    • Issue #1470: Corrected duration format in examples for timeRequired.
    • Issue #2188: Added service to rangeIncludes of orderedItem property.
    • Issue #2184: Added LodgingBusiness to rangeIncludes of numberOfRooms property.
    • Issue #2166: Added number formatting advice to description of number, ratingValue, value, lowPrice, highPrice, totalPrice properties.
    • Issue #2082: Fixed RDFa example for website.
    • Issue #2077: Fixed JSON-LD example for availableLanguage property.
    • Issue #2055: Updated text to correctly quote princess Leia in example for spokenByCharacter property.
    • Issue #2036: Fix broken “url” term in ItemList example.
    • Issue #2013: Fixed typo in hotels.Html.
    • Issue #2003: Improvements to descriptions of campground and CampingPitch, identifying the differences between British English and American English for these and associated terms.
    • Issue #1989: Replaced meta with link in isLiveBroadcast example.
    • Issue #1937: Added a direct link to developers page from schemas page.
    • Issue #1605: Made CatholicChurch subclass of church.

    Changes in pending section:

    • Issue #1909: Added a courseWorkload property for use with CourseInstance type.

    Site improvements

    • Issue #2054: Corrected spelling in README.md file.
    • Issue #2215: Corrected display of enumeration value terms that are more than 1 subclass away from a parent enumeration value (e.g., EvidenceLevelA).
    • Issue #2212: Various fixes to examples. Identified by linter tests.
    • Issue #2209: Updates to Travis test.
    • Issue #2233: Fixed non-disply of acknowledgement text on term pages.
    • Issue #2229: Stopped the duplication of references for some types in *-types.Csv download files.
    • Issue #2234: Fix to JSON-LD context file—properties with text in their range should not have ‘@type: “@id”'set.
    • Issue #2218: Split Travis tests into stages for easier reporting/analysis.

    3.5 (April 1, 2019) Version 3.5. See planning pages. This release moves a number of terms from pending into the Schema.org core. It also simplifies and clarifies the Schema.org extension model, reducing our emphasis on using named subdomains for topical groups of schemas. New terms introduced in pending area include improvements for describing projects, grants and funding agencies; for describing open-ended date ranges (e.g., datasets); and a substantial vocabulary for educational and occupational credentials.

    Vocabulary

    Core changes

    Moved from pending to Core:

    • Issue #1004: BroadcastFrequencySpecification, FMRadioChannel, AMRadioChannel, broadcastFrequency, broadcastFrequencyValue, broadcastFrequencyValue, hasBroadcastChannel.
    • Issue #1125: PreOrderAction.
    • Issue #1262: BedType.
    • Issue #1293: EndorsementRating.
    • Issue #2053: SpeakableSpecification, cssSelector, xpath.
    • Issue #1541: menuAddOn.
    • Issue #1670: subjectOf.
    • Issue #1689: EmployerAggregateRating, reviewAspect.
    • Issue #1698: Occupation, educationRequirements, estimatedSalary, experienceRequirements, occupationalCategory, occupationLocation, qualifications, responsibilities, skills, hasOccupation, relevantOccupation, QuantitativeValueDistribution, median, percentile10, percentile25, percentile75, percentile90, MonetaryAmountDistribution.
    • Issue #1723: FAQPage.
    • Issue #1741: MediaSubscription, ActionAccessSpecification.
    • Issue #243: WorkersUnion.
    • Issue #743: Distillery.

    Other changes in Core section:

    • Issue #2017: Added clip to expected values of video, audio properties. Also amended caption to be expected on AudioObject, documented a pattern for describing downloadable captions and their media format using another MediaObject with an encodingFormat.
    • Issue #2034: Added startTime, endTime properties as expected for MediaObject type.
    • Issue #2038: Add person as an expected type for byArtist property.
    • Issue #2112: Added slogan property, analogous to logo.
    • Issue #2104: Improved description of elevation property.
    • Issue #1030: Noted that encodesCreativeWork is inverseOf encoding.
    • Issue #2021: Added startOffset and endOffset to clip.
    • Issue #1765: Removed superseded status from and generalized descriptions for spatial and temporal properties plus expanded range to CreativeWork.

    Changes in pending section:

    • Issue #1365: In temporalCoverage, added proposed ISO 8601 notation for open-ended date ranges, for example, 1972–2001.
    • Issue #383: Added initial vocabulary towards improved support description of project-based funding, for example to use with dataset citations. See also DINGO draft, another outcome from the same Wikidata/Wikibase workshop. New types for Grant, MonetaryGrant, project, FundingAgency, FundingScheme, and ResearchProject. Developed in collaboration with the Wikidata/Wikibase community.
    • Issue #1591: Added jobLocationType property.
    • Issue #2083: Added applicantLocationRequirements property.
    • Issue #2085: Added membershipPointsEarned property.
    • Issue #2140: Added 3DModel type.
    • Issue #2109: Added callSign property, and RadioBroadcastService type.
    • Issue #2111: Added broadcastSignalModulation and broadcastSubChannel for use with BroadcastFrequencySpecification.
    • Issue #1758: Added ArchiveOrganization, ArchiveComponent types, and archiveHeld, holdingArchive properties.
    • Issue #1759: Added collectionSize and materialExtent properties.
    • Issue #1978: Educational and occupational credentials vocabulary and examples from the eocred-schema community group. Added a type EducationalOccupationalCredential and properties credentialCategory, educationalLevel, competencyRequired, recognizedBy. Minor changes to validIn, validFor, educationRequirements, educationalCredentialAwarded, and qualifications (linking EducationalOccupationalCredentials to JobPostings, occupations and courses).
    • Issue #1968: Replaced use of hasPart and isPartOf properties on trip type, with new subTrip and partOfTrip properties.
    • Issue #1375: Renamed the various properties and labels from geospatiallyXyz to geoXyz for brevity, for example, geoIntersects.
    • Issue #1448: Added manuscript, poster, drawing, and SheetMusic subtypes of CreativeWork.
    • Issue #1976: Added ShortStory subtype of CreativeWork.
    • Issue #1816: Added play subtype of CreativeWork.

    Site improvements

    • Added section to Github README discouraging large projects without prior discussion.
    • Linked Travis-CI from Github README.
    • Clarified and simplified the extension mechanism.

    3.4 (sdo-telesto; June 15, 2018) Version 3.4 (working name “sdo-telesto”). See planning pages. Clarifications around series. Introduced an example showing ClaimReview of a clip within a larger media object. A lot of small fixes and maintenance edits. Numerous additions to the pending area, including employment and occupation vocabulary, a claim type to help explore richer modeling around fact-checks, tourism-related vocabulary. Other pending area improvements include amendments to the legislation vocabulary, trust project additions, and the proposal of an inverse of the about property, subjectOf.

    Vocabulary

    Core changes

    • Issue #1902: Improvements to HowTo examples. See also #1895.
    • Issue #1761: Changed mis-abbreviation of Friday as “Fri” to “Fr.” Per the guidelines for openingHours, Friday should be abbreviated as “Fr.” In one set of examples, it was abbreviated as “Fri.”
    • Issue #1820: Hyperlink to new, recommended DOI resolver.
    • Issue #1894: Added reference to cryptocurrencies in description for “paymentAccepted” and “currenciesAccepted” properties. For consistency, also added to description of properties: “discountCurrency,” “price,” “priceCurrency,” “currency.” Updated the description additions to reference ISO 4217 formats, cryptocurrency Ticker symbols, LETS and other currency types including single inline examples for each.
    • Issue #1090: Clarified series, CreativeWorkSeries, EventSeries. The series type is now intangible, rather than having an inappropriate CreativeWork supertype. We have two subtypes of it, CreativeWorkSeries (successor to the original definition of series, which was too generally named), plus the proposed EventSeries type (currently in the pending area). These domain-specific series types also have corresponding supertypes in CreativeWork and event. See also issue #447.
    • Issue #1155: Clarified that encodingFormat, fileFormat mean the same thing, converged their definitions, and made encodingFormat the preferred term.
    • Issue #1824: Fixed typos in tickerSymbol.
    • Issue #1833: Added expected type ItemList to acceptedAnswer, suggestedAnswer.
    • Issue #1799: Fixed an error in an example for question type.
    • Issue #1686: Added an extra ClaimReview example based on a TV clip.
    • Issue #1762: Added schema mapping declarations indicating that “name” is a subproperty of rdfs:Label, and equivalentProperty of dcterms:Title.
    • Issue #1729: Improvements around HowTo, including declaration of CreativeWork as its supertype, and introduction of “step” property instead of “steps.”
    • Issue #1732: Changed HowToSection and HowToStep (which support the HowTo type) to have CreativeWork as their supertype.
    • Issue #1723: Introduced FAQPage, which represents a WebPage presenting one or more “frequently asked questions.”
    • Issue #1935: Extend the contentRating property to also expect the rating type.
    • Issue #2020: Extend the byArtist property to also expect the person type.

    Changes in pending section

    • Issue #1670: Add a subjectOf to thing as an inverse property of about.
    • Issue #1689: Add EmployerAggregateRating as a subtype of AggregateRating to differentiate ratings for the entity as an employer and the reviewAspect property to review and rating to allow authors to specify the facet or relevant part of being reviewed.
    • Issue #1698: Add a occupation type with the supporting properties, many of which existed for JobPosting. A Person's occupation can be specified with hasOccupation. The Occupation's estimatedSalary is expressed as a MonetaryAmountDistribution, a subtype of the new QuantitativeValueDistribution type.
    • Issue #1787: Narrowed the range of legislationJurisdiction to AdministrativeArea rather than place. Broadened the range of legislationIdentifier to include URL, to match with the description of the property.
    • Issue #1815: Added proposal for TouristDestination, TouristTrip, and trip plus examples.
    • Issue #1886: Add basic sdPublisher and sdDatePublished (and in #1866, sdLicense) markup, for example, for when structured data created elsewhere.
    • Issue #1866: Added a proposed sdLicense property.
    • Issue #1776: Introduced DefinedTerm as a super type of the existing proposed term CategoryCode, to support non-categorical controlled terms.
    • Issue #1828: Added a claim type, to make explicit a concept suggested in the ClaimReview fact-checking markup.
    • Issue #1773: Add a MediaSubscription type with the supporting properties to describe a subscription for viewing media. Also added ActionAccessSpecification to allow authors to specify when a ConsumeAction requires a subscription or a purchase.
    • Issue #1744: Improvements around legislation vocabulary. Changed definition of “legislation” to make it clear that it is a document. Deleted numerous inverse properties. Fixed range on legislationType (CategoryCode). Added legislationJurisdiction as a subproperty of spatialCoverage. Fixed definitive into DefinitiveLegalValue to align with other values. Expanded the definition of legislationIdentifier to clarify that it can apply to string or URIs. Fixed and adjusted the examples accordingly.
    • Issue #1709: Added an occupation type and examples, to represent professions.
    • Issue #1389: Updated SpeakableSpecification-related properties, “xpath” and “cssSelector” to explain how they relate to “WebPageElement.” Allowed both properties to be expected on that type.
    • Issue #1525: Improvement suggested by trust project, including addition of AskPublicNewsArticle, ownershipFundingInfo and diversityStaffingReport properties, editorial tweaks to dateline property. Second phase of additions included knowsAbout, knowsLanguage proposals (#1688.)
    • Issue #1950: Added CorrectionComment and a property correction to point to it, as suggested by the trust project.

