This session address the development of a significant new digital humanities resource in the form of a world-historical dataset. The session provides an overview of the project and details of two key elements in its early stages of development. The commentator will set this project in the context of other complex, multidisciplinary datasets. The Collaborative for Historical Information and Analysis (http://chia.pitt.edu) is a multi-institutional collaborative of scholars in humanities and in social, natural, and information sciences. The Collaborative, structured as a Research Collaborative, Headquarters, and a wider informal network, has recently received major support from the National Science Foundation, which has provided CHIA with an award in its Building Capacity and Community program. On the Research Collaborative side, the award is to strengthen the organizational and technical infrastructure of linking participating institutions that are collecting data on historical population, climate, and other topics with a crowdsourcing tool to demonstrate the feasibility of building a continuously growing collection of diverse historical data and metadata. On the Headquarters side, the award is to assemble and develop knowledge on repository design to develop a repository sufficient to house the incoming data and permit global and interactive analysis. The Collaborative for Historical Information and Analysis’s future plans include expanding its collection and processing of historical data, broadening its community of social and natural science researchers, analyzing historical patterns of global change, and sharing its resources with researchers, policy-makers, teachers and students. CHIA is headquartered at the University of Pittsburgh with formal affiliates at the University of California – Merced and the International Institute of Social History (Amsterdam) plus two affiliates at Harvard University (Institute for Quantitative Social Sciences and Center for Geographic Analysis). Other associated groups are at Boston University and Michigan State University (participating in the NSF-funded project) as well as the University of Portsmouth, the University of California, Irvine, and the Council for the Development of Economic and Social Research in Africa (CODESRIA).

Three papers discuss varying aspects of CHIA and its work. Patrick Manning of the University of Pittsburgh, director of the project, provides an overview of the project objectives, philosophy, structure, and its practical milestones. Ruth Mostern of the University of California - Merced, a member of the governing Executive Committee, presents on the issue of soliciting, linking, and evaluating scholarly datasets. Kai Cao, the technical lead on the project at the University of Pittsburgh, describes the work of creating the prototype archive.. The discussant, Ian Johnson, the developer of a parallel project at the University of Sydney, will comment on both generalities and specifics of the CHIA project as presented.

Big Data in history will provide a new, comprehensive level of documentation on the past. Currently available historical information, while immense in its overall quantity, is scattered and dispersed. Libraries and archives in great cities hold treasure troves of data on trade, politics, and religion for national and imperial centers, but each archive is separate from the others, and the totality of their records provides scanty information on people of rural areas. The idea of Big Data in history is to digitize a growing portion of existing historical documentation, to link the scattered records to each other (by place, time, and topic), and to create a comprehensive picture of changes in human society over the past four or five centuries.

The Collaborative for Historical Information and Analysis (CHIA), formed in 2011, now has five institutional members based in the U.S. and Europe, four additional, informal associated institutions, and expects to grow further through links with such organizations as the Council for the Development of Economic and Social Research in Africa (CODESRIA, based in Dakar). The Collaborative, directed by an international Executive Committee, is administered at the University of Pittsburgh. Its collaborative projects include the creation of a prototype and a full archive able to contain consistent and documented world historical data, along with systems for gathering and incorporating new data and for analyzing and visualizing the data. Data included are expected to begin with demographic, economic, social, political, health, and environmental variables, and to display their patterns and interactions. This presentation will trace the development of the overall project over the past five years and identify the main problems to be taken on in the next three years.

Tasks addressed from 2007 have included initial articulation of the objective, overall design of the research project and work on the initial steps of implementing several aspects of that design. In particular project members have emphasized recruiting and sustaining collaborating groups at several institutions. In addition, work has included collection of various sorts of historical data, design of the repository for worldwide historical data, development of an ontology describing world-historical data at various levels of detail, and systems of cleaning and documenting data.

At an empirical level, major advances have taken place in collecting, displaying, and linking U.S. data on disease, climate, and population.

The NSF-supported project includes three years of a projected ten-year project to create a global- historical dataset, with the hope that it would be taken over and sustained through the efforts of UNESCO or the World Bank. The CHIA project pledges to maintain open-source, open-access, non-proprietary standards throughout its work in constructing a world-historical archive. We have allocated the range of our activities among three basic missions. Mission 1. Gather historical data. Mission 2. Aggregate data up to the global level. Mission 3. Visualize, analyze, and mine the data.

Currently funded infrastructure projects include:

Additional areas of activity

To understand global social patterns as they exist today, it is increasingly clear that we need to understand how they have evolved over recent centuries. The Collaborative for Historical Information and Analysis responds to this need and takes historical analysis into the realm of Big Data. It is expected that the data resources will grow to several terabytes in size. This project will stimulate development of more efficient research collaborations, enabling systematic large-scale consolidation of diverse historical data sources. Once collected and integrated, the data repository and analytical system will allow scholars to address a wider set of questions testing hypotheses about long-term and short-term social change at the global scale and catalyzing an expansion of the evidence base in humanities and social sciences. For example, our understanding of important societal issues can advance by linking health to demography and by incorporating climate and health factors into economic studies. Disciplinary theory will advance through interaction among the various scientific fields, so that a global network of humanities and social-science researchers will emerge.

The project addresses the global dynamics in humanities issues and social-science variables over the past several centuries. Contemporary globalization and concerns about future global trends naturally raise questions about past patterns of global change. What were the interactions of population, economy, governance, and social inequality with each other and with climate and disease? Historical social science, focused at national and subnational levels, has scarcely addressed global issues. Our group expects to collect, document, and analyze historical data to permit cross-disciplinary analysis of human society over time. The overall topic is immense, but we believe we have found an orderly and productive way to work on it.

How are we to create consistent data at regional and global levels over time? Our group, rather than tunneling within a single discipline, seeks to coordinate data collection and research in multiple disciplines. We advance an explicit focus on the global and historical character of human society. The existing data sources are mostly used for regional comparative efforts; they vary widely in degree of consistency, reliability, completeness, as well as in data representation format. The task of large-scale data utilization can only be resolved via collaborative efforts within a large network of researchers.

What criteria distinguish this global strategy in humanities research from other large-scale projects? Our project is not simply to archive large quantities of data but to define and link them into a single overarching set of interacting, historical data. We require a coherent metadata framework to link data to their sources and each other. Creating this mass of new metadata—as we incorporate, integrate, and aggregate data—requires a strong ontological base and a crowdsourcing procedure to link many contributors. Our collection of base data on population worldwide is to go back four centuries and to include migration and other extra-census data—it is thus complementary to rather than competitive with the Terra Populus (University of Minnesota) collection of census data.

The Collaborative responds to an imperative of the current moment: the need to understand global social patterns not only as they exist today but as they have evolved over some four centuries. The program argues that some national resources should be put into research at this broad level, to clarify and diffuse a global strategy in social-science research. It also focuses on population and climate as key layers of global data. In research agenda, CHIA addresses human interaction with the natural world, global population change, patterns in social inequality, and local and global patterns of governance.

A signal characteristic of world-historical data is that much of it will need to be assembled piecemeal from datasets created by specialists. Some of these datasets, such as those which concern climate information, are quite large. However, at the global historical scale, even climate data needs to be integrated based on local and regional analyses. Ocean sediment samples, ice cores, or dendrochonologies may offer centuries of continuous information, but they concern particular locations. Census and epidemiology datasets often record information about millions of people, but they are episodic in nature and regional in spatial scale. Beyond these types of data, contributions range downward in size and upward in analytical complexity. The history of commodity exchange, for instance, requires meticulous reconstruction from bills of lading and tax documents that may be difficult to locate, trust, and make commensurable. One of the challenges of the CHIA effort concerns this work. The solution involves three tasks.

A key element of the overall project of the Collaborative for Historical Information and Analysis is construction of an appropriate archival system. Three CHIA affiliates collaborate in the initial stage of archive construction by creating a prototype archive and retrieval system based on a “faceted search,” to be created and tested by mid-2013. The collaborating groups are the World-Historical Dataverse at the University of Pittsburgh (Pitt) (with the author as lead developer), the Institute for Quantitative Social Science at Harvard (with Gustavo Durand as principal developer), and the Center for Geographic Analysis at Harvard (with Benjamin Lewis as developer). The archive will be based initially on the Dataverse Network (DVN), to enable linkage of multiple data files (which themselves include explicit spatial and temporal data) so as to develop data that can be searched so as to identify patterns at the global level and interactions among variables at various spatial and temporal levels. Multiple data files will be stored in the DVN system as “studies,” and will be accessed by the system of retrieval.

The “faceted search” is the key element of the archive. It is a search portal, which enables users to define selected data by space, by time, and by topic in a text box. The search is “faceted” in the sense that when the user adjusts the range on one dimension (e.g., space), the range on the other dimensions adjusts appropriately. Once the search criteria are entered, the program identifies the studies that meet the search criteria and displays their geographic distribution within the bounding box on the map. With that, the user can then explore the studies by clicking the dots /links visualized in the mapping area.

Four categories of data have been identified for this task. The four categories of development are:

A graduate researcher at the University of Pittsburgh has been engaged to collect these data.

