Confronting disciplinary fragmentation

ResearchSpace is designed around an empirical ontology

ResearchSpace can use any properly constituted ontology, but its design aligns with the empirical and semantic approach of the CIDOC CRM.
Confronting disciplinary fragmentation

ResearchSpace is designed around an empirical ontology

ResearchSpace can use any properly constituted ontology, but its design aligns with the empirical and semantic approach of the CIDOC CRM.
Confronting disciplinary fragmentation

ResearchSpace is designed around an empirical ontology

ResearchSpace can use any properly constituted ontology, but its design aligns with the empirical and semantic approach of the CIDOC CRM.

The humanities are centered around textual narratives – journal articles and books. These narratives use various styles, language and categorisations. Historians, social scientists, archaeologists, sociologists, psychologists, anthropologists, politicians, lawyers, philosophers, etc., use different categories to talk about the same people, events, processes and things. All these elements make alignment difficult and misinterpretation common. Despite the advantages of text as a descriptive narrative, unstructured text is impossible to synthesize in detail leaving humanities disciplines fragmented across their many specialisations and with other disciplines.

In data network frameworks the potential of interconnection are apparent, but these have also focused on artificial categorisation (vocabularies) which are impossible to align either technically and meaningfully. In other words, data system suffer from exactly the same type problem as textual narrative but without any form of mitigating contextualisation.

However, with semantic data frameworks based on humanities knowledge systems there is a possibility of crossing disciplinary boundaries. For difference to be compared and reconciled and for similarities of detail and pattern to be established – reducing fragmentation.

The ResearchSpace and CIDOC CRM provides that framework. Many other Linked Data schemas are not ontologies, although they may use the same technical conventions, and simply introduce more artificial categories without context resulting in more fragmentation, resulting no effective progression.

The humanities are centered around textual narratives – journal articles and books. These narratives use various styles, language and categorisations. Historians, social scientists, archaeologists, sociologists, psychologists, anthropologists, politicians, lawyers, philosophers, etc., use different categories to talk about the same people, events, processes and things. All these elements make alignment difficult and misinterpretation common. Despite the advantages of text as a descriptive narrative, unstructured text is impossible to synthesize in detail leaving humanities disciplines fragmented across their many specialisations and with other disciplines.In data network frameworks the potential of interconnection are apparent, but these have also focused on artificial categorisation (vocabularies) which are impossible to align either technically and meaningfully. In other words, data system suffer from exactly the same type problem as textual narrative but without any form of mitigating contextualisation.

However, with semantic data frameworks based on humanities knowledge systems there is a possibility of crossing disciplinary boundaries. For difference to be compared and reconciled and for similarities of detail and pattern to be established – reducing fragmentation.

The ResearchSpace and CIDOC CRM provides that framework. Many other Linked Data schemas are not ontologies, although they may use the same technical conventions, and simply introduce more artificial categories without context resulting in more fragmentation, resulting no effective progression.

The humanities are centered around textual narratives – journal articles and books. These narratives use various styles, language and categorisations. Historians, social scientists, archaeologists, sociologists, psychologists, anthropologists, politicians, lawyers, philosophers, etc., use different categories to talk about the same people, events, processes and things. All these elements make alignment difficult and misinterpretation common. Despite the advantages of text as a descriptive narrative, unstructured text is impossible to synthesize in detail leaving humanities disciplines fragmented across their many specialisations and with other disciplines.

In data network frameworks the potential of interconnection are apparent, but these have also focused on artificial categorisation (vocabularies) which are impossible to align either technically and meaningfully. In other words, data system suffer from exactly the same type problem as textual narrative but without any form of mitigating contextualisation.

However, with semantic data frameworks based on humanities knowledge systems there is a possibility of crossing disciplinary boundaries. For difference to be compared and reconciled and for similarities of detail and pattern to be established – reducing fragmentation.

The ResearchSpace and CIDOC CRM provides that framework. Many other Linked Data schemas are not ontologies, although they may use the same technical conventions, and simply introduce more artificial categories without context resulting in more fragmentation, resulting no effective progression.