    Changes in other named extensions

    • Issue #1882: Fix to OccupationalTherapy—made OccupationalTherapy a simple class instead of a typeof schema:MedicalSpecialty
    • Issue #1677: Added (and #1900 improved) autos documentation.

    Site improvements

    • Issue #1634: Aliased “rdf:HTML” to “HTML” in JSON-LD context definitions.
    • Issue #1879: Fix to duplicate and missing properties listings in the CSV outputs for types and enumeration values (see also #1882).
    • Issue #1878: Improvements towards a more responsive HTML design, thanks to Dan Scott.
    • Issue #1857: Updated site search function to use Google custom search v2 API.
    • Issue #1646: Updated various links to use https when referring to this site.
    • Issue #1722: Continuous integration tests now check RDFa syntax.
    • Issue #1946: Linked experimental D3/RDFS and OWL representations from the developers page.

    3.3 (sdo-enceladus; August 14, 2017)

    Version 3.3 (working name “sdo-enceladus”). See planning pages, specifically #1569 for details.

    This release introduces pending types around review, NewsArticle and article that allow more fine-grained distinctions to be made. It adds a TouristAttraction type to the core, and a draft PublicToilet type as a pending proposal. It adds a HowTo vocabulary that generalizes recipe-like structured instructions, and adds some useful properties to EmailMessage. It also adds a substantial vocabulary for describing legislation to the pending section.

    Vocabulary

    Core changes

    • Issue #1459: Improvements to TouristAttraction. Add a new property touristType to TouristAttraction, “attraction suitable for type(s) of tourist. For example, Children, visitors from a particular country, etc.” Extend the domain of availableLanguage to TouristAttraction, of isAccessibleForFree to place, adapting the definitions accordingly. Add new property publicAccess to place. “A flag to signal that the place is accessible by public visitors.”
    • Issue #1525: Expanded definition of NewsArticle from “A news article” and added a news overview document introducing some proposed additional terms, and discussing the relationship between news and related markup for education/learning, TV, radio, and so forth.
    • Issue #647: Add a HowTo type, building upon and generalizing the existing recipe vocabulary. This introduces estimatedCost, steps, supply and tool as new properties, and re-uses performTime, prepTime, totalTime. We also introduce yield, which generalizes the existing recipeYield, and note that performTime generalizes the existing cookTime. Supporting types HowToItem (with a requiredQuantity property), HowToSupply and HowToTool, HowToSection and HowToStep (a kind of ItemList) are also provided, to allow more explicitly structured steps to be described. This effort has had many contributors over the years—thanks to all who have helped!
    • Issue #1590: Add bccRecipient, ccRecipient, and toRecipient for use with EmailMessage.
    • Moved to Core from pending:

      • None proposed.

    Additions to pending section

    • Issue #1457: Add a schedule type, with supporting properties: repeatFrequency, repeatCount, byDay, byMonth, byMonthDay, exceptDate, eventSchedule. These terms are used alongside the existing startTime and endTime properties. The main purpose of this new vocabulary is to capture recurring events in a manner compatible with a subset of the iCalendar approach.
    • Issue #1156: Add legislation, LegalForceStatus, LegislationObject, LegalValueLevel and other supporting terms. These terms are based on the European legislation identifier (ELI) ontology and the work of the ELI taskforce. ELI is an initiative of some national legislation publishers endorsed by EU countries and institutions, to identify, describe and link legislation on the web.
    • Issue #1624: Add a draft PublicToilet type to the pending section.
    • Issue #1577: Add estimatedSalary to JobPosting.
    • Issue #1589: Add CriticReview and UserReview to distinguish common subtypes of review.
    • Issue #1576: Add EmployerReview to indicate reviews that are by a current or former employee, and that are reviews of that organization's role as an employer.
    • Issue #1525: Add subtypes of NewsArticle: ReportageNewsArticle, AnalysisNewsArticle, OpinionNewsArticle, ReviewNewsArticle, BackgroundNewsArticle to distinguish common kinds of NewsArticle. Also add a subtype of article, AdvertiserContentArticle and a news overview document. Added a subtype of organization, NewsMediaOrganization, alongside several “best practice”-oriented properties: actionableFeedbackPolicy, correctionsPolicy, diversityPolicy, ethicsPolicy, masthead, missionCoveragePrioritiesPolicy, unnamedSourcesPolicy, verificationFactCheckingPolicy (apart from ethicsPolicy and diversityPolicy, these are all sub-properties of publishingPrinciples). These changes were based in large part upon the work of the trust project.
    • Issue #1437: Add SatiricalArticle type.

    Site improvements

    • Issue #1582: Added dedicated documentation page describing background and examples for banking and finance terms.

    3.2 (sdo-callisto; march 23, 2017)

    Version 3.2 (working name “sdo-callisto”). See planning pages, specifically #1292 for details.

    This release introduces new core vocabulary including mechanisms for describing fact-check reviews, courses, and menus. It also introduces a number of newly-proposed terms via the “pending” area, to encourage discussion, implementation feedback and other improvements.

    Vocabulary

    Core changes

    • Issue #1209: Added maximumAttendeeCapacity and remainingAttendeeCapacity for describing places and events.
    • Issue #1361: Broadened about property to be applicable on events.
    • Issue #1364: Broadened genre property to be applicable on BroadcastChannel.
    • Issue #1488: Added identifier as a common super-property of numerous identifying properties.
    • Issue #1288: Added types menu, MenuItem, MenuSection, and properties hasMenuItem, hasMenuSection. A proposed menuAddOn property has also been added to pending for further design discussion and implementor feedback. Introduced hasMenu property that supersedes menu, both for consistency with the other new menu properties, and to avoid a property and type having essentially the same name.
    • Moved to Core from pending:

      • Issue #195: Added course-description terms: Course, courseCode, coursePrerequisites, hasCourseInstance, CourseInstance, courseMode, instructor. Thanks to the schema course extension community group at W3C, initiated by LRMI and led by Phil barker.
      • Issue #1061: Migrated ClaimReview and claimReviewed for fact-checking reviews into Core section.
      • Issue #1100: Added accessMode, accessModeSufficient, accessibilitySummary properties to support accessible digital publishing. This proposal came from the digital publishing community, and specifically from the Epub 3.1 accessibility working group.

    Additions to pending section

    • Issue #1541: Add a menuAddOn property (originally proposed with other menu-related terms).
    • Issue #1125 added PreOrderAction (a subtype of TradeAction)
    • Issue #1253: Added several terms proposed by the FIBO project and its community group.

      • Types: BrokerageAccount, ExchangeRateSpecification, InvestmentFund, MoneyTransfer, MortgageLoan, RepaymentSpecification.
      • Properties: accountMinimumInflow, accountOverdraftLimit, amount, bankAccountType, beneficiaryBank, cashBack, contactlessPayment, currency, currentExchangeRate, domiciledMortgage, downPayment, earlyPrepaymentPenalty, exchangeRateSpread, floorLimit, gracePeriod, loanMortgageMandateAmount, loanPaymentAmount, loanPaymentFrequency, loanRepaymentForm, loanType, monthlyMinimumRepaymentAmount, numberOfLoanPayments, recourseLoan, renegotiableLoan.

    • Issue #1262: Added BedType.
    • Issue #1293: Added EndorsementRating.
    • Issue #1375 added GeospatialGeometry and some draft definitions for associated properties, an informal collaboration with the W3C spatial data on the web working group. Initial properties are: geoContains, geoCoveredBy, geoCovers, geoCrosses, geoDisjoint, geoEquals, geoIntersects, geoOverlaps, geoTouches, geoWithin.
    • Issue #1389: Added new type SpeakableSpecification, with properties cssSelector and xpath, alongside a property on WebPage called speakable to refer to a SpeakableSpecification. The purpose (documented primarily on the speakable property) is to indicate sections of a document that are highlighted as particularly suitable for being spoken/read. These sections are indicated via URL/ID values, XPath and/or CSS selectors.
    • Issue #1397: Added CompleteDateFeed to indicate completeness of feeds.
    • Issue #1423: Added a draft WebAPI.
    • Issue #1425: Added a draft measurementTechnique property, to support richer dataset description particularly for scientific data.
    • Issue #1495: Added a consortium type, as well as a LibrarySystem type to represent a particular kind of library consortium.
    • Issue #743: Added a distillery type.
    • Issue #894: Added CategoryCode, CategoryCodeSet, codeValue and inCodeSet to explore better description of enumerated value lists.
    • Issue #1004: Added broadcastFrequency, broadcastFrequencyValue and BroadcastFrequencySpecification for describing (primarily radio but also TV) frequencies. Also supporting types FMRadioChannel, AMRadioChannel, and property hasBroadcastChannel (inverse: providesBroadcastService).

    Archived into attic section

    • The proposed variablesMeasured property has been removed from pending. A non-plural variableMeasured property replaced it.
    • broadcastSignalModulation was retired august 17, 2016 from pending section.
    • The courses proposal while in pending originally had additional terms—courseCredits, courseCreditsUnit, grantsCredential, previewUrl—which should be archived here.

    Site improvements

    • Issue #1027: Published a basic robots.Txt file to avoid a template file being indexed by search engines.
    • Issue #1272: Created an internet of things (IoT) section, mailing list and discussion document. At this stage no specific schemas are proposed, only some basic infrastructure for collaboration and schema review.
    • Issue #1398: Added HTML anchor to support links to the “incoming” section of pages documenting types (see example).
    • Issue #1483: Added a sitemap.Xml file.
    • Issue #53: Added Travis-CI monitoring for test cases and syntax checking of examples.
    • Issue #1351: Improvements to the output formats of downloads section in developers page.