In addition to the above outline, the presentation will address as many as possible of the details involved in constructing the archive, and will discuss how it is to be used in later phases of project activity. For each study, metadata are to be defined and implemented to ensure that each individual value is fully defined. As a next step, the localized files that are entered into the archive are to be aggregated in order to yield continental and global summaries of data. Besides, we will define and impose the elements of a mid-level ontology for the data, to define relations among all the topical categories, and allow users to use unstructured tags within it.

The archive and faceted search, once implemented, will enable users to envision the breadth of the world-historical analytical system, which is the ultimate objective of the work. The associated website will appear as a storytelling platform. We also expect to rely on social media to spread the word on the archive to convey the idea that its continued development can build a resource of global and interdisciplinary interest.

This archive and faceted search, after its initial development, is to be articulated with other aspects of the overall CHIA project. To begin with, based on the responses and evaluations of users, we will carry out revisions of the faceted search created in this Phase 1, advancing it to lower the barriers of entry for participants and expand the reach of the project. Further, we will encourage the collection of additional categories of data and develop models to improve the analysis of data. In perhaps the most important next stage, we will link the archive and faceted search to the crowd-sourcing data- input application under separate development at Pitt. With the linkage of these two applications, it will become possible for large quantities of data to be incorporated and integrated into the world- historical archive.

Many digital tools have been and are being developed aimed at transcribing, annotating and publishing editions of literary or historical texts making use of crowd sourcing for collaborative research. This panel discusses the question how well digital scholarly editions produced by such tools reflect the theoretical notions of digital scholarly editions, and how such may be assessed based on both empirical examination of current practices and text theory in the digital era.

The practice of preparing and producing digital editions is increasingly supported by purpose made and specialized digital tools, many of them involving crowd sourcing. Although an exhaustive survey and typology of these tools is still missing, by and large we can see that most of these tools are highly similar in functionality, text model, and editorial process. As such they express a fairly straightforward transformation from the physical book to a digital metaphor of the book, roughly along a trajectory of transcription, annotation and publication. Quite contrary text theory in the digital era seems to express a different scholarly ideal of representation of text, far more rooted in notions of instability (McGann 2001), fluid text (Bryant 2002), transclusion (Nelson 1982), text as process (Buzzetti 2002, Gabler 1987), transmedialization (Sahle 2010), and distributed editions (Zundert 2011) for instance. Also the usage patterns of emergent digital technologies and their applications such as Web2.0/3.0, crowd sourcing, cloud based services, open notebook science, data as service, multi device enabled layouts –to name but a few– seem to favor shaping the representation of digital text more in line with theory than with the practices of current scholarly digital edition tools.

If we follow Internet pioneer Cailliau (Cailliau 2012) information at our fingertips will become essentially undocumented, in the sense of not being a conventional cover-to-cover document, not even as a metaphor. Rather specific parts and facets of information will adapt to different devices and context, rather reminiscent of the concept of Nelson’s envisioned docuverse (McKnight 1991). It should therefore be critically examined if it is still opportune and adequate to speak of a digital edition as a document, and if a digital editing process should necessarily lead to a single or physical publication to serve maximum scholarly expressiveness. Overall the tools in use mostly let the metaphor of the book be inferred as the de facto model for digital publication. But as users and scholars choose and adapt new technologies and new forms of engagement with information, should scholarly publishing and digital editing follow these patterns of usage? Does an audience oriented approach such as the ideas on minimal and maximal editions expressed by Vanhoutte (Vanhoutte 2011) strike a middle ground between theory and practice? What is the intellectual loss or gain if the metaphor of the book prevails over using the medium of the Internet as an expression of text as process?

It is these kinds of questions about the theoretical underpinnings of the digital scholarly edition that arise at the intersection of shaping technologies, standing scholarly practice, and changing usage. Trying to establish some practices of quality a number of comparative studies have been conducted into transcription tools and crowd sourcing tools for digital editions, notably by members of this panel. Most of these studies have been based on analysis and comparison of functional requirements, usability aspects, and user feedback. However, a text theory based aspect of evaluation of such tools and editions is mostly lacking.

This panel will explore the issue of practice and theory based quality assessment of digital editions, building on the results of a comparative text theory based empirical survey of tools for digital scholarly editions the design of which is the subject of our preparatory paper presented at the NeDiMAH Expert Meeting on Digital Scholarly Editions held in conjunction with the 2012 conference of the European Society for Textual Scholarship. The panel focuses on several prominent digital tools and projects for preparing digital scholarly editions with varying approaches. From this broader view specific themes and issues will be examined, such as: The metaphor of the book as enabler or inhibiter of new avenues for research. User surveys and feedback as shaping forces of tools for digital editions. The role of users, editors, researchers, and funders in determining quality aspects. The digital edition as an expression of text in flux versus the iconic object. Text models for distributed documents. Is the generic or the specific a hallmark of quality of tools for digital editions? Crowd sourcing and open notebook science as determining aspects of digital editions. Visualization of instability of text as a scholarly quality of the digital edition. The relationship between formalization of editorial process and the instability of text.

Archaeology has a long history of innovative work with information and computing technology. While there are a small number examples in the late 1950s, the most influential comes courtesy of James Deetz’s seminal work on Arikara ceramics. Carried out in the early 1960s, Deetz’s project used the IBM704 mainframe at the MIT Computation Laboratory to discover "stylistic coherence" on over two thousand rim sherds from central South Dakota Medicine Crow site. Deetz’s work was extremely important as it suggested that computers were excellent tools for statistical, typological, chronological, or stylistic analysis of large and complex sets of data (a hallmark of archaeology).

Since these early days, digital archaeology has remained intently focused on the analysis, interoperability, and preservation of digital data. By the mid-1980s, however, the personal computer had reached a point where they became effective tools for archaeological visualization and imagery. Desktop applications such as GIS, which allowed for the visualization, analysis, and modeling of socio-spatial data, and CAD, which facilitated the production of detailed and geometrically accurate archaeological maps at various scales without time-consuming redrafting, became central in the digital archaeological ecosystem.

In recent years, along with many other disciplines in the humanities and social sciences, archaeology is entering a new age in which information, computing, and communication technology is having transformative impact on all aspects of the field. The archaeological domains and activities in which digital approaches, methods, and technologies are relevant have grown well beyond the traditional trinity of data, GIS, and CAD. All aspects of research (including field and lab methods), teaching, outreach, publication, and scholarly communication are being impacted in new and unpredictable ways by “digital.” Quite frankly, gone are the days in which digital archaeological methods were siloed off from the main body of scholarly practice. In many ways, one might argue that we have entered an age in which all archaeology is digital archaeology and all archaeologists are digital archaeologists.

In is within this context that this session will highlight a series of innovative projects and practices that represent the forefront of work in digital archaeology. Special attention has been made to highlight projects which represent a variety of domains within digital archaeology including digital data, public engagement, data & topic modeling, crowdsourcing, linked-open data, and digital fieldwork records management. All papers in the session also speak to the changed and changing nature of scholarly and professional practice within archaeology, addressing new approaches to collaboration, community engagement, citizen scholarship, cyberinfrastructure, preservation & access, capacity building, and sharing. A second, but no less important, goal of this session is to challenge the rather curious separation that exists between digital archaeology and the digital humanities by clearly placing the two domains parallel to one another and recognizing the fact that they both have much to learn from one another. The ultimate goal in this regard is to foster and support fruitful discussions and collaboration between digital archaeologists and digital humanists.

Topic modeling is very popular at the moment in the digital humanities. A recent tutorial on getting started with this tool explains them as tools for extracting topics or injecting semantic meaning into vocabularies: "Topic models represent a family of computer programs that extract topics from texts. A topic to the computer is a list of words that occur in statistically meaningful ways. A text can be an email, a blog post, a book chapter, a journal article, a diary entry – that is, any kind of unstructured text" (Graham, Weingart, and Milligan 2012). In that tutorial, 'unstructured' means that there is no encoding in the text by which a computer can model any of its semantic meaning.

Archaeological datasets are rich, largely unstructured bodies of text. While there are examples of archaeological datasets that are coded with semantic meaning through xml and Text Encoding Initiative practices, many of these are done after the fact of excavation or collection. In the field, things can be rather different, and this material can be considered to be 'largely unstructured' despite the use of databases, controlled vocabulary, and other means to maintain standardized descriptions of what is excavated, collected, and analyzed. This is because of the human factor. Not all archaeologists are equally skilled. Not all data gets recorded according to the standards. Where some see few differences in a particular clay fabric type, others might see many, and vice versa. Archaeological custom might call a particular vessel type a ‘casserole’, thus suggesting a particular use, only because in the 19th century when that vessel type was first encountered it reminded the archaeologist of what was in his kitchen – there is no necessary correlation between what we as archaeologists call things and what those things were originally used for. Further, once data is recorded (and the site has been destroyed through the excavation process), we tend to analyze these materials in isolation. That is, we write our analyses based on all of the examples of a particular type, rather than considering the interrelationships amongst the data found in the same context or locus. David Mimno in 2009 turned the tools of data analysis on the databases of household materials recovered and recorded room by room at Pompeii. He considered each room as a 'document' and the artefacts therein as the 'tokens' or 'words' within that document, for the purposes of topic modeling. The resulting 'topics' of this analysis are what he calls 'vocabularies' of object types which when taken together can suggest the mixture of functions particular rooms may have had in Pompeii. He writes, 'the purpose of this tool is not to show that topic modeling is the best tool for archaeological investigation, but that it is an appropriate tool that can provide a complement to human analysis....mathematically concrete in its biases'. The ‘casseroles’ of Pompeii turn out to have nothing to do with food preparation, in Mimno’s analysis.