The Problem with Vocabulary

All organisations use vocabularies to categorise their collections. These mostly revolve around object records, for example, object types, periods (particularly production), materials, techniques, subjects, etc. Depending on the organisation and the scope of the collection record these can be relatively simple or more complicated, particularly ones that use hierarchical structures. While most commentators recognise that local vocabularies reflect organisation specialisms, local references and dialects, a regular case is put forward to aligning these vocabularies with umbrella vocabularies, maintained by a consistent authority organisation.

  • Keyword vocabularies are not a good way of navigating quantities of aligned data with the type of information and language that many audiences require. Locally, vocabularies are more useful because organisational users understand the underlying conventions used in their application – but even local systems are often less than satisfactory. For wider users, without insider knowledge and familiarity, vocabularies, whether aligned or not, are not suitable mechanisms.
  • Documentation vocabularies were developed to support internal administrative functions and act as a consistent vehicle for finding aids but don’t address significance and relevance or provide further context. They are founded on physical arrangement not contextual arrangement. Since collections are increasingly being published publicly and have a wider role for community collaboration and information exchange, these categorisations don’t reflect the interfaces needed by those wider audiences.
  • Where organisations have local terminologies but align them with an umbrella vocabulary for aggregation and finding aid purposes, users are searching not using the local terms assigned by the author. While this helps consistent searching it covers up specific local terms hiding them from initial view, but which have a different nuance and significance to the aggregated term they are assigned to. This affects research and covers up the concepts that general users, would recognise.
  • Even if things are aligned by essential vocabularies, it doesn’t mean that they have much in common. Vocabularies belong to the database world in which things are fitted into fixed profiles regardless of their different qualities )(every object has the same profile regardless of history). As such they project a narrow view of the objects place in the world.
  • The cultural heritage sector has never been able to establish and maintain alignments of vocabulary even when a lowest common denominator is employed. The issue has been given a new lease of life with Linked Data but top down technical solutions don’t address bottom up issues and processes or changing needs in the public domain.
  • Alignment based on essential vocabularies can skew local representations, even if the original record is found. Gaps between umbrella vocabularies and local vocabularies and be difficult to understand or even identify.
  • Using Linked Data as a means to align vocabularies misses the point about Open Data and engaging new audiences through digital information. Linked Data was originally part of an initiative called the Semantic Web. The Semantic Web proposes the use semantics to provide context for data alignment. Transferring vocabularies from legacy systems to Linked Data only perpetuates the static, artificial and narrow professional documentation and maintains a division between semantic and non-semantic approaches, transferring unsuitable presentations into the public domain, and addressing flaws in the data itself.

The use of CIDOC CRM and patterns of information that incorporate wider contextual knowledge provides a means to address the needs of wider audiences and, at the same time use this real world context act as a “semantic glue”. By providing an ontological semantic framework, users can view the variations of vocabularies that are used in the same context and make their own minds up about which variation come within their scope of investigation.

The Problem with Vocabulary

All organisations use vocabularies to categorise their collections. These mostly revolve around object records, for example, object types, periods (particularly production), materials, techniques, subjects, etc. Depending on the organisation and the scope of the collection record these can be relatively simple or more complicated, particularly ones that use hierarchical structures. While most commentators recognise that local vocabularies reflect organisation specialisms, local references and dialects, a regular case is put forward to aligning these vocabularies with umbrella vocabularies, maintained by a consistent authority organisation.
  • Keyword vocabularies are not a good way of navigating quantities of aligned data with the type of information and language that many audiences require. Locally, vocabularies are more useful because organisational users understand the underlying conventions used in their application – but even local systems are often less than satisfactory. For wider users, without insider knowledge and familiarity, vocabularies, whether aligned or not, are not suitable mechanisms.
  • Documentation vocabularies were developed to support internal administrative functions and act as a consistent vehicle for finding aids but don’t address significance and relevance or provide further context. They are founded on physical arrangement not contextual arrangement. Since collections are increasingly being published publicly and have a wider role for community collaboration and information exchange, these categorisations don’t reflect the interfaces needed by those wider audiences.
  • Where organisations have local terminologies but align them with an umbrella vocabulary for aggregation and finding aid purposes, users are searching not using the local terms assigned by the author. While this helps consistent searching it covers up specific local terms hiding them from initial view, but which have a different nuance and significance to the aggregated term they are assigned to. This affects research and covers up the concepts that general users, would recognise.
  • Even if things are aligned by essential vocabularies, it doesn’t mean that they have much in common. Vocabularies belong to the database world in which things are fitted into fixed profiles regardless of their different qualities )(every object has the same profile regardless of history). As such they project a narrow view of the objects place in the world.
  • The cultural heritage sector has never been able to establish and maintain alignments of vocabulary even when a lowest common denominator is employed. The issue has been given a new lease of life with Linked Data but top down technical solutions don’t address bottom up issues and processes or changing needs in the public domain.
  • Alignment based on essential vocabularies can skew local representations, even if the original record is found. Gaps between umbrella vocabularies and local vocabularies and be difficult to understand or even identify.
  • Using Linked Data as a means to align vocabularies misses the point about Open Data and engaging new audiences through digital information. Linked Data was originally part of an initiative called the Semantic Web. The Semantic Web proposes the use semantics to provide context for data alignment. Transferring vocabularies from legacy systems to Linked Data only perpetuates the static, artificial and narrow professional documentation and maintains a division between semantic and non-semantic approaches, transferring unsuitable presentations into the public domain, and addressing flaws in the data itself.
The use of CIDOC CRM and patterns of information that incorporate wider contextual knowledge provides a means to address the needs of wider audiences and, at the same time use this real world context act as a “semantic glue”. By providing an ontological semantic framework, users can view the variations of vocabularies that are used in the same context and make their own minds up about which variation come within their scope of investigation.

