A Representation Protocol for Traditional Crafts

: A protocol for the representation of traditional crafts and the tools to implement this are proposed. The proposed protocol is a method for the systematic collection and organization of digital assets and knowledge, their representation into a formal model, and their utilization for research, education, and preservation. A set of digital tools accompanies this protocol that enables the online curation of craft representations. The proposed approach was elaborated and evaluated with craft practitioners in three case studies. Lessons learned are shared and an outlook for future work is provided.


Introduction
Crafts, handicrafts, traditional, or "heritage crafts" [1] are an integral part of cultural heritage [2,3], and are in many cases endangered due to the "declining number of practitioners and apprentices" [4]. In contrast to other forms of intangible cultural heritage, traditional crafts bring together multiple tangible and intangible dimensions, as "the most tangible manifestation of intangible cultural heritage" [5]. A classification of craft dimensions as tangible and intangible by ICCROM [6] identifies tangible dimensions as the materials, tools, workspaces, and the produced artefacts, and intangible dimensions refer to knowledge and skill but, also, to the social, technological, and historical context that led

Craft Documentation
UNESCO's recognition of traditional crafts as ICH [4] generated awareness and compliance of craft descriptions according to scientific and ethical guidelines. Descriptions following these guidelines can be found in the Representative List of the ICH of Humanity of UNESCO [11] and national inventories, e.g., [12][13][14][15]. These craft descriptions emphasize cultural content, social components, ownership, IPR, and authenticity certification, but do not include sufficient technical descriptions to support craft understanding and re-enaction. In 1990, UNESCO published a data collection guide for traditional crafts [16]. Though technologically outdated, it identifies the essential elements to be recorded: artifacts, materials, tools, and the crafting process. Photographic documentation is deemed necessary to record the practicalities of the making process, which are often simpler to show visually than describe verbally, such as the way to hold and manipulate a tool.
Ethnography [17] identifies and describes the activity of a social unit as "textual reconstructions of reality" [18]. Recently this has been applied in workshops, with examples in carpentry [19], glasswork [20], and textile manufacturing [21]. The Living Human Treasures program of UNESCO focused on talented tradition bearers and practitioners who, through ethnographic documentation, could contribute to the transmission of their knowledge and skills to the younger generations [22].
Cinematography and digital technologies facilitated the recording of directly comprehensible, visual examples. Video dictionaries of crafting gestures were proposed in [23]. In [24], a worn camera was proposed to capture the viewpoint of the practitioner. The 3D recording of crafting motion was used to enable any viewpoint in [25]. In [26], digital ethnographic tools were proposed.

Prior Work on Craft Representations Adopted by This Work
The Crafts Ontology (CrO) in [27] is based on CIDOC-CRM [28] and enables the formal representation of crafts. It is an application ontology obtained by integrating several existing ontologies, notably (a) the CIDOC-CRM, a top ontology and an ISO standard forming the conceptual backbone of the CrO, (b) the Narrative Ontology [29,30], a domain ontology for the representation of narratives, (c) the FRBRoo, a domain ontology for bibliographic records, resulting from the harmonization of FRBR with CIDOC-CRM and (d) OWL Time [31], a domain ontology recommended by W3C for the representation of time.
We use the CrO to formally represent crafting processes, as in [10]. The digital narratology approach in [9] is adopted for the representation of narratives that contextualize artifacts, collections, and techniques. Furthermore, it is used to model the stories comprising the technological, social, and historic context of crafts.
The Mingei Online Platform (MOP) [9] is a Web-based authoring platform that integrates the aforementioned craft process and narrative models and provides a GUI for the instantiation of craft representations. This GUI facilitates this process and user collaboration, allowing different access roles. The MOP is built on top of ResearchSpace [32,33], a research platform that introduces modules for CIDOC-CRM and integrates a repository of digital assets that assigns IRIs to the resources managed. For the resources that have other identifiers in Wikidata or any other knowledge base, connective links are asserted in the MOP knowledge base. Each digital asset is endowed with metadata, including semantic annotations from controlled vocabularies for objects, activities, and locations.

Step 1. Understanding and Recording
A mapping of the topics and knowledge items to be documented is established. The pursued outcome is an identification of the entities required to comprehend (a) the practice of the craft and (b) the social and historical context of this practice by a community.
Prior knowledge on the topic catalyzes discussions with practitioners and communities. This preparatory task regards secondary source research. The targeted sources provide an identification of the basic concepts and their appellation. Recordings of objects (endurants) and events (perdurants), such as human actions, complement the documentation by visually capturing the craft tools, workshops, machines, and products.