To date, this is the only example of topic modeling applied to archaeological data. As such, it is novel in the digital humanities for applying the tools of data mining not to texts, but to things. In this paper, I explore the use of topic models on another rich archaeological dataset, the Portable Antiquities Scheme database in the UK. The Portable Antiquities Scheme is a project "to encourage the voluntary recording of archaeological objects found by members of the public in England and Wales". To date, there are over half a million unique records in the Scheme's database. I use topic modeling of this database to tease out archaeological patterns -the discourses of topic modeling, to use Ted Underwood's phrasing - over both time and space. In order to visualize these discourses, I map them both in geographic and relational space, using the network analysis program Gephi. The constellation of ideas (the resultant ‘topics’) that make up the various discourses in the data can be represented as nodes while the strengths of the associations suggested by the topic model can be represented as edges. This two-mode graph (words and ‘topics’ or ‘discourses’) can be queried for deeper structure. I look at the modularity of this graph to determine ‘communities’ of ideas or discourses. I then lay this network against real geographic space by time-slice to understand changes over time and space in the Portable Antiquities Scheme data. I agree with Mimno's suggestion that this is an appropriate tool for the digital archaeologist, but try to understand the limitations, caveats, and lessons for digital humanities more generally, from this application.

Over the past few decades, archaeologists have begun to realize the benefits of providing archival records in digital form. Whether information is collected electronically or digitized from pre-existing materials, digital archaeological data should be readily accessible from anywhere in the world. These developments have increased the productivity of scholars who no longer need to visit the actual archive and eased the strain on those projects that must accommodate visiting researchers in addition to their normal daily operations. On the other hand, the creation of digital archives has drastically impacted the ways in which we interact with documents and artifacts that form the basis of archaeological research. Financially constrained projects have lagged behind their better-funded peers in the process of digitization and dissemination of electronic records. As a result, instead of providing greater access to a wider range of archaeological data, the process of archival digitization runs the risk of further privileging the evidence of those surveys and excavations with greater financial resources. In addition, many of the existing digital archaeological archives concentrate upon artifacts to the point that the archaeologist’s field journal, arguably the most important evidence in establishing context, is rarely presented in its original form. Even where diverse forms of information are provided, a digital archive often encourages the study of objects and documents in isolation from one another and without the benefit of the institutional memory that often aids in their interpretation. While a traditional archive allows an individual to conduct their research through physical interaction with a number of different archival materials at once, often in the presence of those who discovered and prepared them, many digital archives simply rely on keyword searches to generate lists of electronic records that to the untrained eye appear to be of equal value as forms of evidence.

In this paper, we present the Archaeological Resource Cataloging System (ARCS), an open-source digital asset management application created for the Ohio State University Excavations at Isthmia in order to address these issues. Developed at Michigan State University through an NEH Digital Humanities Startup Grant, ARCS utilizes a web-based interface that allows authorized users to upload and “tag” digital resources consistently according to generally accepted metadata standards that can be further refined to reflect any project’s unique terminology. These resources can then be searched, sorted into collections and connected to one another through the creation of virtual links without affecting the integrity of the original data. In this way the essential interrelatedness of the various forms of archaeological data is preserved in a flexible electronic format. In order to foster a better sense of community among researchers, each resource is also provided with a discussion tool that allows users to ask questions or identify mistakes, thereby making use of others’ knowledge and experience to cultivate the development of the dataset. In addition, because ARCS depends on the collective effort of a community of users, the system generates a permanent record of all additions and modifications of resources so that errors can be easily corrected and dependable users more clearly identified.

Perhaps most importantly, because it is an open-source application that relies on multiple users to develop and manage the digital assets of an excavation or survey, ARCS offers an affordable option for archaeological projects that lack a dedicated digital archivist or IT specialist. Digital data can be added as it is made available and, once uploaded, new resources can be linked to body of evidence that continues to grow in size and detail. In the end, ARCS not only retains the many benefits of more traditional research involving physical documents at an actual archive but in many ways also speeds and simplifies the process of archaeological investigation.

In October 2012, Historic St. Mary's City (HSMC) launched a digital exhibit and social media campaign focused on the 19th-century component of their museum. HSMC, an archaeology and living history museum, has traditionally focused its 17th-century component, which was Maryland's first capital city. The digital exhibit, however, allowed the museum to begin interpreting additional centuries without disrupting the 17th-century landscape. Additionally, the digital exhibit, HSMC is able to develop an approach that focuses on public communication and engagement, allows for transparent research methods and interpretations, and provides flexibility when integrating the content into future programming and on-site exhibit. Through a combined approach of a content based digital exhibit, research blog, and social media, the digital exhibit, called "All of Us Would Walk Together", provides an example of digital archaeology that incorporates contemporary concepts of public archaeology through digital exhibitions and research methods. This paper will discuss how this has been put into action.

During the past few decades, community engagement has become a critical component of African American archaeology. Starting with the public excavations at the African American Burial Ground project in New York City, researchers have begun to incorporate local communities and descendants in the development and implementation of research projects and museum exhibits. Establishing a transparent, reciprocal, and pragmatic back-and-forth has become a valued and integral part of the research process for many archaeologists. Online tools have also been used as a means for public engagement, in particular at the Levi Jordan Plantation and Rosewood. More recently, archaeologists have adopted social media as a means for engaging communities and stakeholders, such as at the Michigan State University Campus Archaeology Program, Florida Public Archaeology Network, and at Mt. Vernon. Although each approach has highlighted different topics and methods for engagement, each has found the use of the web and digital social media to be a beneficial means of engaging the public.

At HSMC, the interpretation of the 19th century had not been the primary focus of the museum. This was particularly evident when the structures relating to the 19th century, including a manor home, its outbuildings, and a former duplex slave and tenant quarter, were physically moved to a different location in 1992 due to its conflict with the 17th-century interpretation. While the buildings continued to be used as a bed and breakfast, they were not used as an interpretive component of the museum, causing memory of the 19th century to be lost to the public. Recently, this story has begun to resurface, due to a number of factors at the museum. Included was the reacquisition of the manor home and outbuildings from the owners of the bed and breakfast, and funding opportunities to interpret the duplex quarter through a digital exhibit and a physical exhibit. In addition to interpreting the site, the goal of the exhibits was also to build a relationship with the African American community and to reinstate the 19th-century story into the public consciousness.

The digital exhibit consists of two components: a traditional exhibit space and social media. The exhibit space presents a number of webpages devoted to the interpretation of the historical and archaeological data that has already been analyzed. These pages trace the transition from slavery to freedom for the African Americans who lived on the site, and uses historical and archaeological evidence to develop the narrative. Interspersed are links to blog posts that discuss how the evidence was gathered or used by researchers to draw the conclusions. Additionally, each exhibit page has a comment field, where the public can ask questions, offer their own interpretation, or provide additional commentary. This allows for two-way communication between the public and researchers, while also allowing the public to engage in the interpretive efforts. Lastly, these pages will be linked to the physical exhibit through the use of QR codes or augmented reality to provide more flexibility to the interpretive efforts at the site of the duplex. In doing so, the exhibit becomes flexible, transparent, and engaged physical space, fitting within the parameters of an engaged cultural heritage project.

The use of social media, through a blog, Twitter, and Facebook, adds extra depth to these efforts. The blog provides the most flexibility and transparency by allowing the exhibit space to be amended, added to, or modified in a transparent way. This provides a great deal of flexibility to the site that only a digital exhibit can provide. For example, if the research results in a major change to the exhibit, a blog post can be written to discuss the change and why it happened. In the exhibit, a link to this post can be added to demonstrate that the research process is a fluid, ongoing process. The blog and Twitter are also used to highlight research at comparable sites or examining comparable themes. This ties the archaeological work conducted at this site to larger themes in the discipline, and begins to build relationships with other institutions, while providing additional resources and access to new scholarship to the public. The blog and Twitter are both instrumental in the preservation and exhibit building process, as they make it more transparent: the public can watch and participate in the decisions about what will be included in the exhibit, in addition to understanding what types of constraints are placed on the construction of an exhibit. Twitter also allows lab and fieldwork to be shared in realtime with the public. Lastly, Twitter provides access to a larger, global network, particularly an African American network, a demographic that uses Twitter more than others. Facebook, on the other hand, is being used to connect with HSMC's current online fans through its official Facebook account.

The project itself has a high set of goals, and is approaching it with a multifaceted approach. While the digital component is a crucial step, it is only part of a larger program of public engagement. For example, HSMC has been actively soliciting feedback from local community members and seeking to engage them through the formation of an advisory board. This reiterates an important tenant of engaging in digital public archaeology: that one cannot rely solely on one approach. Nonetheless, the use of the digital space does provide us with additional exhibit and interpretive space, and gives us a great deal of flexibility when dealing with the public, research, and presentation. Most importantly, the use of the digital arena allows this all to be transparent, reciprocal, and dynamic.