    Quick fixes and examples

    • Issue #1475: Clarified that event can have an audience property.
    • Issue #1494: Broadened lender to allow organizations as possible values.
    • Issue #1294: Restored material as a super-property of artMedium that can be used with any product or CreativeWork. Also marked artworkSurface as another specialized sub-property of material. The material property can now also have product as its value, allowing branded and manufactured materials to be described in more detail.
    • Issue #1290: Add the word “event” to isAccessibleForFree definition.
    • Issue #1162: Updated text for sameAs—freebase mention migrated to reference Wikidata.

    3.1 (sdo-makemake; August 9, 2016)

    Version 3.1 (working name “sdo-makemake”). See planning pages, specifically #1212 for details.

    This release is primarily concerned with hotel/accomodation vocabulary, improvements around dataset, and bug fixes.

    Vocabulary

    • Issue #915: Added substantial vocabulary for hotels and related accomodation, alongside a dedicated documentation page. New vocabulary includes LodgingBusiness, accommodation, and several subtypes which can be used (via multiple typing) alongside the existing offer type. Other terms include starRating, amenityFeature, LocationFeatureSpecification, floorSize, permittedUsage, petsAllowed, checkinTime, checkoutTime…
    • Issue #1203: Fixes for accidental rollbacks in 3.0. See Github for comprehensive analysis and details. Restored several definitions: paymentStatus lacked PaymentStatusType from #518; breastfeedingWarning, healthCondition, prescriptionStatus and secondaryPrevention have been migrated into health-lifesci.schema.org as planned; SocialMediaPosting has supertype of BlogPosting (from #526) rather than Article; genre expects URL values (from #346); affiliation is a sub-property of memberOf (per #596); cookTime, prepTime and totalTime definitions hyperlink to ISO 8601 details; publisher property can reference Person (fix for #1198, restoring #611); videoFormat is again expected on BroadcastEvent, ScreeningEvent (and BroadcastService), rather than TelevisionStation (per #577); parentOrganization and subOrganization are noted as mutual inverses (per #g535); codeSampleType description improved per #343; netWorth definition now correctly omits Organization (per #585).
    • Issue #900: Extended isAccessibleForFree to be expected potentially on any event or CreativeWork.
    • Issue #996: Added DateTime to expected values for startDate, endDate, commentTime.
    • Issue #1083: Improvements primarily around dataset, generalized spatial property of dataset to become spatialCoverage of any CreativeWork; similarly generalized temporal as temporalCoverage. Indicated that spatialCoverage is a subproperty of contentLocation. Broadened sponsor to apply to creative works such as datasets, and added a more specific subproperty to indicate funder. Amended definition of MediaObject to match its use as a supertype of DataDownload. Amended fileFormat to allow URL as a way of indicating niche or unregistered file formats (common for scientific datasets). Added a draft variablesMeasured property to the pending extension.

    Site improvements

    • Issue #1234: Enhanced incoming properties display (instances of […] may appear as values for the following properties) to include properties from extensions.
    • Issue #479: Updated and clarified the datamodel document, including conformance section.
    • Issue #1213: Fixes for site performance problems.
    • Issue #1214: Fix for RDFa formatting errors in per-term pages.
    • Issue #1248: Fix for need easy way to download rdf definition files. Introduces a vocabulary definition download section providing access to RDF representations of the vocabulary and extensions in triple, quad, JSON-LD, and turtle format.

    Quick fixes and examples

    • Issue #1151: Broadened version to allow textual values as well as numbers.
    • Issue #1257: Fixed all labels to match corresponding term IDs exactly.
    • Issue #1260: Fixed product example, microdata was missing itemprop attribute.
    • Issue #1201: Clarified deliveryLeadTime for the case of on-site pickup.
    • Issue #1283: Removed dentist from deprecated type ProfessionalService, added LocalBusiness as a supertype.

    The following early access fixes were published between 3.0 and 3.1 releases:

    • Issue #1159: Fixed problems with W3C validator, caching, and misclassification of manufacturer property which in in core rather than health-lifesci extension.
    • Issue #1165: Improved wording for term pages that are not in core, to avoid publishers worrying that they need to change their markup.
    • Issue #1136: Fixed term labels to always be alphanumeric (no “bib:” prefixes etc.)
    • Issue #1168: Fixed unit tests to run against extension schema data as well as core.
    • Issue #1169: Fixes for property/type associations in health-lifesci and pending drafts, identified by improved unit tests.
    • Issue #1174: Fix for missing definition for “ingredients.”

    3.0 (sdo-deimos; May 4, 2016)

    Version 3.0 (working name “sdo-deimos”). See planning pages, specifically #911 for details.

    This release introduces three new extensions.

    The meta.schema.org extension contains terms primarily designed to support the implementation of the Schema.org vocabulary itself.

    The pending.schema.org extension is a staging area for work-in-progress terms which have yet to be accepted into the core vocabulary. Pending terms are subject to change and should be used with caution.

    The health-lifesci.schema.org extension is a new home for our existing medical and healthcare related terms, as well as potential future collaborative work.

    The initial releases of bib.schema.org and auto.schema.org have also been finalized.

    In addition to numerous vocabulary-related improvements, we have added a new “how we work” document providing an overview of the project's approach to schema development, collaboration, versioning and change review. In particular it documents the use of the new pending extension, which gives earlier access to proposed new schema designs, as well the parallel webschemas.org site which shows the latest work-in-progress version of the entire schema.org site.

    Vocabulary

    Core changes

    • Issue #845: Added vocabulary for suitability to restricted diets in recipe (suitableForDiet, RestrictedDiet).
    • Issue #1005: Generalized some properties from product to apply also to service-audience, brand, logo, isRelatedTo, isSimilarTo.
    • Issue #880: Added disambiguatingDescription to explicitly indicate short Wikipedia-like disambiguating descriptions.
    • Issue #1048: Added DigitalDocument and subtypes SpreadsheetDigitalDocument, PresentationDigitalDocument, TextDigitalDocument, NoteDigitalDocument. Also a hasDigitalDocumentPermission property of these whose value is a DigitalDocumentPermission type. A permissionType property of each DigitalDocumentPermission has values that are from a DigitalDocumentPermissionType enumeration: CommentPermission, ReadPermission, WritePermission. The grantee property indicates in each case the person, organization, contact point, or audience that has been granted that permission.
    • Issue #888: Added PreSale to the ItemAvailability enumerated values.
    • Issue #962: Broadened definition of sponsor property.
    • Issue #961: Improved wording of validThrough, and added JobPosting to allow markup for job posting closing dates.
    • Issue #1020: Added new subtypes of CreativeWork: Message and conversation. Marked EmailMessage as a subtype of message.
    • Issue #1079: Created ComputerLanguage, amended programmingLanguage to expect it (and text); amended language to focus on natural languages. Noted that BCP 47 should be used for natural language tagging, via alternateName property. Amended availableLanguage to indicate that text values are expected, that BCP 47 is encouraged, and to cross-reference with inLanguage.
    • Issue #1092: Amended contributor, actor, translator, director, composer to apply to event.
    • Issue #969: Integrated improvements from FIBO to core terminology around financial products, including the addition of MonetaryAmount (superseding DatedMoneySpecification) and CompoundPriceSpecification, alongside improved wording around CreditCard, PriceSpecification.
    • Issue #923: Created schema.org entries for Monday–Friday and PublicHolidays, based on the GoodRelations originals. Also clarified OpeningHoursSpecification and added a specialOpeningHoursSpecification property to allow for general opening hours to be overridden.
    • Issue #541: Added enumerations for male and female, but allow other values as well.
    • See also issue #492 below; many terms have been moved from the core into health-lifesci.
    auto.schema.org
    • Issue #922: Finalized first release of auto.schema.org. Removed “pre-final preview release” description.
    bib.schema.org
    • Issue #911: Finalized first release of bib.schema.org. Removed “pre-final preview release” description.
    health-lifesci.schema.org
    • Issue #492: Moved a large number of medical/health terms from core into a health-lifesci extension.
    pending.schema.org

    The pending extension is a staging area for work-in-progress terms which have yet to be accepted into the core vocabulary.

    • Issue #1050: Added contentReferenceTime.
    • Issue #1062: Added HealthInsurancePlan, a US health insurance schema.
    • Issue #1110: Added accessMode, accessModeSufficient and accessibilitySummary.
    • Issue #447: Added EventSeries.
    • Issue #271: Added quotation and spokenByCharacter.
    • Issue #1061: Added ClaimReview and claimReviewed for fact-checking reviews.
    • Issue #1059: Created pending.schema.org to enable publication of work-in-progress drafts for wider review.
    • Issue #1045: Added LinkRole as a subtype of role, and a linkRelationship property.
    • Issues #243: Added WorkersUnion.
    • Issue #1004: Added broadcastFrequency to BroadcastService to specify the over the air frequency.
    • Issue #195: Added snapshot of the evolving course-related terms being developed by the course extension community group.

    Site improvements

    • Issue #1059: Created hosted extensions for “meta” vocabulary and for “pending” vocabulary under review. Moved class, property, domainIncludes, rangeIncludes, inverseOf, supersededBy into meta.
    • Issue #256: Implementation of HTTP HEAD for all pages.
    • Issue #1095: ETag and last-modified headers only for successful page requests.
    • Issue #1098: Added categorization of terms listed on extension home page.
    • Issue #990: JSON-LD context file should enumerate all terms. This will make it easier to combine the use of schema.org with other vocabularies and extensions.
    • Issue #1075: Added markdown format processing of comment fields. This makes the schema editing process easier, especially for cross-referencing related terms.
    • Issue #178: Integrated rdflib python library. This improves our ability to parse RDFa and JSON-LD schema definitions, and makes unit tests easier to run.
    • Issue #1109: We now display when a type is superseded.