Understanding
This task then enables understanding of the craft practice and context, through interaction with practitioners and CH professionals, in particular through (a) collaborative creation workshops, where TC communities are facilitated in the definition of the craft roles, traditions, and contextual elements of the craft, and (b) ethnography to comprehensively document the crafting processes is proposed. Practitioners explain and demonstrate the making of artifacts, identify the required skills to do so, and recommend a teaching or demonstration process. Communities and CH professionals explain the history and social context of a community and the traditions followed within and along with craft practice. Preparation includes stakeholder mapping to identify stakeholders and identify the pieces of information to be sought from each stakeholder group.
Understanding craft practice requires the identification of at least the following types of concepts and their appellations. Physical items are materials, artifacts, and manufacturing tools or machines, protective or traditional clothing, and workspaces. Actions refer to the transformation of materials into intermediate or final craft products. Quantitative data include the measurement of quantities relevant to craft practice such as the amounts of materials, number of workers, as well as production and economic figures. Narrations and stories of direct relevance to craft representation are collected. They are selected based on the historical events that have influenced the history and economy of social groups, technological advances, and artistic movements relevant to the craft of study. In [34], ways to identify pertinent topics are proposed. The output of this task is preliminary documentation, in the following forms. For the crafting process, a vocabulary of (a) physical items (i.e., products, materials, tools, and machinery) and (b) crafting actions for the transformation of materials into products. For the craft context, a collection of narratives pertinent to the (a) evolution of the craft's genre and technological evolution and (b) the social and historical context of the practice of the particular craft instance at a place and/or by a community. Stemming from the above, a catalog of recording, or digitization, of targets is prepared, for the next task. Optionally, objects of historic or symbolic significance to the story of a craft can be appended to the digitization targets. For all of the above, references to sources where each piece of information was discovered are included.

Recording
This task concerns the recording of objects (endurants) and events (perdurants). Recordings of endurants are photographic documentation and 3D digitization. Event recordings are acquired through audio, video, and 4D reconstruction. The digitization also includes contextualization items, literature, and other information-carrying media, e.g., documents, technical drawings, images, and audio recordings. The digital assets are associated with technical metadata to interpret them appropriately as computer files. The multiplicity of types is facilitated by the concept of digital assets, which are linked to the knowledge base independently of their file type and storage location.

Endurants
Items of interest are materials, tools, machines, products, information carriers, and sites that are related to the craft practice and its context. The conventional recording of endurants is through photography and 3D reconstruction. Comprehensive guides in the photographic recording of CH artifacts and sites are recommended for this purpose [35,36]. For 3D documentation, in [37] the data processing pipeline is comprehensively explored, along with recommendations for the use of conventional digitization modalities. The digital assets are images and 3D models of structure and appearance, typically encoded as 3D textured meshes.
The digitization of endurants regards structures of a wide variance of spatial scales, in indoors and outdoors environments. The choice of 3D scanning modality depends on subject size, material, and environment type. Often digitization employs multiple scanning modalities, each operational in a specific scale or environment. Rooms and outdoor areas require the combination of laser scanning and aerial photogrammetry to systematically cover. For smaller artifacts, photogrammetric reconstruction and active illumination sensors are nowadays simple and widely accessible. In [38], an approach to the full scope of 3D data curation through the collection, processing, archiving, and distribution from multiple modalities is proposed.
In some cases, the historic or artistic significance of artifacts requires their meticulous digitization for conservation purposes. In other cases, as in the case of ordinary tools, it is only the structure that is of importance to the representation and a less detailed model may suffice. In most cases, post-processing is required when using multiple scans, to spatially register them and harmonize their appearance.

Perdurants
Digitization of perdurants mainly regards the craft practice and archive data regarding demonstrations of extinct techniques or documentary recordings of historic events. The collection of the latter follows the conventional practices of linked audiovisual data in the semantic web.
For craft practice recordings, the events of interest are practitioner postures and gestures. The digital assets are audio, video, and 4D motion recordings, which we call "animations". The applicability of MoCap and video modalities depends on the type of environment. Inertial MoCap is more suitable than optical in the cluttered space of workshops, due to reduced installation requirements and independence to occlusions. For the digitization of motion using video, markerless visual methods are employed [39]. Albeit significantly less accurate than MoCap recordings, they require solely a camera and are the only way to treat archive video.

Parts
Objects and animations have parts of individual importance. For objects, spatial regions may be associated with meaning, e.g., a symbol, or an affordance, e.g., a handle. Human motion is interpreted in actions, gestures, or postures per body member. The articulation of static or dynamic scenes in, respectively, spatial or temporal parts, contributes to the analysis of the recorded object or activity. To facilitate the segmentation of animation recordings into body postures and gestures, an editing tool was developed [40].

Technical Implementation in the MOP
The collected digital assets are linked to the knowledge base as "media objects". These recordings are abstracted as media objects when inserted in the knowledge base. Media objects are the way the MOP treats digital assets. New digital assets that are not web resources, i.e., new media objects, are made such by uploading them to the MOP repository, associating them with new URIs, and assigning them to the Web server of the MOP. They are identified by an IRI and served by the MOP Web server via the HTTP protocol. No operation is required for resources already online. When data collection is complete, all media objects are similarly treated, albeit their storage location is transparent to the user.
Though digital assets are abstracted in the MOP, the format in which these are encoded is relevant to the sustainability of their digital preservation. In this respect, it is important to use open formats and provide metadata on digitization parameters, so that they are convertible to future ones.

Step 2. Knowledge Elements
The basic elements of the craft representation are instantiated. These elements are the conceptual entities identified in Step 1. For crafting dimensions, these are the materials, objects, places, actions, and products involved and can be found in the vocabulary and storyboard. For contextual dimensions, these entities are the places, persons, events, and objects mentioned in the collection of contextualization narratives. The types of the basic knowledge elements are Product, Person, Material, Object, and Event.
The instantiation of knowledge elements refers to the creation of a record for each via the assertion of semantic metadata and relations, as well as the linking of digital assets. Thus, knowledge elements contain curated information encoded as knowledge statements and links and their instantiation is a digital curation task. The user task includes the selection of the entity type, the provision of the semantic metadata for that entity, and the linking of digital assets relevant to the entity.