The Digital Index of North American Archaeology (DINAA) is a project to create interoperability models for archaeological site databases in the eastern United States, funded by the National Science Foundation (#1216810 & #1217240). The core research team consists of researchers from the Department of Anthropology and Archaeological Research Laboratory at the University of Tennessee, the Alexandria Archive Institute, and the Anthropology and Informatics programs at Indiana University. Open Context (http://opencontext.org) will be used as the primary platform for data dissemination for this project. Our aims are to work with the databases held by State Historic Preservation offices and allied federal and tribal agencies in Eastern North America, with the goal of linking data across state lines for research and management purposes. Redacted of sensitive items, such as site location, data linkages will promote extension and reuse by government personnel in state and federal agencies, and domestic and international researchers. The project will mint stable Web-URIs for each site record, and in doing so, we will help lay the foundations for future Linked Open Data applications in North American archaeology, architectural history, and historical studies.

This project repurposes government curated datasets to support innovative humanistic and social science research. Governmental archaeological site files in North America are important loci for documentary information on known archaeological sites. Their most basic function is to contain data about site types and information quality pursuant to heritage preservation legislation at the federal level, but potentially state and local levels as well. However, as a matter of practice these files, often as relational databases, contain many other data fields that describe important archaeological findings, and other data that serve environmental and bureaucratic functions for management and protection of heritage resources. The ways in which data about archaeological sites are recorded and communicated have an important origin in theoretical models about past behavior, and also have important implications on the professional comprehension of the data at large and the use of the data to rank planning and preservation priorities.

Efforts to collect and compile archaeological data have a long history, and information about archaeological sites and collections is maintained by every state and territory. Only rarely, however, have these data been compiled and examined at large geographic scales, especially those crosscutting state lines, and never to the extent and for the research and management purposes proposed in this project. Data from some 15 to 20 states (>half a million sites) east of the Mississippi will be integrated with a common ontology, based on existing standards, and adapted in collaboration with researchers and government personnel in state and federal agencies. The ontology will classify site files according to cultural affiliation and chronology as well as agency assessments of historical significance.

Linkage of site file and other datasets will facilitate studies of past human adaptation spanning large areas, and lead to greater collaboration between archaeologists and scientists in other disciplines. As examples, the linkage of archaeological data at broad new scales will permit, for the first time, the exploration of exciting new research topics, such as how the human populations in North America responded to climate change, population growth, and/or anthropogenic environmental issues over the past 13,000 years.

The availability of output online in the form of maps and data tables (at significantly reduced spatial resolution, to protect sensitive locations) will enhance public awareness, education, and appreciation for scientific research in general and archaeology in particular. The demonstration that primary archaeological data can be integrated and used to address fundamental questions at such scales will stimulate similar efforts worldwide. Finally, by creating translating routines rather than dictating procedures, this project will foster archaeological cooperation through cyberinfrastructure with a high ratio of benefits to costs.

The project helps achieve broader archaeological concerns regarding professional data management training, research ethics and outreach education. It will foster novel networking and data integration among multiple partners, as well as research and educational activities across multiple disciplines and geopolitical boundaries. Publicly accessible data products, at coarse scales to preserve site location security, will also be available for download and reuse as shapefiles, CSV data tables, and RDF and N3 triples for other Web and desktop applications, including desktop GIS investigation. The project will provide specific instructions for the open source GIS applications QGIS gvSIG, and uDIG, and the widely used proprietary application ArcGIS, in order to foster education in geographic information science within archaeology and related disciplines. The project will fund graduate and undergraduate students, and will assist their training in critical information management skills for the 21st century. The project addresses head-on a major challenge facing research communities worldwide: how to link disconnected and incompatible data systems in such a way that the combined data are useful for important scientific research.

The integration of site file data at continental scales in a new and unique informational infrastructure will allow, for the first time, the exploration of the North American archaeological record across multiple temporal periods and geographic regions. The geographic scale and extent of data integration proposed is currently unprecedented in American archaeology yet, we believe we have demonstrated that it is readily achievable. With proper attention these data have the potential for continued growth as developed by the professional archaeological community, and as the resulting datasets become more inclusive they may transform the practice of our profession.

The Portable Antiquities Scheme (PAS) was inaugurated in 1997, following the revision of the ancient law of Treasure Trove and the subsequent implementation of the Treasure Act in 1996. This project has been through several funding phases all using public money and encourages the voluntary recording of archaeological objects discovered by members of the public in England and Wales (Scotland is subject to different legislation) and items that meet the stipulations of the Treasure Act.

One of the key pillars of the PAS has been its digital presence, which has now been online in some form for over 13 years and this paper will discuss the impact that the digital arm of the project has had on a national and international audience. The PAS has been hailed by many as a model of public archaeological engagement and decried by others for allowing the mining of the archaeological record for personal gain; however this paper will show some of the PAS’ many successes. For example the PAS has had contributions from over 20,000 individuals worldwide, a new facility has been created for contributors to record their own discoveries through taxonomy driven interfaces and over 350 projects utilise these data for informing their research - for example Oxford University’s EngLaID project. The project has also absorbed and enhanced internationally renowned resources such as Oxford University’s Celtic Coin Index and Cardiff University’s Iron Age and Roman coins of Wales database to provide the largest national, single search node for the study of Roman and Celtic coinage.

The project website records a huge array of metadata about objects, that we often have but one chance to record; images, textual description, measurements, spatial data and user generated comments and audit logs. Over 820,000 objects have been recorded on the PAS database and these are made available for all to view, comment and reuse within their own research or their own websites under a Creative Commons ‘by attribution share-alike’ licence. This liberal approach to licensing content has not been seen widely in the UK and European archaeological sector and the launch of the PAS’ Staffordshire Hoard microsite in September 2009 showed how well received this approach would be. Over 250,000 unique visitors in one day used the innovative microsite to learn more about the amazing Anglo-Saxon hoard and many more viewed the images that had been disseminated via Flickr. The PAS has also utilised and archived social media platforms with varying degrees of success and is a major case study within Lorna Richardson’s forthcoming PhD and complements that of its host organisation (the British Museum.)

This paper will show how the PAS website impacts on the public with specific reference to stories of international interest – such as the Staffordshire Hoard, the Frome hoard of 52,503 Roman coins, the Crosby Garrett helmet and the Staffordshire Moorlands Patera. It will also discuss how these successes have been reached on a minimal digital budget (less than £5000 per annum) via the use of open source technology and through the buy in of its audience. The website has been internationally recognised, winning the prestigious Museums and the Web ‘Best of Web’ award for ‘research or online collection’ (recent winners include the V&A and the Metropolitan Museum of Art.)

The author will demonstrate how a variety of digital techniques that have been employed to complement and enhance data collected via our network of archaeologists, volunteers and citizen scientists; for example geo enrichment through the use of Yahoo!, Geonames, Pleiades and Google Maps, text extraction through OpenCalais and Autonomy, the implementation of a Solr based faceted multi-core search engine and also how a wide variety of application programming interfaces (api) have been employed throughout the site. It will also show how the author has been influenced by ground breaking projects such as Open Context, Pleiades and Pelagios and a variety of Museum sector projects.

The paper will also touch on the recent steps towards providing the PAS data through linked data (for example the release of over 50,000 annotations for Pelagios), towards integrating linked data into the site (for example dbpedia enriched resources) and the CIDOC-CRM mapping process. It will also discuss 3D , computed tomography and PTM/RTI imaging projects conducted by Brighton and Southampton universities that have used PAS sourced objects and hoards of coins to provide research material. If time allows, the paper will also demonstrate the success achieved through the use of Flickr as a disseminator for a wide variety of archaeological images; how the author has leveraged news sources such as the Guardian and UK parliamentary records for background debate on the portable antiquities debate (ethical and fiscal) and how it has impacted on the public psyche within the UK and further afield.

The spaces we inhabit and interact with on a daily basis are made up of layers of cultural activity that are, quite literally, built up over time. While museum exhibits, historical and archaeological narratives, and public archaeology programs can communicate this cultural heritage, they do not generally allow for rich, place-based, and individually driven exploration by the public. In addition, museum exhibits rarely explore the binary nature of material culture and the preserved record of human activity: the presented information about material culture and the process by which scholarly research has reached those conclusions. In short, the scholarly narrative of material culture, cultural heritage, and archaeology is often hidden from public consumption.

In recent years, mobile devices as well as the development and maturation of augmented reality (broadly construed) have offered both platforms and models for mobile cultural heritage applications to address the former issue. Mobile applications such as The Museum of London’s Streetmuseum Londinium, Florida Public Archaeology Network’s Destination: Civil War, and the forthcoming CHESS Acropolis Museum mobile application facilitate public interaction with cultural heritage and archaeology in a place-based context. However, the latter issue, the scholarly narrative of the process by which cultural heritage and archaeological information was uncovered and information was generated, is often left unaddressed and unexplored in mobile cultural heritage applications.

It is within this context that this paper will introduce and explore msu.seum. Developed during the 2011 Michigan State University Cultural Heritage Informatics Fieldschool directed by Ethan Watrall in collaboration with the Michigan State University Campus Archaeology Program and Campus Archaeology Fieldschool, msu.seum is a mobile application that allows users to interact with the rich cultural archaeological heritage of the historic Michigan State University campus, and explore the processes by which the Campus Archaeology Program helped reveal it. Building on the idea of “campus as museum,” msu.seum connects cultural heritage directly to place, highlighting both what is known about the MSU Campus and the scholarly narrative of the associated archaeological and historical research. msu.seum functions less like a “check-in” app and more like a rich, exploration-based museum tour guide that exposes the cultural heritage and archaeology of Michigan State University to staff, faculty, students, alumni, and the general public.