    Quick fixes and examples

    • [core] issue #1097: partOfOrder and partOfSystem are no longer marked as sub-properties of isPartOf since they are not relations between creative works.
    • [core] issue #1102: Examples fixed to avoid “content” attribute on span element in microdata.
    • [core] issue #1066: Cleanup around dataset/DataCatalog association: Renamed containedDataCatalog as includedInDataCatalog, marked it as inverse of dataset and improved wording of both definitions.
    • [core] issue #1073: Removed redundant types listed for branchCode—place already covered LocalBusiness and store.
    • [core] issue #1081: Added AudiobookFormat to the BookFormatType, for consistency. Note that the bib extension also has a dedicated type for describing audio books.
    • [core] issue #950: isBasedOnUrl now called isBasedOn; isBasedOnUrl still exists but is marked as supersededBy isBasedOn. Extended range to include (for now just) CreativeWork, product.
    • [core] issue #929: Marked subEvent and superEvent as inverses of each other.
    • [core] issue #929: The colleague property now also expects URL values (and for JSON-LD therefore expects things not strings by default).
    • [core] various typo fixes (including #1014).
    • [core] issue #966: Fixed bug in PropertyValue JSON-LD example.
    • [core] issue #912: Added education/learning examples (from Phil barker/LRMI): learningResourceType, educationalAlignment, AlignmentObject, educationalFramework, alignmentType, targetName, targetUrl, audience, EducationalAudience, typicalAgeRange, timeRequired, educationalRole.
    • [core] issue #919: Fixed Javascript errors caused by bad markup in 404 pages (published December 10, 2015).
    • [core] issue #899: Bugfix for a few pages that were returning 500 error codes (published December 10, 2015).
    • [core] issue #904: Simplified JSON-LD examples for periodical (published December 10, 2015).
    • [core] issue #1015: Fixed spelling of “content” in FoodEstablishment example.
    • [core] issue #997: Expanded sharedContent to coverage of SocialMediaPosting example.
    • [core] issue #992: Removed an example from QAPage since the example did not illustrate the term's use.
    • [core] issue #836: Markup fixes to examples of PropertyValue.
    • [core] issue #143: Fixed openingHours examples to use span element instead of time, and some time examples to use datetime attribute instead of content.
    • [core] issue #3905: CSS fix to actions document.
    • [core] issue #511: Microdata examples fixed to use “href” example on “link” element instead of “content” on “meta.”
    • [core] issue #533: OpeningHoursSpecification example: Various fixes including better use of “time” element.
    • [bib] issue #984: Bug fixes for bib.schema.org/Chapter example.

    2.2 (sdo-phobos; November 5, 2015) Version 2.2 (working name “sdo-phobos”). This release combines a range of small additions across several areas of schema.org with an ongoing concern for cleanup and integration of existing vocabulary. Notable additions include ExhibitionEvent, barcode, DataFeed, containsPlace, GeoCircle, geoRadius, geoMidpoint, providerMobility and actionPlatform. A number of changes improve our handling of service description, including the deprecation of the ProfessionalService construct. We have kept all the specific types that were previously considered ProfessionalService but discourage the use of the type “ProfessionalService” since it is easily confused with “service.”

    Vocabulary

    • Issue #773: Added actionPlatform as a simpler mechanism to specify the platform for EntryPoint targets.
    • Issue #823: Marked homeLocation, workLocation as subproperties of location, and marked location as expecting textual values.
    • Issue #818: Added OfferCatalog for specifying a catalog of offers by an organization or service.
    • Issue #817: Added providerMobility to describe whether a service provider is at a fixed location or has a dynamic location.
    • Issue #445: Added ExhibitionEvent and added a superproperty of workPresented and workPerformed to cover event-to-work relationships such as a work being exhibited.
    • Issue #113: Added GeoCircle as a subtype of GeoShape, providing a more usable mechanism than circle structured text; added geoRadius, geoMidpoint to describe the size and location of these circular areas.
    • Issue #323: Added locationCreated to CreativeWork.
    • Issue #811: Renamed containedIn as containedInPlace, and added an inverse containsPlace.
    • Issue #810: postalCode and addressCountry are now an expected property for GeoShape and GeoCoordinates. Adjusted addressCountry to allow textual values, matching the existing definition.
    • Issue #801: Broadened award, category, offers to be applicable on service. Extended itemOffered to expect service as a possible value. Added LegalService as a supertype for notary, attorney. Marked ProfessionalService as deprecated, with some explanation, and added a brief account of the relationship with service to some of the more service-oriented local business types for example, LegalService, HomeAndConstructionBusiness. A cluster of construction-related local businesses formerly treated as ProfessionalService continue as HomeAndConstructionBusiness subtypes. Marked attorney as deprecated in favor of LegalService per #784. Added hoursAvailable to service.
    • Issue #411 clarification and cleanup around areaServed, now marked as a super-property of the more specific eligibleRegion and availableAtOrFrom properties. Cross-referenced eligibleRegion and ineligibleRegion. Marked serviceArea as supersededBy areaServed. Updated the associated types for areaServed to cover those used by its sub-properties.
    • Issue #808: The address property now anticipates text values, to support unstructured address data.
    • Issue #802: Referenced world Geotic system from GeoCoordinates properties (elevation, longitude, latitude); made address an expected property for GeoCoordinates and GeoShape.
    • Issue #796 added offeredBy property to relate an offer to a person or organization. Inverse of makesOffer.
    • Issue #791 added branchCode to represent codes identifying stores and other business locations.
    • Issue #790 broadened fileFormat to be usable with CreativeWork, and improved the description.
    • Issue #687: Added barcode as a subtype of ImageObject.
    • Issue #688: Added DataFeed and DataFeedItem to represent a data feed of structured items and supportingData for using DataFeed to provide data to a software application.
    • Issue #767: Added countryOfOrigin property to movie, TVEpisode, TVSeason, and TVSeries.
    • Issue #768: Added broadcastOfEvent property to BroadcastEvent to allow for specifying the event being broadcasted.
    • Issue #432: Added organization to domain of alumni and to range of alumniOf properties.

    Site improvements

    • Updated FAQ to clarify CC licensing of docs/examples and use of https: URLs in structured data.
    • Updated feedback form to permit per-term feedback and to be explicit that feedback messages may be shared with the wider community.
    • Issue #769: Sorting of multi-inheritance breadcrumbs on term pages.
    • Issue #761: Updated linking structure in “more…” section of term pages.
    • Listing of extension contents on extension home pages (eg. bib.schema.org)
    • Term counts added to docs/schemas.Html—the core vocabulary currently consists of 638 types, 980 properties, and 219 enumeration values.
    • List of enabled extensions added to docs/schemas.Html.
    • Various internal software fixes, unit tests and cleanup.

    Quick fixes and examples

    • Issue #764 added example for makesOffer.
    • Issue #733 added example showing library availability for book. Also added example for library and linked it to openingHoursSpecification as a good example for that too.
    • Bulk update to examples: Various fixes to html syntax, missing quotes and <meta> tags that should be <link> tags.
    • Issue #478: Replaced <meta> tags that should be <link> tags in periodical examples.
    • Several small markup fixes to various examples.

    See also issue list for this release.

    2.1 (sdo-ganymede; August 6, 2015) Version 2.1 (working name “sdo-ganymede”) includes many small fixes, some new vocabulary, as well as the first release of the “hosted extensions” mechanism, featuring bib(liographic), and auto(mobile) extensions for final review. A pre-release of this version of schema.org was available for review.

    Vocabulary

    • Issue #611: Added person as expected value for publisher property.
    • Issue #596: Documented that affiliation is a sub-property of memberOf.
    • Issue #207: Added a bridge type under CivicStructure.
    • Issue #346: Added URL as expected value for genre property.
    • Issue #374: Added report type under article, and a reportNumber property.
    • Issue #577: Amended videoFormat to indicate that it is expected on BroadcastEvent and ScreeningEvent, rather than TelevisionStation.
    • Issue #535: Broadened domain of parentOrganization to allow any organization, rather than only LocalBusiness. Noted parentOrganization and subOrganization as inverses.
    • Issue #531: Added the lyrics property to MusicComposition.
    • Issue #518: Renamed paymentDue property to paymentDueDate. Changed paymentStatus property to expect an enumerated value of type PaymentStatusType in addition to text. The options are: PaymentComplete, PaymentDue, PaymentPastDue, PaymentDeclined and PaymentAutomaticallyApplied.
    • Issue #137: Deprecated UserInteraction and its children in favor of a new type: InteractionCounter. InteractionCounter allows for reporting aggregated statistics by service and type. Thanks elf Pavlik for collaboration.
    • Issue #526: Added new types LiveBlogPosting, SocialMediaPosting, and DiscussionForumPosting, alongside supporting properties.
    • Issue #334: Updated the controlled values for accessibilityHazard shown in examples to follow latest community recommendations: “noFlashing” is now “noFlashingHazard,” “noMotionSimulation” is “noMotionSimulationHazard,” and “noSound” is “noSoundHazard.”

    Site improvements

    • Issue #429: Hosted extensions for example, bib.schema.org are now enabled and linked from “core” pages with appropriate navigation and styling.
    • Issue #655: Extension definitions now include an isPartOf property to link them to the extension.
    • Issue #527: Added materials from kickoff workshop in 2011.
    • Issue #78: Multiple supertypes are now displayed properly. It is possible for several schema.org types to apply to a single thing, and for a type to have multiple parent types. This is relatively rare, but important to display. For example, anything that is a LocalBusiness is considered both an organization and a place; so all subtypes of LocalBusiness have both as supertypes. We can now show this more clearly, for example, ElectronicsStore.
    • Added CSS for explicit styling () in releases page during steering group review of candidate releases.
    • Issue #557, #660, #695, #469: Several DataType-related navigation fixes, following recent modeling changes.
    • Issue #548: Added a favicon.Ico image. The text is tiny and not intended for humans to read.
    • Behind the scenes: Cleanup and improvements to python code, unit tests, CSS, filetree layout.

    Quick fixes and examples

    • Issue #343: Improved description of sampleType.
    • Issue #498: We now write “a URL” consistently, rather than sometimes also using “an URL.”
    • Issue #585: Fix description of netWorth, which erroneously included organization.
    • #570: Added RDFa and microdata examples for WatchAction.
    • Added new accessibility examples.
    • Numerous small documentation fixes for typos, markup validity and textual clarity.
    • Fixes to several examples including byArtist, place, question/answer, …
    • Issue #552, #644: Fixes for book examples.
    • Issue #382: Removed examples that do not reference WebPage from WebPage.
    • Issue #643: Moved bulk of mainEntityOfPage documentation into dedicated background notes.

    See also issue list for this release.

    2.0 (sdo-gozer; May 12, 2015) Version 2.0 consolidates and integrates schema.org's existing vocabulary. This version features many updated definitions, type/property associations and term names that improve consistency and usability. It also introduces a new extension system. Version 2.0 adds various new terms including mainEntityOfPage and major contributions from the automotive ontology community.