Conceptual Usage
To introduce the conceptualization of knowledge elements and our event-oriented representation, we present two examples. The first regards a contextual event and the second a crafting event.
The first example illustrates the event of a building acquisition, which hosted a craft workshop that we study. The building was built by its original owner and then sold to another person. The building was an equipped textile manufacturing workshop. In Figure 2, we illustrate the digital assets gathered from Step 1. To instantiate the event, we need the basic knowledge elements about the place of the event, the date and time, and the participants. These are illustrated below, as entities for H. Gotzes, G. Diepers, and an address at Krefeld, Germany. We underscore the description of the building, which goes from generic (building) to more craft-specific (workshop). The location is a knowledge element that is referred to by other entities, besides the event which we study, such as the construction of the building or its renovation (shown only for illustration). The figure also illustrates the use of the geographical database of GeoNames. Given the association of this location with Krefeld, it is possible to reason that this location is in Germany and Europe. Finally, the figure illustrates two more events (the construction and a renovation event), which are implicitly related in the knowledge base, through the location and the participants. The second example is illustrated in Figure 3 and comes from the glassblowing craft. The term "bubbling" refers to a glass-forming technique, where the practitioner insufflates air into a gathering of molten glass with the aid of a blowpipe to create a glass bubble. In the illustration, the representation of the event is linked to the required knowledge elements. The association of digital assets with knowledge elements exemplifies objects and actions. In our knowledge base, the blowpipe tool is associated with two usage postures: standing (free blowing) and sitting (using stabilizing supports). These postures and performances are documented by (i) a semantic description and (ii) digital assets that exemplify the posture and the action. In the example, the first case is linked to three digital assets.

Basic Knowledge Elements
Each knowledge entity has one unique system handle, which can be in any language. Each entity may have an arbitrary number of appellations to support multiple languages and idioms.

Events
Following [28], we call events the changes of state in cultural, social, or physical systems brought by phenomena or influenced by other events. In the ontology, events are instances of class E5 Event. We are interested in two types of events, which are used as the building blocks of contextualization narratives and crafting processes. Respectively, they are called "contextual" and "crafting" events and are used to represent changes in (a) social, technological, and economic systems and (b) materials that are transformed into craft products, respectively. Events are organized in the ontology by three types of relations: (1) Mereological, relating events as parts of other events. This allows the hierarchical organization of events in a coarse-to-fine detail analysis. (2) Temporal, enabling the representation of the order or simultaneous occurrence of events as described by Allen's temporal logic [41]. (3) Causal, relating events that in normal discourse are predicated to have a cause-effect relationship in the curator's opinion.
Events may relate to an arbitrary number of places, objects and locations. Persons and social groups are related as participants to the event. Places are related to the place(s) of occurrence. Objects and materials may be used or produced in the event.

Materials
Materials extend class E57 Material. In our ontology we extend the class with an optional Place attribute to represent the material origin, which can be relevant to the quality, reputation, authenticity, and cost of the produced craft article.

Objects
Objects are used to represent craft products, tools, machines, and other objects of historical significance in craft contextualization. Both extend E22 Man-Made Object. Pertinent digital assets, such as photographic and 3D documentation, are associated with the entity. The entity contains optional links to the Materials that compose it as well as a Place for its location. Through the representation of the crafting process, craft products can be associated with the crafting process, its place of occurrence, and the practitioner(s) involved in the process.

Places
Places extend class E53 Place and may contain one or multiple links to geographical locations. For crafting processes, both the geographical location and type of the place are of relevance. The geographical location is relevant to the origin of materials or the manufacturing site of craft products. The classification of places is of relevance to represent the type of place where crafting processes occur, such as a workshop or a harvesting site. For contextualization narratives, geographical locations enable the documentation of historical events. In addition, some places may be imaginary, as in legends or tales, and are not associated with a geographical location. The implementation is integrated with a database of location names [42], which provides geographical coordinates and hierarchical organization of locations into cities, regions, and states. The definition of arbitrary, user-defined locations is also supported.

Persons and Social Groups
Person entities extend class E21 Person and are used to refer to individuals. Social groups refer to organizations of individuals and extend E74 Group. In process representations, persons and groups are used to represent individual practitioners and workgroups. Specific individuals are of relevance as participants in crafting processes. Furthermore, persons are semantically classified as to their role in that process. In narrative representations, person and social group entities are used for historical figures and social groups, such as companies, parties, guilds, or art movements.

Technical Implementation in the MOP
In the MOP, dedicated forms for the composition of knowledge each knowledge element type are implemented, as supported by the CrO. Several efficiency facilities are provided for the interlinking of digital assets and locations. For all knowledge elements, their semantic annotation using the Getty Arts and Architecture Thesaurus [43], the UN-ESCO Thesaurus [44] and, optionally, additional thesauri from National Documentation Centers. For location names, we have integrated a geographical reference [42].