The Problem with Vocabulary

All organisations use vocabularies to categorise their collections. These mostly revolve around object records, for example, object types, periods (particularly production), materials, techniques, subjects, etc. Depending on the organisation and the scope of the collection record these can be relatively simple or more complicated, particularly ones that use hierarchical structures. While most commentators recognise that local vocabularies reflect organisation specialisms, local references and dialects, a regular case is put forward to aligning these vocabularies with umbrella vocabularies, maintained by a consistent authority organisation.
  • Keyword vocabularies are not a good way of navigating quantities of aligned data with the type of information and language that many audiences require. Locally, vocabularies are more useful because organisational users understand the underlying conventions used in their application – but even local systems are often less than satisfactory. For wider users, without insider knowledge and familiarity, vocabularies, whether aligned or not, are not suitable mechanisms.
  • Documentation vocabularies were developed to support internal administrative functions and act as a consistent vehicle for finding aids but don’t address significance and relevance or provide further context. They are founded on physical arrangement not contextual arrangement. Since collections are increasingly being published publicly and have a wider role for community collaboration and information exchange, these categorisations don’t reflect the interfaces needed by those wider audiences.
  • Where organisations have local terminologies but align them with an umbrella vocabulary for aggregation and finding aid purposes, users are searching not using the local terms assigned by the author. While this helps consistent searching it covers up specific local terms hiding them from initial view, but which have a different nuance and significance to the aggregated term they are assigned to. This affects research and covers up the concepts that general users, would recognise.
  • Even if things are aligned by essential vocabularies, it doesn’t mean that they have much in common. Vocabularies belong to the database world in which things are fitted into fixed profiles regardless of their different qualities )(every object has the same profile regardless of history). As such they project a narrow view of the objects place in the world.
  • The cultural heritage sector has never been able to establish and maintain alignments of vocabulary even when a lowest common denominator is employed. The issue has been given a new lease of life with Linked Data but top down technical solutions don’t address bottom up issues and processes or changing needs in the public domain.
  • Alignment based on essential vocabularies can skew local representations, even if the original record is found. Gaps between umbrella vocabularies and local vocabularies and be difficult to understand or even identify.
  • Using Linked Data as a means to align vocabularies misses the point about Open Data and engaging new audiences through digital information. Linked Data was originally part of an initiative called the Semantic Web. The Semantic Web proposes the use semantics to provide context for data alignment. Transferring vocabularies from legacy systems to Linked Data only perpetuates the static, artificial and narrow professional documentation and maintains a division between semantic and non-semantic approaches, transferring unsuitable presentations into the public domain, and addressing flaws in the data itself.
The use of CIDOC CRM and patterns of information that incorporate wider contextual knowledge provides a means to address the needs of wider audiences and, at the same time use this real world context act as a “semantic glue”. By providing an ontological semantic framework, users can view the variations of vocabularies that are used in the same context and make their own minds up about which variation come within their scope of investigation.