    Vocabulary

    • Issue #262: Added new vocabulary for car: The new vocabulary includes a number of properties such as: numberOfAirbags, numberOfAxles, numberOfForwardGears, fuelConsumption, fuelEfficiency vehicleEngine, EngineSpecification… Thanks to Martin Hepp, Mirek Sopek, Karol Szczepański and others in the automotive ontology community group. We anticipate substantial additions that build on this foundation will be contributed as an extension.
    • Issue #301: Added a generic property/values mechanism, building upon PropertyValueSpecification, StructuredValue and value; added PropertyValue and additionalProperty. See examples for usage. This is a lightweight mechanism to allow markup of properties that are not easily mapped to existing schema.org properties. Whenever possible, it is generally better to use a dedicated schema.org property if one exists.
    • Issue #301: Added mainEntityOfPage (and an inverse, mainEntity) to indicate which of the entities mentioned in a page is its primary topic.
    • Issue #225: Added a schemaVersion property, for use by applications that require more precise referencing of schema versions.
    • Issue #266: Added OrderItem to allow ParcelDelivery to be tracked per item, not just on an entire order.
    • Issue #170: Clarifying CommentAction, comment by deprecating UserComments. Several mentions of UserComments removed from other terms. “inLanguage” supersedes “language.” Added a resultComment property. UserInteraction and sub-types will be deprecated.
    • Issue #300: Answer is now a sub-type of comment. The “question” property expects “question” type for its values (rather than text as in AskAction previously). Added resultComment on ReplyAction. Cleaned up examples.
    • Issue #130: Amended true and false to be instances (rather than sub-types) of Boolean.
    • Issue #218: Amended DataType to model it as a sub-class of class, with each datatype for example, Boolean, text, date, DateTime, time, etc. being classes that have a typeOf DataType. DataTypes can have sub-types, for example, integer and float are subtypes of number.
    • Issue #328: Added ScreeningEvent to support movie/video screenings.
    • Issue #314: Added MovieClip and VideoGameClip as parallels to TVClip and RadioClip.
    • Issue #122: Added gtin12 property for product.
    • Issue #417: Renamed many types and properties, which had overly generic names. This is no cause for alarm, the superseded names can still be used, although we encourage consumers and producers of schema.org data to move towards the newer terminology. CreativeWorkSeason was season, CreativeWorkSeries was series, SoftwareSourceCode was code, TaxiService was taxi, actionApplication was application, actionOption was option, artMedium was material, artworkSurface was surface, availableOnDevice was device, branchOf was parentOrganization, broadcastTimezone was timezone, codeSampleType was sampleType, containsSeason was season, datasetTimeInterval was temporal, executableLibraryName was assembly, exerciseCourse was course, exerciseRelatedDiet was diet, incentiveCompensation was incentives, includedDataCatalog was catalog, isAccessibleForFree was free, jobBenefits was benefits, runtimePlatform was runtime, serviceArea was area, serviceOutput was produces, softwareRequirements was requirements, targetCollection was collection.

    Site improvements

    • Issue #429: Implementation of new extension system.
    • Issue #6: Improved 404 handler for missing pages, with support for old-style “/person/minister” extension URLs and action-related “/query-input” pseudo-properties.
    • Issue #318: Added an experimental JSON(-LD) view of the hierarchy. This combines D3 with RDF schema. Thanks to Gregg Kellogg and Sandro Hawke for collaboration.
    • Various behind-the-scenes improvement to the site software.

    Quick fixes and examples

    • Issue #119: Added a link to GS1's page giving background information on /globalLocationNumber.
    • Issue #471: Added a JSON example for exampleOfWork and workExample.
    • Issue #131: Amended category to be applicable to product
    • Issue #8: Fixed description of box to follow the original rNews definition (instead of accidental copy of polygon). Added explanatory text to GeoShape regarding commas/whitespace.
    • Issue #224: Changed expected value of openingHours to be text rather than duration, since latter is for ISO formats not our custom micro-syntax.
    • Issue #112: Clarified that upvoteCount and downvoteCount apply to questions, answers and comments (rather than just to questions).
    • Issue #103: Clarified description of StructuredValue.
    • Issue #333: Clarified description of BreadcrumbList.
    • Issue #456: Generalized numberOfEmployees, to apply to organization as well as to BusinessAudience.
    • Issue #439: Clarified numberOfItems with multi-page pagination.
    • Fixed the mapping from DataCatalog to dcat:Catalog
    • Issue #341: Amended programmingLanguage to expect a language rather than a thing.
    • Issue #365: Extended award for use on organization, product.
    • Issue #145: Marked serviceAudience as supersededBy the more general audience property.
    • Issue #165: Renamed ingredients to recipeIngredient, improved the description, and added ItemList as a possible value for recipeInstructions.
    • Issue #171: Removed class as a value of query property.
    • Issue #183: Comment property now applicable to an RsvpAction, and can have comment as values.
    • Issue #193: aggregateRating and review properties are now expected on brand, event, service.
    • Issue #247: We now write “antonym” instead of the incorrect “antagonym” in several action definitions.
    • Issue #316: Marked warrantyPromise as supersededBy warranty.
    • Issue #213: Amended eligibleRegion and ineligibleRegion to have the same associated types.
    • Issue #315: Updated BroadcastService, marking it as a kind of service rather than just thing.
    • Issue #426: Broadened supersededBy to be usable with classes and enumerations.
    • Issue #366: Reworded description of publication to use “item” instead of a list of types (which had omitted book), and marked it as applicable to CreativeWork.
    • Issue #288: Added properties related to boarding passes to FlightReservation.
    • Issue #329: Expanded BroadcastService and added new types CableOrSatelliteService, BroadcastChannel, TelevisionChannel, and RadioChannel to support TV/radio schedules. Thanks, Yves Raimond for careful review.
    • Issue #335: Added gameLocation to VideoGameSeries.
    • Issue #336: Added actor and director to series.
    • Issue #364: Replaced non-existent award type with book in WinAction type. Thanks, unor.
    • Issue #386: Fixed extra spaces in labels.

    See also issue list for this release.

    1.93 (sdo-stantz; February 4, 2015) Lots of bugfixes and small improvements, plus new terms to describe visual artworks and invoices.

    Vocabulary

    • Added VisualArtwork type and supporting properties—artEdition, artform, material, surface. Thanks, Paul Watson.
    • Added an invoice type to represent invoices/bills, with additional properties totalPaymentDue, minimumPaymentDue, accountId, scheduledPaymentDate, billingPeriod, paymentStatus, referencesOrder.
    • Fix to allow offers on AggregateOffer. Thanks, Simon Spero, Zinebb.
    • Expanded the list of types that musicBy applies to, matching those of actor and director. Thanks, Simon Spero.
    • Adding priceSpecification to TradeAction type, fixing #305.
    • Adding deliveryMethod to OrderAction type, fixing #306.
    • Adjusted description of width, height and depth properties to be “the ___ of the item,” rather than list specific types.

    Site improvements

    • Added support for linking to #examples within per-term pages. Thanks, Wes turner.
    • Add rangeIncludes to RDFa markup missing from property pages. Thanks, John Walker.

    Quick fixes and examples

    • Fixes to several examples, thanks to Gregg Kellogg.
    • Fixes to book example, thanks to unor.
    • Added name property in CreativeWork examples; improved MedicalScholarlyArticle example, marking up the abstract using description—thanks to Dan Scott.
    • Corrected a case typo in numberOfEmployees. Thanks, Stéphane Corlosquet.
    • Fixed mistaken use of “MovieGameSeries”; reported by both Gregg and Simon.
    • Fixed valuePattern definition; it takes text values, not number. Thanks, Holger.
    • Fixed definitions for fromLocation, toLocation; they do not support numeric values. Thanks, Simon Spero.
    • Fixed producer property; it takes an organization value, rather than applying to organizations. Thanks, unor.
    • Fixed typo: clinicalPharmacology is supersededBy clinicalPharmacology. Thanks, Marc.
    • Fixed an example that used “address” instead of PostalAddress. Thanks, unor.
    • Added 100+ periods/full stops (“.”) to end each definition.

    See also issue list for this release.

    1.92 (sdo-venkman; December 11, 2014) ItemList and BreadcrumbList changes; sports-related improvements; added a music vocabulary; video games. Also weight/height on person; control and RSVP actions; role names; dated monetary values.

    Vocabulary

    • Updated ItemList to allow for richer kinds of lists (including ordering), rather than just strings; added BreadcrumbList
    • The weight and height properties can now be with the person type. Added netWorth to person. Added EmployeeRole.
    • Added SportsOrganization as a super-type of SportsTeam, representing sports organizations, including sports teams, governing bodies, and sports associations. Added a sport property to indicate the associated sport(s) for a SportsOrganization; also added coach and athlete properties. Added numberedPosition property to OrganizationRole. Added homeTeam and awayTeam properties to SportsEvent, alongside a super-property of those, competitor.
    • Added a VideoGame type with two super-types: SoftwareApplication and a new game type. We anticipate a need for future work around game, see issue #169. VideoGame has cheatCode, gameServer, gamePlatform, playMode and gameTip properties, plus a few supporting types. A property softwareHelp has also been added to SoftwareApplication to indicate supporting documentation.
    • While integrating VideoGame an issue was raised around the need to describe series of games. As a result we have made some changes to the series type, which previously emphasized only TV and radio series: Added VideoGameSeries, BookSeries, and MovieSeries. The recently added periodical type is also now modeled as a series subtype. Some redundant type/property associations were also tidied up. The existing trailer property can now be used with VideoGame and VideoGameSeries.
    • Added ControlAction (previously proposed as OperateAction). Add expectsAcceptanceOf on ConsumeAction (previously proposed as “contingentOnOffer”). Added ineligibleRegion (on offer, previously proposed as “notAvailableAtOrFrom”).
    • Add rsvpResponse, additionalNumberOfGuests properties to the existing RsvpAction type (see also github issue and details). This change also adds an enumeration RsvpResponseType: RsvpResponseYes, RsvpResponseNo, RsvpResponseMaybe. Finally, we add a new possible ActionStatusType: FailedActionStatus, alongside an error property linking an action to more information on the cause of the failure.
    • Add roleName to role, superceding namedPosition (see also issue and details).
    • Add to MusicAlbum properties: albumProductionType, albumReleaseType, albumRelease (and inverse, releaseOf). Supporting enumerated types: MusicAlbumProductionType (CompilationAlbum, DemoAlbum, DJMixAlbum, LiveAlbum, MixtapeAlbum, RemixAlbum, SoundtrackAlbum, SpokenWordAlbum, StudioAlbum). MusicAlbumReleaseType (AlbumRelease, BroadcastRelease, EPRelease, SingleRelease). Add new types: MusicRelease, MusicComposition, the latter with properties musicArrangement, composer, firstPerformance, includedComposition, iswcCode, lyricist, musicCompositionForm, musicalKey, recordedAs (inverse: recordingOf). Expand track to allow ItemList values, for lists of MusicRecording, and genre to apply to MusicGroup.
    • Broadened the description of logo property, to match the types it applies to. See also github issue and details.
    • Added a DatedMoneySpecification type for non-price monetary values, by analogy with PriceSpecification and its subtypes. Added amount and currency properties and tweaked startDate and endDate property definitions to fit.
    • See also blog post…

    Site improvements

    • Added links to supporting blog posts for role, action and periodical.
    • Added “alt” attributes to all image examples.
    • Fixed the full hierarchy page.