Step 3. Representation
The entities represented in Step 2 are related to craft representation. This includes the representation of pertinent crafting processes and contextualization of events contributing to narratives. Both representations are comprised of events organized by relations. Crafting events occur each time an individual, handcrafted product is made. We say that all the expressions of a particular crafting process follow the same schema. In this step, this schema is represented according to [10]. The schema of the classes introduced in this section is elaborated in Appendix A.

Fabulae
Craft context representation is based on the collection of narratives defined in Step 1. For this purpose, we adopt the computational treatment of narratology proposed in [30]. Central to this approach is the term "Fabula", which denotes the series of events that make up a narrative, in chronological order. The Fabula may contain parallel and simultaneous events but has only a single version.
Each narrative is individually modeled, using the events that comprise its Fabula and which were created in the previous step. Common events can be shared across fabulae. Events are linked inside the Fabula by mereological, causal, and temporal relations. Mereological relations are important to analyze complex events into simpler ones. Causal relations encode the curator's account of events. The Fabula of each narrative is instantiated using the MOP.

Process Schemas
A process schema encodes the actions to be performed to execute in a particular crafting process. Examples of schemas are wedding ceremonies, recipes, or soccer games. A defining feature of crafting process schemas is that they contain branching points in their workflow. The decision to take one action or another relies on the judgment of the practitioner and can be influenced by a range of factors.
Visual, unambiguous, and formal encoding of schemas is borrowed from UML activity diagrams [45] to unambiguously encode process schemas. The visual nature of activity diagrams facilitates the collaboration between technical and heritage partners.
Following the conceptual segmentation of the crafting process created in Step 1, process schemas are analyzed in process schema steps. The analysis ends when a targeted level of detail is reached. This level is set to that of simple actions that practitioners can identify. This recursive decomposition yields a hierarchy of schemas that start from a coarse description down to a fine analysis of elemental actions. The finest sub-steps are individual postures and elemental gestures, such as "grasping a hammer" or "a strike of the hammer upon a nail".
In the schema and activity diagram, crafting events are connected by transition nodes (Simple, Fork, Merge, Join, and Branch), which suffice for the definition of crafting workflows. Fork, Merge and Join nodes enable the representation of simultaneous and synchronized actions, performed by multiple persons or body members. Branching nodes represent conditional actions of the practitioner, due to which executions of the same process schema may lead to the execution of different actions. Transition nodes relate process steps temporally and causally, while sub-step decomposition organizes them hierarchically.

Technical Implementation in the MOP
The process schema is transcribed into a semantic representation, using the MOP. The authoring of both process schemas and fabulae is supported by the MOP. In Figure 4, the initial panels for authoring a Fabula and a process schema are shown. Detailed information on Narrative authoring is provided in [9], while the semantic representation of process schemas in the MOP is covered in depth in [10].

Step 4. Narratives
Narratives are authored and semantically represented. Narratives implement the ways that fabulae are presented or narrated. Contextual events are used in narratives and are events that have occurred in the past. Narratives are represented following [29,30]. In [9], these ideas are implemented in the MOP.

Narratives
Narratives implement the ways that fabulae are presented or narrated. A narrative may have multiple narrations. Each narration contains references to events of the Fabula in some particular order that is not necessarily chronological. Individual narrations may differ because we wish to present them in multiple ways, for various audiences, and through a multitude of presentation modalities. Individual narratives may focus on particular subsets of the Fabula. Each narrative contains a reference to a Fabula and a reference function that determines the order by which events in the Fabula are narrated. In this way, knowledge elements and digital assets associated with the event are accessible to the narration. Individual narrations of a narrative may be adapted per language, age, or special needs. Last but not least, narrations may differ due to their adaptation to the format of the presentation medium, whether it is audio, visual, textual, or multimodal.
To represent the orchestration of stimuli presented during a narration on a particular medium, we introduce the class of Presentations. For each event referenced by the narration, a presentation segment is authored. This segment contains the links to the digital assets to be presented during the narration, or their spatiotemporal arrangement. Authoring presentation segments is facilitated by the MOP, through the retrieval of knowledge entities and digital assets.

Processes
Processes are comprised of the events that occurred during the crafting process. If these events conform to a process schema, we identify it as a process that executes that schema. A process is one of the potential flows of a process schema and, thus, two different sets of events may still conform to the same schema.
Like a Fabula, a process contains a reference function, which identifies and chronologically orders the events that occurred during its execution. Likewise, the knowledge entities and digital assets linked to these events are available for the presentation of the process. The digital assets are recordings of the events that took place and the digitization of the objects involved in them.

Technical Implementation in the MOP
Multiple media can be employed in the delivery of process and narrative presentations, starting from verbal and visual and reaching up to immersive and interactive narrations. The authoring of presentations in segments that follow the narrative events and crafting process steps is facilitated by the MOP, which retrieves the knowledge elements and digital assets in pertinence to each event.
The presentations entail the spatiotemporal orchestration of stimuli for each segment, in individual media channels. Authoring of narrative and process presentations is relevant to the medium, as it imposes requirements and constraints on the format and content type. The MOP provides native support for web pages, verbal narrations, and audio guides. Figure 5 shows the primary form in the MOP for creating a narrative based on a Fabula or process schema and the way that events are presented within the Fabula.