Currently available for iOS, msu.seum’s content is organized into a series of thematic “exhibits” that reflect the development of Michigan State University: Beginnings: 1855-1870, Foundation: 1870-1900, Expansion: 1900-1925, and Legacy: 1925-1955. A fifth exhibit, Discovery: Archaeology, explores more recent research and excavations carried out by the MSU Campus Archaeology Program. The idea behind this exhibit-based model is that the user’s experience mirrors the physical layout of a museum. Instead of being contained within a physical structure, however, exhibits are distributed around the MSU campus. Each exhibit contains a collection of locations that users are free to visit and experience at their leisure (either within the context of a visit to the Michigan State University campus or during their regular, daily campus activities)

Each exhibit location contains information and rich media (video, audio, and imagery) about that location and well as the narrative about the work (excavation, survey, etc) carried out at the location by the MSU Campus Archaeology Program.

The paper will explore the features of the application itself, as well as the process by which it was designed. Special attention will be paid to discussing the unique and highly collaborative environment in which msu.seum was developed.

Beyond its value as a tool to allow the public to interact with and explore the cultural heritage and archaeology of the Michigan State University (and the associated scholarly narrative as to how that knowledge was developed), we feel very strongly that msu.seum can act not only as a model for the design and development of other mobile and place-based campus cultural heritage applications, but for mobile and place-based cultural heritage applications in general.

In this paper, we introduce the concept of dyadic pulsations as a measure of sustainability in online discussion groups. Dyadic pulsations correspond to new communication exchanges occurring between two participants in a discussion group. A group that continuously integrates new participants in the on-going conversation is characterized by a steady dyadic pulsation rhythm. On the contrary, groups that either pursue close conversation or unilateral communication have no or very little dyadic pulsations. We show on two examples taken from Usenet discussion groups, that dyadic pulsations permit to anticipate future bursts in response delay time which are signs of group discussion collapses. We discuss ways of making this measure resilient to spam and other common algorithmic production that pollutes real discussions.

Can a discussion group be characterized by looking solely at the interaction patterns of its participants? Symmetrically, can the patterns of interaction of a given participant identify its role in a discussion group? Can we predict from these patterns the evolution of the interaction inside a group, spotting for instance the early signs of a group decomposition process? Our research aims to establish a new mathematical approach to distinguish between different types of discussion group participants as well as between different types of discussion groups, both with their typical ways of interacting (interaction "signatures") and life cycles.

The analysis of interaction dynamics has recently received an increased focus of attention in the network science community. Mathematical methods (Newman, Barabási and Watts 2006) have been used to identify signatures characterizing the mode of exchanges of famous scholars, comparing for instance patterns in the correspondence of Charles Darwin and Albert Einstein (Oliveira and Barabási 2005). Fluctuation patterns and delays in letter responses indicate prioritization strategies that can be modeled and simulated. These methods permit also to draw comparisons with modern forms of electronic exchanges where similar patterns, corresponding to universal scaling laws (Barabási 2005; Bunde, Eichner, Havlin and Kantelhardt 2004), can be found. Interestingly, all these analyses can be conducted without considering the semantic or pragmatic nature of the exchanges.

Patterns in correspondences networks are of great interest for research in Digital Humanities. For instance, the Stanford’s Republic of Letters project (http://republicofletters.stanford.edu) use “big data” and “distant reading” approaches to offer new visualization tools and test various hypotheses about the Enlightenment. However, mathematical analysis of such networks is not yet common in the Digital Humanities community.

Drawing on research on social networks (Wasserman and Faust 2009 [1994]) and computer mediated communication (Smith and Kollock 1999; Turner, Smith, Fisher and Welser 2005; Welser, Gleave, Fisher and Smith 2007), we analyze the activity and correspondence pattern of participants in Usenet newsgroups about religion and spirituality. Our paper reports an on-going analysis of large data set that consists of more than 1.5 million unique Usenet messages. Usenet is one of the Internet's oldest discussion systems still in widespread use. Unlike more recent platforms like Facebook and Twitter, Usenet hasn't stored its messages in a single central location and offers due to its open nature easier access to its data. Furthermore, Usenet's threaded conversations are organized by topics, with the advantage of allowing comparisons between topics.

In this paper, we introduce the concept of dyadic pulsation (DP) as a complementary measure to reply time for measuring the vitality of a given group. In our representation, a pulsation corresponds to the creation of new communication dyads, i.e. the first direct communication between two users A and B. When B first replies to A, a pulsation of type A (for asymmetric) is emitted. When A replies again to a message of B, a second pulsation of type M (for mutual) is produced. A group that continuously integrates new members in the on-going conversation is characterized by a steady dyadic pulsation rhythm, mixing type A and type M pulsations. On the contrary, groups that either pursue close conversation or unilateral communication (e.g. news feeds, announces without discussion) have no or very little dyadic pulsations.

Our working hypothesis is that evolutions in the pulsation rhythms are earlier predictors of the evolution of group dynamics. Figure A shows an example of a group maintaining a good average response time for a long period followed by an apparently unanticipated explosion of response time. At some point group members simply stop to answer timely to the messages of the discussion group. Interestingly, although no anticipated sign of this evolution could be spotted in the delay time graph, the pulsation graphs shows a progressive reduction of the frequency on the creation of new communication dyads. The grey lines in the lower bar of show pulsations of type A while the black lines show pulsations of type M. Figure B shows the change in delay time after the appearance of a succession of type A and type M pulsations, indicators for the formation of new links between discussion participants.

For dyadic pulsations to be a reliable predictor, it is mandatory that they are resistant to the various forms of spam common to digital communication. This is one reason for the distinction of the two types of dyadic pulsations. Indeed, dyadic pulsations of type A can result of spam bots posting messages to a group. However, it is very unlikely that any real user answers to those messages. Thus, the presence of type M pulsations guarantees that new correspondence partners entered the group studied.

We are conducting a larger study to test the relevance of this measure in the particular case of different online discussion groups related to topics about religion and spiritualty. Our hope is to validate the hypothesis that dyadic pulsations on a group level permit to distinguish between different modes in a discussion group's life cycle.

More generally, we believe that this measure can be relevant to characterize the rises and declines of activity in correspondence networks, including literary correspondence networks.

The Book of Hours is perhaps the most ubiquitous type of surviving medieval manuscripts. Each unique book contains a wealth of information about liturgical practices, social and private use of books, manuscript production and decoration, and personal and local choices for the creation of private devotional collections. Each Book of Hours contains a Kalendar that lists feast days for the entire year. These document the standard days for liturgical celebration, but are filled with local and temporal variants that paint a picture of the ways in which practices change according to regional interests over the centuries. Any attempt to curate and control a project devoted to cataloging Kalendar variants in a single team or location suffers from the sheer volume of potential inputs to the project. Further, the standard form of the Kalendar invites widespread contributions from interested scholars and the broader public. This paper describes a comparative Kalendar built from distributed resources served from multiple participating tools and repositories.

This distributed environment relies on repository/tool interactions to share image data in multiple tool environments, and harvest the tool output to build a discovery and viewing environment that can grow as more data is added from any participating tool or repository. These interactions rely on two principle interchange protocols: the SharedCanvas data model, used to aggregate distributed resources; and the International Image Interoperability Framework image API for standardized image resource access.

The SharedCanvas data model (http://www.shared-canvas.org) specifies a mechanism for representing a real-world manuscript object as a collection of two or more blank “canvases” each with a URI and an aspect ratio. Information, including digital image surrogates, is then associated with this canvas using the OpenAnnotation specification. This approach decouples information “about” the real-world object from the digital surrogate that could be used to illustrate that object on the web. Because it is quite possible that a single real-world resource will have multiple digital surrogates all purporting to represent the same thing, this approach makes it possible to build a digital facsimile utilizing a choice of digital surrogates without affecting the other information associated with the object (like transcriptions, scholarly commentary, or other metadata).

The International Image Interoperability Framework’s image API (http://lib.stanford.edu/iiif) specifies a RESTful web service to deliver a requested image with a number of specified parameters. These include: image format, size, rotation, cropping, and color. The API allows standardized image retrieval from disparate repositories, leading to the ability of tool developers to pull content in an efficient way from any participating host repository. More importantly, this API allows contributors to the Kalendar project to work with from a number of different interfaces and use image data served from a number of different repositories, allowing scholars with a regional or temporal focus to compare books held in different modern repositories.

The participating repositories in this effort are the Stanford University Libraries (serving content from the Walters Art Museum manuscript collection, the Parker Library, Corpus Christi College, Cambridge manuscript collection, and Stanford’s own collection), along with other participants of the Digital Manuscript Technical Council (http://dms.stanford.edu/). This selection of content offers a variety of geographically and chronologically diverse Books of Hours, leading to an initial dataset that highlights both the consistency of some Kalendar elements over three centuries (13th, 14th, and 15th centuries), and the variants that provide insight into localized practices.