    Quick fixes and examples

    • Several minor bugfixes.

    See also issue list for this release.

    1.91 (September 12, 2014) Offer/price documentation fixes, cleanup and community contributions.
    • Updated text of the price property to include practical usage guidance, alongside links to information from GS1 to the gtin-related offer properties.
    • Updated all our examples to follow that guidance; primarily by using priceCurrency and the content = attribute.
    • Noted our thanks to the OpenDomain project for our domain name.
    • Updated the text of the “image” property to match its expected types. Thanks, Dan Scott.
    • Changed spelling of “supercededBy” to the more conventional supersededBy. Thanks, Sachini Aparna Herath.
    • Noted that “logo” and “photo” are sub-properties of “image.” Thanks, Sachini Aparna Herath, again.
    • Fixed two syntax errors in examples (store opening hours RDFa; book, PublicationVolume microdata). Thanks, Gregg Kellogg.
    • Added Tolkien-based examples for exampleOfWork/workExample. Thanks, Dan Scott, again.
    • Fixed a bug with our UTF-8 support. Thanks, Richard Wallis.
    1.9 (August 18, 2014) Bibliographic improvements (periodical and more) alongside seller/broker/provider clarifications.

    Several bibliographic improvements, thanks to the work of the W3C BibExtend community group. Improved handling of periodicals, articles and multi-volume works, introducing types and supporting properties for periodical, PublicationVolume and PublicationIssue. Adds new article properties: pageStart, pageEnd and pagination. CreativeWork gains a convenience inverse of isPartOf called hasPart, as well as a new pair of inverses, workExample and exampleOfWork that link specific and general perspectives on creative works. Many thanks to all the BibExtend group members, to Richard Wallis for chairing, and to Dan Scott for implementing.

    Provider/seller vocabulary: This update also clarifies how to use a set of related terms, integrating sections of schema.org that had developed separately. Full details are documented at W3C and in Github, alongside an overview document. The provider property describes a service provider, service operator, or service performer, while seller describe the entities which sell or offer a service on behalf of an actual service provider. We introduce a more general broker property (replacing the over-specific bookingAgent), as well as deprecating vendor and merchant in favor of seller. Regarding flights and parcel delivery, carrier is superseded by provider. Please also note that flightNumber should now be written in full (i.e., “UA110” rather than just “110”). See the supporting documents for details.

    Finally, episodeNumber, seasonNumber and clipNumber (for TV/radio) have been harmonized with the periodicals design (issueNumber and volumeNumber). In each case, both integer and text values are anticipated. A common superproperty, position can also be used. This reflects implementation experience and establishes a model that can be applied elsewhere, for example, ongoing work on describing music.

    This update also improves the display and navigation of supersededBy relations between schema.org properties (e.g., see seller).

    The terms of service document was also updated to note that Yandex is now a schema.org sponsor and to adopt the patent terms used by W3C.

    (announcement)

    1.8 (July 28, 2014) Added WebSite type; broadened isPartOf property. Added a new CreativeWork type, WebSite: “A WebSite is a set of related web pages and other items typically served from a single web domain and accessible via URLs.” also adopts the suggestion made by the BibExtend group and other collaborators, to broaden the isPartOf property, which now relates any CreativeWork to any other CreativeWork. This release also makes the potential actions documentation available in HTML. Several markup fixes from Stéphane Corlosquet, alongside improvements to the consistency of encoding/associatedMedia property definitions thanks to Dan Scott. (announcement)
    1.7 (July 8, 2014) Allows image with ImageObject; adds event organizers, map improvements. This release amends the image property, noting that ImageObject is a reasonable value. It also adds an organizer property to event. Changes around map: We add and prefer a “hasMap” property which supersedes the older “map” property, and we add a mapType property which comes with some enumerated values: ParkingMap, SeatingMap, TransitMap, VenueMap. A map might be (but need not be) also an ImageObject. (announcement)
    1.6 (June 16, 2014) Roles, various fixes, site navigation improvements. This release introduced the notion of a role type to schema.org. See the blog post for details. The role mechanism applies across all of schema.org, and allows simple statements to be elaborated or qualified, for example with temporal information. Other changes in this release include the addition of a license property, and some documentation of properties used internally by schema.org's documentation system (property, class, supersededBy, inverseOf). Fixed an embarrassing typo—“dead”—in the diet property—thanks, Dan Scott. Various other small changes—listed in the preview announcement (announcement and candidate release).
    1.5 (May 27, 2014) Tweaks and fixes. Tweaks to definition of “keywords” to indicate that comma-separated entries are expected. Several typo fixes from Stéphane Corlosquet. Adds a new icaoCode text-valued property for airport. Updated description for flightNumber to note “not including the airline IATA code.” Added missing actions-related definitions (urlTemplate, valueName). (announcement)
    1.4 (May 16, 2014) Examples, examples, examples. Added two examples to accompany workPerformed. Added several local business, company contact and music event examples. Several glitch and typo fixes including MedicalEnumeration, thanks to Dan Scott. Announcement blog post
    1.3 (May 1, 2014) Added workPerformed. Added workPerformed, a relationship between an event and a CreativeWork that is performed at the event.
    1.2 (April 16, 2014) Potential actions. Introduces vocabulary that enables websites to describe the actions they enable and how these actions can be invoked. (announcement and blog post)
    1.1 (May 4, 2014) Reservations, Q/A and more.

    Added a reservation schema. Question/answer sites. Order changes. Added EmailMessage.

    (1) this release of schema.org adds a finalized reservations schema, which refines earlier proposals discussed in the web schemas community over the last year. See wiki for details.

    The reservations vocabulary includes: Reservation, ReservationPackage, ReservationStatusType, ticket, flight, EventReservation, FlightReservation, TrainReservation, BusReservation, TaxiReservation, LodgingReservation, RentalCarReservation, ProgramMembership, seat, vehicle, airline, FoodEstablishmentReservation, TrainTrip, car, BusTrip, taxi.

    (2) Q/A sites

    A simple vocabulary for question/answer and FAQ sites. It adds types for question, answer, and QAPage.

    (3) adds a type EmailMessage, analogous to the existing WebPage type.

    (candidate release drafted as “1.0 g”; not formally announced but included in the change to github-based publishing)

    1.0f (February 5, 2014) Not-for-profit offer and other fixes. This version adapts the wording around the offer type to suit non-profit scenarios, such as libraries. Thanks to Dan Scott and the BibExtend group for these proposals. This change broadens the definition of offer and makes associated changes to ItemAvailability and OfferItemCondition, typically rewording “the item for sale” to simply say “the item.” The itemOffered property is now defined more appropriately as “the item being offered.” rather than “…Sold.” The “seller” property is now defined as “the organization or person making the offer,” a compromise to support its use for more kinds of offer. (candidate release and announcement)
    1.0e (December 4, 2013) Accessibility properties and order type. This release includes a schema for describing orders, as well as the accessibility properties for CreativeWork (see discussion). The W3C WebSchemas wiki has the order proposal and draft schema. For accessibility, see the blog post for more background on the collaborations involved. See also wiki materials and the draft schema. (announcement)
    1.0d (November 21, 2013) TV/radio; civic services; ContactPoint event and organization improvements.

    TV/radio improvements in collaboration with BBC and EBU—see guest blog post by Yves Raimond. For example see TVEpisode.

    For background see the TV/radio wiki entry.

    Civic services (which includes ContactPoint) was blogged when first discussed. See also wiki entry. Example type: GovernmentService.

    Event improvements: Wiki entry. Organization improvements: Wiki entry. In particular, this update added department, and subOrganization properties.

    (blog announcement).

    1.0c (August 7, 2013) Actions (in the past). Introduced a (past-tense) actions vocabulary (see wiki) this updated added an action type, supporting properties, and a substantial set of subtypes. It was followed in 2014 by a mechanism for describing potential actions. (announcement)
    1.0b (July 24, 2013) Added sameAs; broadened citation.

    Adds sameAs property to thing that makes it easier to indicate identifying URLs for entities being described (wiki). The citation property has been moved up to CreativeWork, from MedicalScholarlyArticle (wiki). Thanks to the BibExtend group for suggesting this small but useful improvement.

    (announced)

    1.0a (April 5, 2013) Adds LRMI, datasets, audience, technical publishing vocabulary and more.

    For LRMI (learning resource metadata initiative), a collaboration co-led by Association of Educational Publishers (AEP) and creative commons: Added several properties, mainly to CreativeWork. Added AlignmentObject type, for aligning content to educational coding schemes. (see wiki).

    Added dataset type (collaboration building upon work at W3C and elsewhere). See also blog, wiki.

    Added audience and related types, useful for LRMI and other applications (wiki).

    Added technical publishing vocabulary: TechArticle, APIReference, code. See also early blog post and wiki entries for TechArticleSchema, APIReferenceSchema and CodeSchema.

    (announcement)

    0.99 (November 8, 2012) Added good relations e-commerce vocabulary. Added substantial e-commerce vocabulary based on good relations; re-engineered site to accept RDFa/RDFS schema definitions from community.(wiki) (announcement blog post and email from Martin Hepp)
    0.98 (October 18, 2012) Fix for requiresSubscription property. A minor bugfix release, which improved the documentation of requiresSubscription property. We now encourage its Boolean value to be represented as true/false rather than yes/no.
    0.97 (July 26, 2012) Added DateTime, time datatypes. Added two datatypes in preparation for good relations integration. (announcement)
    0.96 (July 18, 2012) Added additionalType property. Added additionalType property (wiki). “An additional type for the item, typically used for adding more specific types from external vocabularies in microdata syntax. This is a relationship between something and a class that the thing is in. In RDFa syntax, it is better to use the native RDFa syntax—the “typeof” attribute—for multiple types. Schema.org tools may have only weaker understanding of extra types, in particular those defined externally.” (announcement)
    0.95 (June 26, 2012) Added medical/health vocabulary.