Step 5. Presentation
Craft presentations are built on top of events and event schemas referenced through the narrative and associated with knowledge elements and digital assets, which can be retrieved to illustrate the narration. Narrations are associated with events and, in turn, with knowledge elements and media objects. Alternative presentations of these narrations are enhanced with objects and actions reproduced in 3D, in a virtual environment. In this environment, human action is re-played from motion recordings by animations performed by virtual humans. In this way, the human motion and handling of tools and machines can be seen from multiple viewpoints and better understood.
For all the presentation modalities of this step, the MOP is utilized as an infrastructure to provide content to the presentation tools. This content can be digital assets, knowledge elements, narratives, and demonstrations. Machine interpretability offered by the MOP ensures that the collected knowledge and digital assets can be accessed through third-party applications. In this way, content can be directly imported and re-used in these applications. Besides conventional digital assets, the knowledge base contains semantic representations of narratives and motion-driven narratives that can be queried and systematically accessed by software clients.

Place-Oriented Presentation Modalities
The presentation of environments is facilitated by spatial presentation modalities. The geographical organization of narratives in Step 4 can be directly utilized in presentation modalities oriented to the presentation of geographical and spatial information. Spatial representations are enriched with locations of interest that are associated with knowledge elements in the CrO and occurred as predetermined locations on the map. These locations and associated digital assets are retrieved directly from the knowledge base or other repositories. This way, spatial presentations can be enriched with any type of digital asset in the knowledge base, such as 3D reconstructions, historical information, video data, etc.
For large-scale representations, map-oriented presentation modalities at the same scale but relevant to the local environment and its resources 3D geophysical maps are proposed [46]. This is demonstrated through the implementation of two modalities. The first is an immersive projection room and its standalone PC version that overviews 3D environments, interactively navigating from bird-eye viewpoints (see Figure 6a). The second is a lightweight presentation of a 3D geophysical map in mobile platforms (see Figure 6b). For inspecting or navigating through 3D environments, a 3D virtual reality (VR) viewer can be employed whether on the computer screen or through an immersive visualization system (i.e., wide-projection, VR headset). In particular, software that produces video output for a predetermined walkthrough of the virtual camera in the 3D environment is required. The primary video can then be enhanced with narration, subtitles, music, etc. This is demonstrated by the two examples below. The first regards the craft of dry stone walling [47]. In the example, traditional settlements for moving husbandry [48], which were built by dry stone walling, are presented via a guided virtual walkthrough. The scene is composed of multiple scans, to support viewing both the inside and the outside of the structures (see Figure 7a). The second example regards the architecture of mastic villages at Chios. The overview shows how the architecture served as fortification of the village, for protection against invaders and pirates (see Figure 7b).

Virtual Exploration
More complex presentations can be implemented through a virtual exploration of the natural environment from a terrestrial viewpoint enhanced with 3D reconstructions of heritage sites and landscapes. This is demonstrated by a 3D game [49]. The terrain of the game was generated by importing the geophysical map into unity3D. Then, flora and fauna were imported together with 3D reconstructions of villages. The concept of the game is to explore the island in different eras and acquire knowledge about the cultivation and trade of mastic in each one, through a mission-oriented approach. Digitized rural environments are also important to the realism of such application. The game is enriched by re-using the digital representations of garments, machines, and original tree structures that were acquired from the Mastic Museum at Chios (see Figure 8).

Virtual Exhibits in Real Places
Virtual museum modalities present collections of digital assets in a physical space, bearing in mind that (a) several museums have a large collection of artifacts in their storage, but no room to display them, and (b) to represent past period exhibitions, to virtually re-visit them, after they have ended. This is demonstrated by virtual museum applications both through the web and physically installed in the museum space ( Figure 9).

Virtual Storytellers
Virtual storytellers enhance the museum visit with narrations of the life and events of people from the craft community. This is demonstrated by enhancing the factory space of the Mastic Museum of Chios with eight virtual humans, workers of the factory that narrate their life and stories [50] (Figure 10).

Web-Based Narratives
The primary presentation method in the MOP is a form of documentation that maps the contents of a craft representation and provides an overview of the craft and its context. This is provided as a digital, online presentation that provides initial orientation on the craft and its context in the form of an introductory narrative, which provides an overview of the craft instance, in an overview narrative. From this presentation, all narratives, assets, and events are available in multiple views. Such views are temporal (timeline), geographical (map-based), event-based (calendar), object-based (galleries of artifacts, tools, etc.), personbased (galleries of key actors), and so on. All of these are available through "templates" in the MOP [9]. The authoring and presentation of Web-based narratives through the MOP are presented in depth in [9]. An example is shown in Figure 11.

Augmented Physical Artifacts
Object-oriented presentation modalities can occur by enhancing contemporary TCs objects with digital stories that bind them with history, society, traditions, and values. Such stories can support new interest in TCs objects by enhancing their meaning, value, and uniqueness. This is demonstrated by a handbag that stands both as a woman's accessory and as a contemporary craft creation that can be experienced to reveal its unique history and identity [51] (see Figure 12). The handbag is augmented by an augmented reality (AR) application that is providing three layers of pattern recognition. The first layer shows the bag which has its own story, its creation. The second layer has multiple patterns which reveal the story of textile weaving in Krefeld. The third layer shows that the stories of the patterns have a story of their own.