The tools used for this project are T-PEN (http://t-pen.org/TPEN) and DM (http://ada.drew.edu/dmproject/), specialized tools for transcription and annotation of medieval manuscript content. A user of either tool has full access to any of the manuscripts served from the participating repositories, and can add transcriptions of Kalendars while also adding structured data to help organize the material for further analysis and display. Both tools produce outputs that link user-generated information about the Kalendars to a region-constraint on the canvas that represents the page of the real-world manuscript. The tools provide a web service for exposing user-generated information that is then aggregated into a discovery and display interface.

This approach reduces the need for any one institution or project to host and serve every image that might be used in the Comparative Kalendar, to provide the user interface tools for transcription, commentary, and analysis, or to control and curate the input and output. The mechanism for user-interaction allows a user in any of the participating tools to choose content from any of the participating repositories. This allows a user to choose the working environment that most suits the tasks they wish to pursue, and do their analytical work within that tool. Since the tools allow standardized extraction of data for additional display, data management, navigation, or additional analysis and commentary, a user can move from tool environment to tool environment to achieve separate tasks (transcription, addition of structured metadata, commentary and notes, etc.).

The end result of this project is a user interface that draws image resources from the participating repositories and user-generated data produced by the participating tools, hosted by each respective participant. New data from any participating node will be automatically added to the interface without the need for human-mediated interaction, leading to a dynamically growing resource. This resource will allow a user to browse a Kalendar and see image, transcription, and commentary data about each entry, or compare across Kalendars to observe clusterings of continuity and variance over the liturgical year depending on location of book use and production, or chronological period. The distributed approach to a standardized text, where variants are of great interest across a large number of disciplines, and which benefits from a broad array of participants providing small bits of detailed data to build up a useful dataset, could serve as a model for further work on other parts of the Book of Hours, or other medieval texts that were copied frequently an reflect changing social, political, or liturgical practices across Western Europe over a period of several hundred years.

While text mining, and similar quantitative analyses of lexical and semantic frequencies, have proven to be highly informative about aspects of literary history, the origin of these methods within the “hard" sciences creates hurdles for their application within the humanities (Pasanek and Sculley 2008, Bei 2008). Fundamental to these statistical strategies is the assumption of communicative equivalence among words as types: to most kinds of frequency analysis, each instance of a particular word retains the same meaning, force and valence of all other instances of that word. A chapter of a book in which the word “money” is relatively frequent suggests, to these methods, that the author is concerned with economics and it can therefore be related to other texts densely populated by economic terms. A critical problem in literary analysis, however, occurs within the field of tropes: in a seventeenth-century poem, a flea may be a small insect, or it may be a complex metaphor for a particular kind of romantic relationship. While newer, mixture-based, text mining methods, such as topic modeling, can use semantic clusters to reveal the difference between homonyms (the associated “topic” words of “bow”— applause, cheer, crowd — would be different than those of “bow” - arrow, fletch, string) they remain unable to detect the nuance of metaphors that are critical to our understanding of literary effect (Ross 2003, Blei 2012). Indeed, if most of the work of a poem comes not through the communicative value of the literal meaning of the words themselves, but instead, through the complex tropology of the formal effects, then we, as digital scholars of literary texts, require a new way to incorporate an understanding of the work of tropes within the semantic field. In particular, at the level of genre identification, where much text mining or frequency analysis finds its home, the genres of allegory and satire, as extended tropes, remain unavailable to these methods. This problem is particularly acute when bringing to bear new quantitative methods on studies of poetry: with its highly figurative language and complex communicative intent, poetry remains on the fringe of quantitative analyses of literature. Most such studies focus exclusively on prose writing, and, in particular, the novel (Stanford Literary Lab 2011, Clement 2008). Given the importance of poetry to literary history, we believe that this lacuna represents a critical problem for the use of digital humanities in the study of literature.

Our project is an attempt to address this challenge through a new approach that combines a digital analysis of both the formal and lexical features of a constrained sample of poetry to test if it is possible to detect and identify the presence of tropological structures within poetic writing. In particular, this paper explores whether trope-based genres, specifically satire and allegory, are identifiable from linguistic and formal artifacts that can be recovered through digital analysis. In literary criticism, these genres are highly dependent upon context: a traditional analytic approach insists that a detailed knowledge of the poem’s intended reception is necessary to recognize the satiric intent of a particular poem, or the allegorical framework that informs it (Fletcher and Bloom 2012). We contend, however, that there are generic markers within the formal and semantic fields of the poem that can be used to identify the presence of these tropes. To perform this analysis, we will draw upon a corpus of eighteenth-century poems selected for their comprehensive use of either satire or allegory, as well as a corpus with the same distribution across the period composed of poems that do not participate in either genre. By carefully choosing period and genre specific texts, our project, the first effort in this direction, seeks to limit the scope of its inquiry. As a way of approaching the larger question of poetic tropes, we ask if within the specific genres of eighteenth-century allegoric and satiric poems, we can train an algorithm to recognize the presence of either genre. A key to this project is our belief that the greatest challenge for the analysis of poetry using quantitative methods, that is, its highly figurative and stylized language, can be turned into an advantage in our approach. Both genres, we argue, operate on a semantic, as well as a formal level, and, therefore, the regularities of these formal structures can aid us in revealing key generic identifiers. Our method operates by identifying both the formal features and lexical regularities within a poem and comparing their relationship: we argue that the match, or more importantly, the mismatch between the formal structure and the semantic fields within poems reveals the presence of these tropological genres. In our project, this relationship becomes one of mediation: what patterns emerge when a topic is mediated through a form for which it is traditionally unsuited?

This is a particularly relevant question for the poetry of the eighteenth century, which formalized many of the generic tropes within specific combinations of style and meaning. Satire, for example, borrowed the rhetorical tropes of high-minded discourse (heroic couplets, iambic meter and poetic epithets) which are undercut by the inappropriateness of the poetic object (a lady’s dressing room, the theft of a lock of hair) to the poetic medium. Similarly, allegory uses specific formal constraints to layer meaning within its lexical patterns. It is this formal shift towards different models of mediation that our project measures by combining text mining strategies, such as frequency analysis and topic modeling, with an automated recognition of poetic structure (like meter or rhyme scheme). While the results that we will present are specific to the genres of eighteenth-century poetic allegory and satire, we nevertheless believe that this represents a crucial first step towards automating the ways in which we recognize figurative language and developing flexible methods that can be extended to other contexts of use.

This project builds upon ongoing work within the Stanford Literary Lab on the subject of poetic discourse. Our project depends upon a corpus of poetry tagged as either satire or allegory (or neither, as a control), which can be used to both train our algorithm and test its efficacy. For this project, we have assembled a corpus of over 1500 poems, written between 1700 and 1800 that have been identified as belonging to one of our two particular genres. As this project seeks to test the ability of the methodology we have devised to identify the presence of satire or allegory within an entire poem, we have limited our corpus to poems that participate fully within the genres of satire or allegory, rather than those that have satiric or allegoric passages only. As the identifying markers of these genres are already highly nuanced, we feel that restricting our study to these two examples better enables us to identify a working model of poetic tropes that can later be improved for finer-grained and more detailed applications. Similarly, this will be the first practical application of an automated method for identifying and cataloguing meter and rhyme scheme within poetic writing that we have developed for the lab. As we argue that the artifacts of poetic tropes exist within the interaction between semantic fields and their formal mediation, we are able to bring our new ability to identify formal aspects of the poem to bear on the analysis of the patterns that these interactions take.

The stakes of this project involve both a push to incorporate poetic writing within our methodology of quantitative analysis, and a challenge to our understanding of the ways that figurative language operates within eighteenth-century poetry. Scholars, such as Fletcher and Bloom on allegory (2012) and Connery and Combe on satire (1995), argue that we are only able to recognize these genres through a detailed understanding of the historical and cultural contexts in which the poem was written. Our methodology, however, argues for an expanded understanding of context: instead of a detailed knowledge of eighteenth-century political or economic history, context, in our study, is a function of the relationship between semantic fields and their formal mediation within a poem. This, we argue, is equally as important to the context of the poem as the geo-political and historical world from which it emerged. While the necessary specificity of our study limits our project's conclusions to the genres of eighteenth-century allegory and satire, we believe that our results have implications for a wider understanding of how figurative language can be incorporated within a quantitative analysis of literature. By carefully limiting our genres and period, we are offering this study as a proof-of-concept test whose results will aid in future quantitative and digital work on poetry. We anticipate that this pilot study can be extrapolated into a wider understanding of figurative language and the tropes in which it operates and that our results from this project can be incorporated into future work that seeks to undertake a wider and more comprehensive analysis of poetic language across multiple centuries.

A premise of the Digital Humanities 2013 conference theme, “Freedom to Explore,” is that users of the languages of the world should be able to express themselves and search for documents in their own language (or one of their heritage), using the script employed to write that language. For many widely used languages, this is generally possible today, because its script (and characters) can be represented by the international character encoding standard Unicode (Unicode Consortium, 2012) and its ISO mate, ISO/IEC 10646 (ISO/IEC, 2012), and because these scripts are supported on today’s computers and electronic devices.