    This release added substantial vocabulary in the medical/health domain (around 100 classes, 200 properties). The additions are mainly organized beneath the MedicalEntity type. An overview document is also provided. From the blog post:

    “This collaborative project drew upon search expertise from the schema.org partners but also gained immeasurably through feedback from expert reviewers including the US NCBI; physicians at Harvard, Duke and other institutions, as well as from several health web sites. Contributions from the W3C healthcare and Lifesciences group and web schemas community also helped bridge the complex worlds of web standards, search and medicine/healthcare.”

    (announcement)

    0.91 (April 21, 2012) Removed plural “s” suffix from repeatable properties; added comment type, SoftwareApplication vocabulary.

    Added new properties per Singularity proposal: 23 changes: actor, album, attendee, award, blogPost, colleague, contactPoint, employee, encoding, episode, event, founder, map, member, parent, performer, photo, review, season, sibling, significantLink, subEvent, track. Each of these originally had a plural “s” to indicate repeatability. After discussion in the W3C WebSchemas group, the consensus was that this was confusing and worthwhile changing.

    This revision added a new type, “Comment,” whose super-type is CreativeWork (wiki)

    Also adds SoftwareApplication and supporting vocabulary (wiki, original proposal), including MobileApplication, WebApplication subtypes.

    Converged URL/Url spelling. After WebSchemas discussion, consensus was: For each of embedURL, contentURL, downloadURL, ensure there is an equivalent property named with mixed-case: “embedUrl,” “contentUrl,” “downloadUrl.”

    TABLE A2. Alphabetical list of vocabulary terns in Schema.org release history as of May 11, 2022
    A
    acceptedAnswer, accessibilityHazard, accessibilitySummary, accessMode, accessModeSufficient, accommodationCategory, accommodationFloorPlan, accomodation, accountId, accountMinimumInflow, accountOverdraftLimit, acquireLicensePage, actionableFeedbackPolicy, ActionAccessSpecification, actionApplication, actionOption, actionPlatform, ActionStatusType, ActivationFee, additionalNumberOfGuests, additionalProperty, additionalType, addressCountry, addressLocality, addressRegion, AdministrativeArea, AdvertiserContentArticle, AggregateOffer, AggregateRating, albumProductionType, albumRelease, albumReleaseType, AlignmentObject, alignmentType, AllergiesHealthAspect, alternateName, alumniOf, amenityFeature, AmpStory, AMRadioChannel, AnalysisNewsArticle, announcementLocation, ApartmentComplex, APIReference, APIReferenceSchema, applicantLocationRequirements, applicationContact, applicationDeadline, applicationStartDate, ArchiveComponent, archivedAt, archiveHeld, ArchiveOrganization, areaServed, arrivalBoatTerminal, artEdition, artMedium, artworkSurface, AskAction, AskPublicNewsArticle, associatedClaimReview, associatedDisease, associatedMedia, associatedMediaReview, associatedReview, associated, AudiobookFormat, AudioObject, AudioObjectSnapshot, AuthenticContent, availableAtOrFrom, availableLanguage, availableOnDevice, awayTeam
    B
    BackgroundNewsArticle, BackOrder, bankAccountType, bccRecipient, BedType, beneficiaryBank, BibExtend, billingDuration, billingPeriod, billingStart, BioChemEntity, bioChemInteraction, bioChemSimilarity, biologicalRole, Bioschemas, blogPost, BlogPosting, BoatReservation, BoatTerminal, BoatTrip, BodyMeasurementTypeEnumeration, BookFormatType, bookingAgent, BookSeries, boolean, branchCode, branchOf, BreadcrumbList, breastfeedingWarning, BroadcastChannel, BroadcastEvent, broadcastFrequency, BroadcastFrequencySpecification, broadcastFrequencyValue, broadcastOfEvent, BroadcastRelease, BroadcastService, broadcastSignalModulation, broadcastSubChannel, broadcastTimezone, brokenlinkcheck, BrokerageAccount, BusinessAudience, businessFunction, BusReservation, BusTrip, byArtist, byDay, byMonth, byMonthDay, byMonthWeek
    C
    CableOrSatelliteService, callSign, CampingPitch, cashBack, CategoryCode, CategoryCodeSet, CatholicChurch, ccRecipient, cheatCode, checkinTime, checkoutTime, chemicalComposition, chemicalRole, ChemicalSubstance, childTaxon, CivicStructure, claimInterpreter, ClaimReview, claimReviewed, CleaningFee, clinicalPharmacology, clipNumber, codeSampleType, CodeSchema, codeValue, collectionSize, CommentAction, CommentPermission, commentTime, compatability, competencyRequired, CompilationAlbum, CompleteDateFeed, CompoundPriceSpecification, ComputerLanguage, ConsumeAction, contactlessPayment, ContactPoint, containedDataCatalog, containedIn, containedInPlace, containsPlace, containsSeason, contentLocation, contentRating, contentReferenceTime, contentURL, contingentOnOffer, ControlAction, cookTime, copyrightNotice, Corlosquet, CorrectionComment, correctionsPolicy, countryOfAssembly, countryOfLastProcessing, countryOfOrigin, courseCode, courseCredits, courseCreditsUnit, CourseInstance, courseMode, coursePrerequisites, courseWorkload, CreativeWork, CreativeWorks, CreativeWorkSeason, CreativeWorkSeries, credentialCategory, CreditCard, creditText, CriticReview, cssSelector, currenciesAccepted, currentExchangeRate, customerRemorseReturnFees, customerRemorseReturnLabelSource, customerRemorseReturnShippingFeesAmount
    D
    DataCatalog, DataDownload, DataFeed, DataFeedItem, datamodel, datasetTimeInterval, DataType, DataType-related, DataTypes, DatedMoneySpecification datePosted, datePublished, DateTime, dcat, dcterms, dds, DefinedRegion, DefinedTerm, DefinitiveLegalValue, deliveryLeadTime, DeliveryMethod deliveryTime, DemoAlbum, departureBoatTerminal, DigitalDocument, DigitalDocumentPermission, DigitalDocumentPermissionType, directApply, disambiguatingDescription, discountCurrency, DiscussionForumPosting, DistanceFee, diversityPolicy, diversityStaffingReport, DJMixAlbum, DModel, doesNotShip, domainIncludes, domiciledMortgage, downloadURL, Downpayment, downvoteCount
    E
    earlyPrepaymentPenalty, editEIDR, educationalAlignment, EducationalAudience, educationalCredentialAwarded, educationalFramework, educationalLevel, EducationalOccupationalCredential, EducationalOccupationalCredentials, EducationalOccupationalProgram, EducationalOrganization, educationalProgramMode, educationalRole, educationalUse, EducationEvents, educationRequirements, eduQuestionType, EffectivenessHealthAspect, ElectronicsStore, eligibleRegion, EmailMessage, embeddedTextCaption, embedURL, emphasised, EmployeeRole, EmployerAggregateRating, employerOverview, EmployerReview, employmentUnit, encodesBioChemEntity, encodesCreativeWork, encodingFormat, endDate, endOffset, EndorsementRating, endTime, EnergyConsumptionDetails, EnergyEfficiencyEnumeration, energyEfficiencyScaleMax, energyEfficiencyScaleMin, EnergyStarEnergyEfficiencyEnumeration, EngineSpecification, EntryPoint, eocred-schema, episodeNumber, EPRelease, equivalentProperty, estimatedCost, estimatedSalary, ETag, ethicsPolicy, EUEnergyEfficiencyEnumeration, eventAttendanceMode, EventMovedOnline, EventReservation, eventSchedule, EventSeries, eventStatus, EvidenceLevelA, exampleOfWork, exceptDate, ExchangeRateSpecification, exchangeRateSpread, executableLibraryName, exerciseCourse, ExercisePlan, exerciseRelatedDiet, ExhibitionEvent, expectsAcceptanceOf, experienceInPlaceOfEducation, experienceRequirements
    F
    FailedActionStatus, FAQPage, favour, fileFormat, filetree, financialAidEligible, firstPerformance, flightNumber, FlightReservation, floorLevel, floorLimit, FloorPlan, floorSize, FMRadioChannel, FoodEstablishment, FoodEstablishmentReservation, fromLocation, fuelConsumption, fuelEfficiency, FundingAgency, FundingScheme
    G
    gameLocation, gamePlatform, gameServer, gameTip, generalised, GeoCircle, geoContains, GeoCoordinates, geoCoveredBy, geoCovers, geoCrosses, geoDisjoint, geoEquals, geoIntersects, geoMidpoint, geoOverlaps, geoRadius, GeoShape, GeospatialGeometry, geospatiallyXyz, Geotic, geoTouches, geoWithin, geoXyz, GettingAccessHealthAspect, globalLocationNumber, GoodRelations, governmentBenefitsInfo, GovernmentBenefitsType, GovernmentService, gracePeriod, grantsCredential
    H
    hasBioChemEntityPart, hasBioPolymerSequence, hasBroadcastChannel, hasCategoryCode, hasCourse, hasCourseInstance, hasCredential, hasDefinedTerm, hasDigitalDocumentPermission, hasEnergyConsumptionDetails, hasEnergyEfficiencyCategory, hasHealthAspect, hasMap, hasMeasurement, hasMenu, hasMenuItem, hasMenuSection, hasMerchantReturnPolicy, hasMolecularFunction, hasOccupation, hasPart, hasRepresentation, hasVariant, HealthAspectEnumeration, healthCondition, HealthInsurancePlan, HealthTopicContent, highPrice, holdingArchive, HomeAndConstructionBusiness, homeLocation, homeTeam, hoursAvailable, HowItWorksHealthAspect, HowOrWhereHealthAspect, HowTo, HowToItem, HowToSection, HowToStep, HowToSupply, HowToTool, HyperToc, HyperTocEntry
    I
    icaoCode, ImageGallery, ImageObject, ImageObjectSnapshot, incentiveCompensation, inChI, inChIKey, includedComposition, includedDataCatalog, includedInDataCatalog, includesObject, inCodeSet, ineligibleRegion, IngredientsHealthAspect, inLanguage, inLanguage’, inProductGroupWithID, InteractionCounter, interactionStatistic, interpretedAsClaim, inverseOf, InvestmentFund, InvoicePrice, isAccessibleForFree, isBasedOn, isBasedOnUrl, isEncodedByBioChemEntity, isInvolvedInBiologicalProcess, isLiveBroadcast, isLocatedInSubcellularLocation, isPartOf, isPartOfBioChemEntity, isRelatedTo, isResizable, isSimilarTo, issueNumber, isVariantOf, iswcCode, ItemAvailability, itemCondition, itemDefectReturnFees, itemDefectReturnLabelSource, itemDefectReturnShippingFeesAmount, ItemList, itemOffered, itemReviewed, iupacName, Iussue
    J
    Javascript, jobBenefits, jobLocationType, JobPosting, JobPostings, jobTitle, jsonldcontext
    K
    knowsAbout, knowsLanguage
    L
    layoutImage, LearningResource, learningResourceType, leaseLength, LegalForceStatus, LegalService, LegalValueLevel, legislationIdentifier, legislationJurisdiction, LegislationObject, legislationType, LibrarySystem, linkRelationship, LinkRole, liographic, ListPrice, LiveAlbum, LiveBlogPosting, loanMortgageMandateAmount, loanPaymentAmount, loanPaymentFrequency, loanRepaymentForm, loanType, LocalBusiness, localised, locationCreated, LocationFeatureSpecification, LockerDelivery, LodgingBusiness, LodgingReservation, lowPrice
    M
    mainEntity, mainEntityOfPage, maintenance, makesOffer, mapType, marginOfError, materialExtent, MathSolver, maximumAttendeeCapacity, maximumEnrollment, measurementTechnique, MeasurementTypeEnumeration, MediaGallery, mediaItemAppearance, MediaManipulationRatingEnumeration, MediaObject, MediaReview, MediaReviewItem, MediaSubscription, MedicalEntity, MedicalEnumeration, MedicalScholarlyArticle, MedicalSpecialty, memberOf, membershipPointsEarned, menuAddOn, MenuItem, MenuSection, MerchantReturnPolicy, MerchantReturnPolicySeasonalOverride, MinimumAdvertisedPrice, minimumPaymentDue, MissingContext, missionCoveragePrioritiesPolicy, MixtapeAlbum, MobileApplication, MolecularEntity, molecularFormula, molecularWeight, MonetaryAmount, MonetaryAmountDistribution, MonetaryGrant, MoneyTransfer, monoisotopicMolecularWeight, monthlyMinimumRepaymentAmount, monthsOfExperience, MortgageLoan, MovieClip, MovieGameSeries, MovieSeries, MusicAlbum, MusicAlbumProductionType, MusicAlbumReleaseType, musicalKey, musicArrangement, musicBy, MusicComposition, musicCompositionForm, MusicGroup, MusicRecording, MusicRelease
    N
    namedPosition, negativeNotes, netWorth, NewsArticle, NewsMediaOrganization, NiemanLab, noFlashing’, noFlashingHazard’, noMotionSimulation’, noMotionSimulationHazard’, nonprofitStatus, NonprofitType, noSound, noSoundHazard, notAvailableAtOrFrom, NoteDigitalDocument, numberedPosition, numberOfAirbags, numberOfAxles, numberOfBathroomsTotal, numberOfBedrooms, numberOfCredits, numberOfEmployees, numberOfForwardGears, numberOfFullBathrooms, numberOfItems, numberOfLoanPayments, numberOfRooms
    O
    occupationalCategory, occupationalCredentialAwarded, OccupationalExperienceRequirements, OccupationalTherapy, occupationLocation, OfferCatalog, offeredBy, OfferForLease, OfferForPurchase, OfferItemCondition, OfferShippingDetails, OpenDomain, openingHours, OpeningHoursSpecification, OperateAction, OpinionNewsArticle, OrderAction, orderedItem, OrderItem, Organisation, OrganizationRole, originalMediaContextDescription, originalMediaLink, ownershipFundingInfo
    P
    pageEnd, pageStart, ParcelDelivery, ParcelService, parentItem, parentOrganization, parentTaxon, ParkingMap, partOfOrder, partOfSystem, partOfTrip, paymentAccepted’, PaymentAutomaticallyApplied, PaymentComplete, PaymentDeclined, paymentDue, paymentDueDate, PaymentPastDue, paymentStatus, PaymentStatusType, performTime, permissionType, permittedUsage, petsAllowed, phane, playMode, PodcastEpisode, PodcastSeason, PodcastSeries, positiveNotes, PostalAddress, postalCode, PostalCodeRangeSpecification, potentialAction, potentialUse, PregnancyHealthAspect, PreOrder, PreOrderAction, prepTime, PreSale, prescriptionStatus, PresentationDigitalDocument, previewUrl, priceComponentType, PriceComponentTypeEnumeration, priceCurrency, priceCurrency’, PriceSpecification, priceType, PriceTypeEnumeration, ProductCollection, ProductGroup, productGroupID, ProductReturnPolicy, ProfessionalService, ProfessionalService’, ProgramMembership, programmingLanguage, programPrerequisites, PronounceableText, PropertyValue, PropertyValueSpecification, providerMobility, providesBroadcastService, publicAccess, PublicationIssue, PublicationVolume, PublicHolidays, PublicToilet, publishingPrinciples
    Q
    QAPage, QualitativeValue, QuantitativeValue, QuantitativeValueDistribution
    R
    RadioBroadcastService, RadioChannel, RadioClip, Raimond, rangeIncludes, ratingExplanation, ratingValue, rdf, RDFa, rdflib, rdfs, ReadPermission, RealEstateListing, ReceiveAction, recipeIngredient, recipeInstructions, recipeYield, recognizedBy, recordedAs, recordingOf, recourseLoan, referencesOrder, releaseOf, relevantOccupation, remainingAttendeeCapacity, RemixAlbum, renegotiableLoan, RentalCarReservation, RepaymentSpecification, repeatCount, repeatFrequency, ReplyAction, ReportageNewsArticle, reportNumber, requiredQuantity, requiresSubscription, ResearchProject, ReservationPackage, ReservationStatusType, restockingFee, RestrictedDiet, resultComment, ReturnFeesEnumeration, returnLabelSource, ReturnLabelSourceEnumeration, returnMethod, ReturnMethodEnumeration, returnPolicyCountry, returnPolicySeasonalOverride, returnShippingFeesAmount, reviewAspect, reviewBody, ReviewNewsArticle, rNews, roleName, RsvpAction, rsvpResponse, RsvpResponseMaybe, RsvpResponseNo, RsvpResponseType, RsvpResponseYes, runtimePlatform
    S
    Sachini, SafetyHealthAspect, salaryUponCompletion, SalePrice, sameAs, sampleType, SatiricalArticle, scheduledPaymentDate, scheduleTimezone, schemaorg, schemaVersion, SchoolDistrict, ScreeningEvent, sdDatePublished, sdLicense, sdo-callisto, sdo-callisto’, sdo-deimos, sdo-deimos’, sdo-enceladus, sdo-enceladus’, sdo-ganymede, sdo-ganymede’, sdo-gozer, sdo-makemake, sdo-makemake’, sdo-phobos, sdo-phobos’, sdo-stantz, sdo-telesto, sdo-telesto’, sdo-venkman, sdopythonapp, sdPublisher, seasonNumber, SeatingMap, secondaryPrevention, SeekToAction, serviceArea, serviceAudience, serviceOutput, serviceType, sharedContent, SheetMusic, ShippingDeliveryTime, shippingDestination, shippingDetails, shippingLabel, shippingRate, ShippingRateSettings, shippingSettingsLink, ShortStory, significantLink, SingleRelease, sizeGroup, SizeGroupEnumeration, sizeproperty, SizeSpecification, sizeSystem, SizeSystemEnumeration, SkiResort, SKOS, SocialMediaPosting, SoftwareApplication, softwareHelp, softwareRequirements, SoftwareSourceCode, SolveMathAction, Sopek, SoundtrackAlbum, spatialCoverage, speakable, SpeakableSpecification, SpeakableSpecification-related, SpecialAnnouncement, specialOpeningHoursSpecification, spokenByCharacter, SpokenWordAlbum, SportsActivityLocation, SportsEvent, SportsOrganization, SportsTeam, SpreadsheetDigitalDocument, starRating, startDate, startOffset, startTime, StatisticalPopulation, StatusEnumeration, StructuredValue, StudioAlbum, subEvent, subjectOf, subOrganization, subproperties, subproperty, subtitleLanguage, subTrip, suggestedAge, suggestedAnswer, suggestedMeasurement, suitableForDiet, supercededBy’, superceding, superEvent, superproperty, supersededBy, supportingData
    T