Mobile Clients
Mobile devices can provide access to the represented knowledge before, during, and after a visit to a craft-related heritage site. Presented information includes textual narrations, events, videos, images, 3D reconstructions associated with the location, and serious games to present craft processes and techniques (see Figure 13).

2D Presentation of Actions
Two-dimensional visualization of actions is particularly useful for conveying motion and instructions on the printed matter [52] and physical surfaces, opening a wide avenue of applications that involve physical objects and surfaces (i.e., mixed reality). Our approach is to use useful concepts from the world of art, in the world of motion visualization. Our goal is to utilize the insight that such visualizations provide to efficiently create meaningful visualizations of human motion, on 2D media. The reason we choose art for this purpose is that artists have long studied the perceptual appeal of motion visualization methods over centuries of experimentation. Painters, illustrators, and directors use motion lines, contrast, and superimposition and juxtaposition of visual frames to facilitate the mental recreation of the depicted motion by the observer. Abstractions, such as motion lines, provide insight into understanding the motion.
To ease this user task, the MotiVo computer-aided authoring system for the visualization of human motion was developed [53]. MotiVo is an authoring system that simplifies the process by offering several visualization tools as integrated components. Using those tools, motion is visualized by parameters, such as the blending of key poses of activity, the visualization of motion trajectories, the application of image filters to visualizations and their combinations. Using MotiVo illustration on glassblowing was implemented, which presents a carafe glassmaking process in the form of a graphical story, enriched with verbal content and visual annotations [54] (see Figure 14).

3D Presentation of Actions
Virtual humans can be used to present craft processes, as well as to demonstrate tool and machine usage. A motion vocabulary item is used to represent an instance of a specific motion segment. These items can be combined and interleaved to represent entire procedures and are considered building blocks of a motion vocabulary. This vocabulary can, in turn, be used to create "sentences" that encode actions and procedures. The vocabulary can be demonstrated through visualizations of practitioner actions and abstractions of tool usage configurations [55] (Figure 15).

Skill Acquisition and Training
First-person acquaintance applications introducing basic skills are implemented, which will allow the manipulation of virtual and real objects and tools. Craft gestures that include tool handling are valuable for training purposes. The interactive training session may occur in the physical or digital environment. This can be demonstrated through a mixed reality installation at the museum of CNAM and in VR using controllers to enable the grasping and handling of tools and with the usage of the actual tools and real-time feedback on craft actions [55] (Figure 16). Training, in this case, refers to the handling of tools for cultivating and harvesting a mastic tree (see Figure 17). The game was built on top of a VR training SDK [56] and guides the player through the process with instructions. The player has then to select the appropriate tool and follow the instructions on how to use it to complete the task at hand. In the example, the task includes cleaning the soil under the tree (left) and creating incisions at the highlighted locations on the tree trunk (right).

Step 6. Preservation
Supporting access and acknowledging CH related to TCs is the first step towards its preservation. TCs have traditionally been associated with individual practitioners, as well as with part-time activities or family businesses. Nowadays, many individually trade their craft products online. The simplification of learning and digitization of promotional content support the operation and encourage the foundation of small enterprises. Thematic tourism makes use of engaging content, experiences, and training, and can enhance the sustainability of TCs.

CHIs and CCIs
The sharing of digital assets and knowledge on TCs provides a resource for researchers and the general public, by availing the opportunity to makers and communities to promote their work. Narratives create new content, stories, and experiences, for content owners, craft makers, and TC communities. Corresponding applications raise interest in broader audiences, including the general public, visitors, tourists, and prospective apprentices. Storytelling and mobile technologies and installations provide comprehensive and captivating TC presentations. Benefits for CHIs and CCIs are the re-use of digital assets in new content and new presentations on TCs, and new ways of appreciating and experiencing TCs, which are important in reaching new collocated and distant audiences.

TC Education and Training
TC training is important, as the ability to teach a TC enables its preservation. Off-site training can be facilitated through the MOP. Corresponding applications show how TCs have been practiced, and include insightful annotations of digital assets, such as illustrated instructions and motion summarization. On-site training is implemented by mixed reality installations and mobile devices where visitors are introduced to a TC through practical tasks, craft gestures, and the use of tools.
In Craft Centers, Maker Spaces, and FabLabs, TC practitioners and LHTs are engaged in the physical making of artifacts to reconstruct lost information on the making of artifacts, when no documentation or testimonies are available anymore. Re-creation of techniques on artifact making or material treatment can be captured, documented, and evaluated as the potential implementation of techniques, shedding light on lost knowledge on craft processes. Educational applications on TCs can contribute to contemporary issues of craft education [57]. In this work, a first step is taken by offering well-documented, introductory TC training experiences, to raise interest and attract potential apprentices.

Thematic Tourism
The wealth and variety of expressions and forms of ICH are steadily becoming a principal motivation for travel around the world. Many forms of cultural tourism are associated with a longer duration of stay [58]. The UN World Tourism Organization (UNWTO) recognizes that an important challenge lies in identifying, protecting, and safeguarding ICH by investing in sustainable tourism development, in consultation with local communities and other stakeholders [59]. In line with the Faro Convention, the value of TCs in "sustainable development, cultural diversity and contemporary creativity" [60], means that profits will contribute and motivate the preservation of TCs.