But those scripts used for lesser-used languages — both modern and historic — that are not in Unicode are subject to problems in representation and searching, since they are not part of the standard. One consequence is that this will leave gaps in the world’s cultural, linguistic, and historical legacy. Even after the scripts are accepted and published in the Unicode Standard (and ISO standard), the languages and their scripts must then be supported in fonts and rendering engines, and locale data information is needed to implement a modern script on cell phones or computer devices. Because of these factors, the goal of making electronic text communication truly global and multilingual remains a high bar, but one that the digital humanities should continue to strive to achieve.

This short paper will report on recent developments in the effort to make the text of lesser-used languages accessible electronically, focusing primarily on issues involved in getting characters into Unicode but also touching on the social aspects of character encoding.

In the past few years, Unicode (and ISO/IEC 10646) has become more widely accepted and better understood, even by laypeople. Indeed, forty-two modern and historic scripts have been added in the various releases over the past six years1, most through the UC Berkeley Script Encoding Initiative (Script Encoding Initiative, 2012). In the past, it was often challenging to explain to language communities the importance of getting a script into Unicode. However, today user communities are themselves approaching the Unicode Consortium (or others involved in character encoding) in order to get their script included in the standard. The Script Encoding Initiative at UC Berkeley has, for example, been approached to help in encoding the ADLaM alphabet in Guinea and the Mandombe script in the Democratic Republic of Congo, two recently devised scripts intended for modern use. Historic script users — who may be heritage language users, “hobbyists” or come from academia or a liturgical setting — have also been active in getting scripts encoded. For example, experts of the Hatran script of the Middle East recently sought help in getting their script encoded.

While it is useful to have the user communities directly involved in the proposal process, proposals written by new authors often require extensive assistance. Typically, the approval process takes much longer for proposals written by new authors in order for all the required technical information to be included in the proposal. A more effective approach, adopted by the Script Encoding Initiative (SEI), has been to encourage user communities to work with a veteran Unicode proposal author, who is familiar with the standards committees’ requirements. However, relying on an outsider to help on a proposal may be looked on with suspicion. One way to allay this fear, an approach also advocated by the SEI project, is to try to encourage collaborative work between an expert in the encoding process and the users, making it clear the ownership of the script clearly belongs to the user community, but the encoding expert (and the standards committees) will help make sure the script will be implementable on today’s computers and fonts. Currently, work on the Nepaalalipi/Newar proposal is following this approach and a report on its success (or failure) will be relayed.

One issue in the encoding process that has repeatedly been an issue for user communities of modern scripts is what name to assign the script. A script name in Unicode is meant to be as an identifier (such as in programming), and typically one that is commonly found in English.2 The English requirement has been problematic, since the English name is often not that of the user communities. Also, the name for a script can vary across languages (and countries), so there can be a tug-of-war between groups as to which name to use. One way to deal with the name has been to encourage user communities to translate character names in the code charts into their own language. Likewise, once the script is approved, fonts can use their preferred name for the glyphs. In dealing with competing names in different languages, one solution has been to get the groups to try to agree on a name that is acceptable to all. This approach was successful for Tai Tham (“Lanna” in Thailand), but has not worked out well for Old Hungarian/Szekely-Hungarian Rovas. (Typically, names for historic scripts are not controversial or contentious, though disagreements have arisen within in the standards committees on a particular name or character.)

The current process of script encoding involves approval by the Unicode Technical Committee (UTC) and the subcommittee on coded character sets in the International Organization for Standardization (ISO). Because ISO represents character encoding in the governmental arena, the subcommittee chair on Coded Character Sets in the past three years has tried to encourage countries of lesser-used scripts to participate in ISO, particularly those in West Africa. However, many countries have not been able to participate due to lack of funding. As a way to involve the user communities, the SEI project has worked with user communities, without direct support or involvement of the government.

Once a script is approved and published in a version of the Unicode Standard (and ISO/IEC 10646), a text in that script is not necessarily ready for interchange: fonts with the correct codepoints must be created as well as an input method, the rendering engine must be updated to support the script (particularly if the script is complex and has special processing for display), and locale data is needed for implementations, including electronic devices. In these areas, one of the more promising areas of assistance for modern lesser-used languages and scripts is the ScriptSource project sponsored by SIL International (SIL International, n.d.), The talk will include a discussion of the Cherokee script as it progressed from initial Unicode proposal to iPad and iPhone implementations, as a model for other lesser-used scripts. The paper will close with thoughts from the speaker on how to improve the current process and possible next steps, such as: engage users and relevant national body standards bodies early in the process of encoding a script (so as to prevent a situation that arose with Khmer3) support funding for projects in which a dedicated person can work with the user communities and instruct them on the encoding process encourage users to vote (via their national standards body) on relevant ISO script encoding ballots.

Various projects around the world focus on analyzing and visualizing the correspondences and social networks of scholars(academics, writers, etc), e.g., the ‘Mapping the Republic of Letters’ project centered at Stanford University (Edelstein et al., 2008) and the project ‘Vernetzte Korrespondenzen’(engl.: Networked Correspondences)(Burch et al., 2012) run by Trier Center for Digital Humanities, German Literary Archive Marbach, and Martin-Luther-University Halle-Wittenberg. In addition to the design and implementation of adequate tools for analyzing and visualizing the big data, an indexing scheme for the content of the letters of the corpus under consideration has to be developed and the letters have to be captured, annotated, and indexed with respect to the indexing scheme.

The capturing, annotating, and indexing process should be supported by an interactive browser based platform which allows the professional philologists (as well as ‘citizen scholars’ from non-academic backgrounds) to quickly enter metadata (e.g., date, name/identification of senders and addressees) in an accurate und simple manner. To meet the simplicity requirement, the platform should accept unformatted entries as far as possible; to meet the accuracy requirement, the entries have to be clear without ambiguity. The two requirements appear to be competitive and conflicting. To sort out the problem, we propose to support unformatted data entry by auto-completion based on external catalogues, e.g., the ‘Deutsche Nationalbibliografie’(DNB, 2013b) provided by the German National Library. The auto-completion tool should provide additional biographical information on the proposed scholars as soon as the philologist enters first data, to help the philologist uniquely identify scholars.

In preparation for the above mentioned ‘Vernetzte Korrespondenzen’ project(Burch et al., 2012) which investigates the correspondences and social networks of German scholars which went into exile during Nazi Germany, we have implemented such a browser based platform.

As soon as the philologist enters a part of the name of the sender or the addressee, e.g., ‘Joseph’, the tool makes proposals of scholars of the 18th and 19th Century that match the requirement, e.g., ‘Eichendorff, Joseph von’, ‘Hübner, Joseph Alexander von’ and so on, together with their dates of birth and death, identification number of the name authority file (‘Gemeinsame Normdatei’) published by the German National Library(DNB, 2013a), and short vitae. Especially these secondary data are of benefit if several scholars with similar(or same) name exist. Once the philologist made her choice, the system pins the scholar’s biographical data to the work space. If the person is unknown to the system which may be the case if the letter was sent, e.g., to the pharmacist next door (whose name is not specified in the letter), the philologist enters this new person into the system. The autocompletion mechanism of the platform uses these new data during future requests.

The entry of the letter’s date poses a further challenge as the determination of the date is harder than you think. For instance, a letter can be written over several days, the date may be hardly readable, it can be given by a holiday, or it can be specified imprecisely. That’s why our platform allows the philologist input the date of the letter in various formats. For instance, the letter can be dated ‘early January 1798’, ‘Eastern 1770’, ‘mid-April 1770’, ‘19/04/1770’, or ‘15/04/1770-20/04/1770’. The conversion of the entry to an exact date or a period is automatically done by the underlying system. ‘Early January 1798’, e.g., is converted to the period 01/01/1798-10/01/1798, ‘Eastern 1770’ to the period 12/04/1770-16/04/1770, ‘mid April 1770’ to 11/04/1770-20/04/1770. After the date had been entered, the platform provides additional information: It runs plausibility checks, e.g., whether the date of the letter is in life of the correspondents, and provides information on historical events that took place immediately before or immediately after that date. For instance, in the case of Eastern 1770, the system would provide the information that the German evangelical theologian August Wilhelm Reinhart passed away on 17 April 1770 and Marie Antoinette got married per procurationem Dauphin Louis-Auguste in the Augustinian Church in Vienna on 19 April 1770. Especially, later on, such data should help the philologist annotate the letters. These latter data can be extracted from digital chronicles, e.g., the ‘Chronik deutscher Zeitgeschichte’(Overesch et al., 1982) for adapting the platform to the project ‘Vernetzte Korrespondenzen’(Burch et al., 2012).

If possible, we will distribute the tool in different digital research infrastructures for the humanities, e.g., DARIAH(DARIAH-EU, 2012), CLARIN(CLARIN, 2011).

This work was supported by the German Federal Ministry of Education and Research (BMBF) [grant number 01UG1354C]

In the last three years there has been an increasing adoption of the variant graph as a suitable computer-internal data model for textual variation in tools and applications such as NMerge (Schmidt & Colomb 2009), CollateX (Dekker & Middell 2011), and Stemmaweb (Andrews & Macé 2013). The variant graph model can be visualized easily and succinctly; these visualizations become for the text researcher the first interface through which to approach the particular similarities and differences in a given text. In this short paper we consider from the theoretical perspective some hermeneutic risks inherent in the static visualization most commonly used at present, when the interface stands between the scholar and the data model (whether produced by a collation tool such as CollateX or NMerge, or generated from a published critical apparatus or text collation). The result of our application of the theory is the tool we present, meant to address these risks in part. Implemented in JavaScript1, the tool allows the user to interact with and modify the model of a particular text from within the browser window. By doing so the scholar can move beyond the 'black box' nature of visualizations produced by analysis tools and test alternate hypotheses for understanding a text.