    targetCollection, targetName, targetUrl, TaxiReservation, TaxiService, taxonomicRange, taxonRank, TechArticle, TechArticleSchema,

    TelevisionChannel, TelevisionStation, temporalCoverage, termDuration, termsPerYear, TextDigitalDocument, tickerSymbol, timeOfDay, timeRequired, timeToComplete, timezone, titleEIDR, tocContinuation, tocEntry, toLocation, toRecipient, totalJobOpenings, totalPaymentDue, totalPrice, totalTime, tourBookingPage, TouristAttraction, TouristDestination, TouristTrip, touristType, TradeAction, trainingSalary, TrainReservation, TrainTrip, TransitMap, transitTimeLabel, TravisCI, TVClip, TVEpisode, TVSeason, TVSeries, typeof, typeof’, typicalAgeRange, typicalCreditsPerTerm,

    U
    UnitPriceSpecification, unnamedSourcesPolicy, upvoteCount, url-addressable, urlTemplate, usageInfo, usecase, UserComments, UserInteraction, UserReview
    V
    validFor, validIn, validThrough, valueName, valuePattern, valueReference, variableMeasured, variablesMeasured, variesBy, vehicleEngine, VenueMap, verificationFactCheckingPolicy, videoFormat, VideoGallery, VideoGame, VideoGameClip, VideoGameSeries, VideoObject, VideoObjectSnapshot, VirtualLocation, VisualArtwork, volumeNumber
    W
    warrantyPromise, WatchAction, WearableMeasurementTypeEnumeration, WearableSizeGroupEnumeration, WearableSizeSystemEnumeration, WebAPI, WebApplication, WebContent, webFeed, WebPage, WebPageElement, WebPageElement’, webschemas, WebSite, Wikibase, Wikidata, WinAction, WorkBasedProgram, WorkersUnion, workExample, workLocation, workPerformed, workPresented, WritePermission
    X
    Xpath, xpath’
    Y
    yearBuilt
    Z
    Zinebb