Case Studies and Lessons Learned
We have applied the proposed protocol in three case studies. For each, an online presentation pilot showcasing the obtained representation was developed. In addition, for each pilot, several installations were developed showcasing the craft in CHIs.

Glassblowing
Glassblowing combines hand and body gestures, the sense of gravity, and a thorough understanding of the material viscosity. Glasswork is a challenging craft because, during production, glass changes state from liquid to solid and vice versa, posing handling requirements. We use glasswork to study dexterous aspects in tool manipulation. The online presentation of this case study can be found in [61].
Results were deployed in the context of a periodic exhibition at the Conservatoire national des arts et métiers (CNAM). A dedicated space at the cathedral, which is part of the museum, was dedicated to the exhibition. For training, a bench was installed in front of the main display together with a glassblowing pipe to be used by visitors. A digital application was installed dedicated to craft training. This application enables users to mimic craft gestures using the tools and the bench. The scenario that was set up was that the presentation application sequence is executed first and then the main projection switches to the craft training applications so that users test what they learned from the craft presentations. The installation is described in detail in [55].

Silk-Weaving
Weaving is one of the oldest crafts developed by humans. Due to the industrial importance of textile manufacturing, there exists an extensive literature on weaving techniques. Thus, in this case study, we focused on the challenges of modelling the rich social and historical context of this particular craft, as well as finding modern and appealing uses of traditional textiles and motifs. The online presentation of this case study can be found in [62].
The historic context regards protoindustrial Germany and how the religious asylum that Krefeld offered to victims of religious prosecutions in the 17th century eventually made Krefeld an industrial center. The story is told through the evolution of one of the oldest companies in Krefeld, the "Paramentenweberei Hubert Gotzes Textilien". The workshop of that company is now an industrial monument called Haus der Seidenkultur and is operated by volunteers that preserve the craft for posterity. The application is oriented in ten spots in the museum, each one connected to a narration that is accompanied by an audiovisual presentation. From a technical perspective, the narratives are retrieved by the MOP. Narrations by virtual humans enhance the presentation of narratives [63]. Details on the installation can be found in [64].
To investigate the combination of traditional motifs and textiles in modern contexts, a contemporary accessory was designed, studying the historical motifs of HdS and transforming them into text-based narratives. This was in line with one of the most important goals of HdS, which is to raise awareness regarding the unique CH possessed by the museum and at the same time connect its legacy with history and tradition. Each motif is linked to a narrative. An AR application provides three layers of pattern recognition, through the camera of a mobile phone. The first layer regards the bag which has its own story, its creation. The second layer regards individual patterns on the handbag that reveal the story of textile weaving in Krefeld. The third layer is the traditional stories linked to the patterns. Upon recognition, these narratives are presented audiovisually to the user. Details on the implementation of the system can be found in [51].

Mastic
This case study explores an indigenous craft, practiced in the south of the Island of Chios, Greece, where a single type of tree has shaped the local trade, culture, and built environment. The resin harvested from the mastic tree is used for a wide range of products, mainly culinary, but also including skincare and medicinal. The knowledge of mastic cultivation, on the Island of Chios, is inscribed on the Representative List of Intangible Cultural Heritage of Humanity of UNESCO. The culture of mastic represents a comprehensive social event for the community which considers mastic as part of its identity. The online presentation of this case study can be found in [65].
The protocol presentation technologies were used to implement an interactive exhibition at the Chios Mastic Museum of PIOP. The main areas of intervention were the main exhibition space of the museum and the rural space outside the museum. In the main space of the museum, this work targeted the presentation of mastic processing. In the rural space of the museum, there is a mastic field with metal sculptures that represent everyday people in the mastic field. The exterior interventions regarded the usage of AR for the presentation of craft activities in the mastic field. At the locations of interest, virtual humans narrate life stories of daily life and work at the factory. One of the requirements of presenting the craft was to display its seasonality. Visitors of the rural space outside the museum can experience mastic cultivation in the field through AR. An application recognizes metallic sculptures that exist in that space, which "come to life" and present mastic cultivation activities. Details of the installation can be found in [50].

Replicability of the Proposed Method
Replicability can be judged from the application of the TC representation protocol in three use cases. In all cases, we were able to apply several scientific approaches for each step and combine the results in a single representation. This became evident both in the craft understanding phase, where several approaches to studying social and historic context were applied, and on the data collection step, where several scientific methods were used for data acquisition. Regardless of the heterogeneity of the results, the MOP has proven sufficient as a representation and no surprises were encountered on the representation of processes and narratives. Another supportive finding for the replicability of this methodology was that, through detaching the representation from presentation in this work, multiple presentation instances were created from different technologies. This was achieved by applying the exporting functionality of the MOP in the representation and leaving it to the developers of the presentation layers to judge the most appropriate way of facilitating the representation in their presentation context. Finally, the included web-based presentation modalities in the MOP allowed the direct preview and dissemination of the represented knowledge through the web.

Collaboration
One of the main challenges faced by the application of this methodology was the need for several scientific disciplines to work together under a unique methodological framework. This was indeed challenging since different scientific approaches, technical tools, and research methods were applied. In this challenge, we learned that the MOP as a single point of representation of research data greatly enhanced the collaboration of the team as it allowed different scientific disciplines to report and document results under a uniform semantic representation. Of course, several adaptations in terminology had to be made for the entire team to have a common understanding of the represented data.

Iterative Refinement of the Representation
Linear execution of protocol steps meant that the entirety of digital assets would have been acquired a priori. However, knowledge acquired in the second step in some cases referred to non-digitized items, which were only then identified, and needed to be digitized as new digital assets in the context of the first step. Moreover, more sophisticated digitizations of assets were acquired later in the timeline of this work, judged so by CH professionals. Thus, the main lesson learned was that the linearity in these steps can be disrupted by refining iterations when needed. However, the methodology was found to be very adaptive to such iterations, considering that new needs resulted in the application of the previous steps only for the acquired assets and their context.

TC Representation Protocol Steps
Lessons learned per application of each protocol step are presented below. Craft understanding: TCs are deeply rooted in the social and historic context of the communities practicing them. Through history and archival research, many dimensions were discovered that led to the formulation of social and historic narratives. Furthermore, the TCs surprised us by revealing hidden dimensions spanning from art elements to historical and social findings. These largely affected both the work conducted in this phase and the consequent phases multiplying the representation and data collection needs.
Data Collection: Sometimes interest in an artefact may rise after its digitization, as when inspected more closely a digital asset may intrigue researchers to revisit to craft understanding to acquire knowledge on the specific subject that was not covered. From a technical perspective, considering that several digitization technologies were employed we learned that in some cases specific requirements of a study could lead to the need for new digitization techniques.
Craft representation and process representation: In these phases, it was important for the team to evaluate the capacity of the online platform to represent the vast amount of knowledge acquired. This posed two major challenges. The first challenge was the curation time needed to represent resources, and the second was the expressiveness of the representation to address complex knowledge elements such as processes. Regarding the first, we learned that significant effort is required from curators and that training and help are needed. Regarding the second, close work of semantic experts, developers, and practitioners were required to conceptualize the representation, create the authoring facilities, and train on the transformation of ethnography to process representations. The adaptation to these requirements led to several redesign iterations of the MOP.
Craft presentation and preservation faced two main challenges. The first was to create online presentations of the acquired representation so as (a) allow researchers to build narrations on the acquired representation, (b) disseminate knowledge through the MOP, and (c) create online educational material and process representations. The second was to ensure compatibility of the knowledge base with external sources both in terms of linking to external sources and importing in knowledge standards such as CIDOC-CRM and EDM. The lessons learned regarded the quality of our representation, which was capable of supporting several online presentations of the represented knowledge. Furthermore, considering that the system was based on existing knowledge standards, disseminating the represented knowledge was as simple as creating semantic associations of CIDOC-CRM encoded metadata to other knowledge standards and creating a SPARQL endpoint to deliver knowledge in standardized formats.
Preservation: The main lesson learned was that having a knowledge base and a rich representation including digital assets greatly enhanced the capacity to deliver results that could be experienced by visitors of a CHI. We learned that content makes a difference, as engaging narrations created through the web platform were disseminated in alternative means and modalities targeting a wide range of uses including information, education, and entertainment. The separation of the representation from its usage allowed us to use a plethora of technical tools to create different forms of presentations, thus unleashing the creative powers of UX designers and developers.

Conclusions
A special type of knowledge included in crafts is "felt" [66], "tacit", or "embodied" knowledge, or otherwise knowledge that is based on the sensory perception of practitioners and the understanding of tool and material affordances. Examples are the haptic sensation of a material (i.e., plaster dampness of the potter, or roughness of a textile), the sensations of heat and smell (i.e., in the glassmaking process), or the color of an object, which are exploited by a skilled practitioner. In the context of using the TC representation protocol, it is underscored that understanding a craft cannot be a theoretical only task. All of the narrations, documentaries, and VR demonstrators cannot recreate "felt experiences". Thus, besides conventional digital tools, acquired representations include the knowledge for craft re-enactment, through the meticulous representation of craft processes and techniques.
This work uses crafts as a first step to the digitization of intangible dimensions, such as purpose and causal dependencies, that are encountered in many vocational domains. The crafting dimensions explored in the context of TCs are not much different that the skills required by contemporary professions in which tools are used, such as mechanics, medicine, industrial assembly, electronic manufacturing, and others. The use of contextualization narratives can serve more demanding challenges, such as the representation of technological history, as well social and historical events and their influence on social groups and their practices.  ); validation, X.Z. and N.P.; formal analysis, X.Z. and N.P.; investigation, X.Z. and N.P.; resources, X.Z. and N.P.; data curation, C.B., H.H., D.K., E.T. and C.R.; writing-original draft preparation, X.Z. and N.P.; writing-review and editing, X.Z. and N.P.; visualization, X.Z. and N.P.; supervision, X.Z. and N.P.; project administration, X.Z. and N.P.; funding acquisition, X.Z. and N.P. All authors have read and agreed to the published version of the manuscript. Acknowledgments: The authors thank Reviewer #2 for meticulous and constructive criticism that led to the improvement of this paper.

Conflicts of Interest:
The authors declare no conflict of interest.

Appendix A
The tables below show the definitions of the classes utilized for the representation of crafting processes.