Most visualizations of data analysis results are by their nature static, and the variant graph has so far been no exception. A typical example of a variant graph visualization is given in Figure 1, which shows part of a collation result of a line of text obtained from CollateX. (Example adapted from the prologue of the Canterbury Tales (Solopova, 2000; Robinson 2000).):

Each of the manuscript witnesses A, B, and C takes a single path through the possible readings, indicating the version of the text that appears in the witness itself. The variant graph is a compelling and concise means by which we can display the phenomenon of text variation. However, from the given example it is clear that there remains room for further interpretation: should spelling variants such as 'teh' and 'the' always be represented as having equal impact on the text with transpositions such as the one between 'drought' and 'march'? Does the alignment of witness C's 'teh rood' make sense as it stands?

The initial model is thus not entirely satisfactory; there is a clear need for the scholar to express more information or to express the variation more exactly. As long as the graph remains static, our example reflects a general danger inherent to visualizations of digital models of humanities data. An interface is meant to make an internal computer model tractable to a user, but paradoxically it also imposes a barrier: it most literally stands between the user and the model. The user can not change the model; he or she can basically only inspect it insofar as the interface visualizes it. In this sense the interface becomes a display practice enforcing one-way communication (Rijcke & Beaulieu 2011). The extent to which a visualization is interactive and alterable will consequently have an inverse relationship to the impression of finality or correctness that the user will come away with. A high degree of interactivity arguably conveys an impression of interpretability and malleability of the data and results visualized. In contrast, a static visualization produced by a computer tool carries with it an aura of correctness or immutability — even if it is not the only such visualization. If the scholar/user is unable easily to interact dynamically with the model, or even to modify it, then the danger arises that the visualization as posited by the computer analysis will be regarded uncritically and accorded a problematic or even false authority.

Current automated collation practices are a good example of this phenomenon. If an automatic collation algorithm such as that employed by CollateX chooses arbitrarily between two equally good text alignment scenarios, then unless the user is able not only to visualize the results but to manipulate them and explore the equally valid alternative scenarios, the ‘answer’ produced by the computer will tend to be seen as the ‘correct’ collation. Interactivity can thus be seen as an inherent and indispensable part of any sort of modeling of humanities data if the model is to be accepted, refined and used more widely in the field. In the case of variant graphs resulting from computed collation, essential interactions include: annotating variants with information about how they are related combining multiple words into one reading (e.g. 'what/so/ever') to better express the mutation as it occurred splitting words into multiple readings to show more finely-grained variation altering or correcting an initial collation

To enable and enhance such interactivity we present a self-contained and generalized JavaScript library for the interaction with and modification of variant graphs produced by online collation services directly in the browser. The library, currently in use by the Greek New Testament Editio Critica Maior project at the Institut für Neutestamentliche Textforschung (INTF)in Münster2and the 'Stemmaweb' application of the Tree of Texts project at KU Leuven3, can be applied to any research context wherein a variant graph would be appropriate. Users can relate parallel variants, merge or split variant readings, and regularize variants where appropriate to more easily home in on the variants under investigation. Figure 2 shows our variant graph example annotated with different types of relationship.

The scholar may make explicit his or her model of variation by defining the types of relationships that can occur between variants, as well as criteria concerning when these variants can or cannot apply (must the variant readings occur in parallel within the text? must they not occur in parallel, e.g. for transpositions? Does the existence of one relationship imply the existence of another?) These interactions with particular instances of a variant text model allow users to adapt, correct, and extend the results of an automated collation engine, thus making the results produced by the software less fixed and more open to scholarly interpretation. To further support interactivity with the model we experiment with methods of machine recognition of user adjustments in context — for instance, we scan the graph using a vector cosine measure that takes into account the two nodes of a user defined relationship and their existing direct adjacent nodes to identify contexts in the graph that are similar to the context where the user adapts the graph, producing automatic suggestions for where the same relationship might apply under the same conditions elsewhere in the graph.

The clear advantage of enabling interactivity is that the problem of one-way communication between visualization and user is resolved: the variant graph model of a particular text becomes a much more tangible and expressive means to engage with and to conduct research into the text in question.

The interactive engagement with a specific instance of a computer model is a first step in our trajectory to enhance interactivity with current state-of-the-art text collation tools. Future research and development work will focus on feeding instance-based manipulations of collation results back into the reference models of online collation services. As a scholar works with a particular graph, the interactions with that instance of the collation model will be captured by decision trees. These trees can be used by the processes backing the collation model to amend the collation and graph construction, not only for specific instances but also for the general model, thus changing collation engines from rather static rule-based automatons into adaptive algorithms that learn from expert input and feedback.

Some humanities scholars argue that the social web “poses a grave threat to the humanities because it lacks the depth, nuance and permanence that make genuine, meaningful interactions about the human condition possible.” They fear that “we risk becoming a world without the comprehensive communication tools needed to keep the humanities alive” (Adamek 2010). Even those who embrace social media all too often dismiss it as a tool that is used primarily for community engagement and self-promotion (Terras 2012). Yet the social web — which we broadly define as the array of technologies that allow individuals to post their thoughts, pictures, and comments in a public forum — when coupled with recent advances in cloud computing, data management and statistical/visual analysis offers significant potential to explore new and enduring humanities questions.

Through careful, rigorous analysis, we believe that the social web affords opportunities for the humanities to realize a contemporary, nuanced understanding of how the public believes our past informs modern society. For example, although there is a strong consensus amongst historians, the broader American public remains conflicted, divided, and confused about the causes of the Civil War. This is evident in any number of recent public polls, such as an April 2011 CNN poll which found that 42% of Americans believed that cause was something other than slavery, and the April 18, 2011 issue of Time, whose cover said “Why We’re Still Fighting the Civil War: The Endless Battle over the War’s True Cause Would Make Lincoln Weep” (CNN 2011; von Drehle 2011). Where polls provide a snapshot of Americans’ views, analysis of online discourse and social media activity affords opportunities to explore modern attitudes towards issues surrounding the Civil War.

Hence, against the backdrop of the on-going sesquicentennial commemorations, this paper examines the intersection of humanities, social sciences, social media, and computing to probe enduring questions around the legacy of the Civil War. To do so, we will illustrate how an ongoing study that uses the Civil War serves as a testbed for examining and developing techniques to conduct traditional types of humanistic inquiry in the context of the social web. Our results demonstrate how careful analysis of the online discourse that occurs across the social web enables deeper understanding of how the Lost Cause Ideology continues to recast the origins of the war and minimize the role of slavery. Simultaneously, we explore how stereotypes of Southerners continue to be propagated and used to shape a memory of the Civil War.

To conduct this study, we demonstrate how contemporary tools developed by industry may be used to analyse the social discourse around the Lost Cause and stereotypes of Southerners. Most scholars researching the social web rely primarily on the use of single keyword search and user-specified hashtags to narrow the size of their datasets to only posts that are immediately relevant to the topic under discussion (for example, Graham 2012; Ross, et al. 2010). However, these methods can be inadequate or problematic for understanding widely diffused social phenomena. For example, when attempting to study a topic such as the Civil War or especially a concept such as “the South” this methodology isn’t adequate as the majority of users do not tag their casual online conversations with these types of metadata nor do they restrict their conversations to a single platform (e.g., Twitter, Facebook etc.). To elicit a holistic view of conversations on the social web, the researcher has to employ strategies that reach across social media platforms and go beyond simple hashtag sets. To overcome these limitations, we leverage software originally developed for business analytics to aggregate content from across the social web (Radian6, 2012) — our search includes not just Twitter and Facebook, but also content from sources such as mainstream news sites, blog posts, and forum posts.

Through our analysis of the social web, we highlight issues and opportunities for scholars who work at the intersection of the humanities, social science, social media, and computing. One issue that must be overcome is that, on a global scale, there are many Souths: not only must we find a way of eliminating references to countries or continents such as South Africa, South America, and the South Pole, we must also filter out conversations related to places such as the South Side of Chicago and even the television show South Park, none of which would be relevant for this research. Similarly, while most Americans would refer to the American Civil War as simply “The Civil War,” there are many similarly named historic and ongoing conflicts that we must filter from our results. The solution is to create “topic profiles” — collections of words that fall into one of three categories: words or phrases that must be present in a post to be included in the results; phrases that must be present along with words from the first category to help categorize results into different topics; and, finally, phrases that, if present in a post, will result in that post being excluded from our results. So, when conducting research into online discussions of southern gender ideals, we create a topic profile similar to the following: 1. Posts that CONTAIN any of the following: “the south”, “thesouth”, “southern” 2. AND CONTAINS any of the following: “belle”, “lady”, “gentleman”, … 3. But DOES NOT CONTAIN any of: “south pole”, “south park”, “south side”, “south africa”, …

By using keywords that were selected following a survey of users of the H-South discussion list, a request from leading scholars (including several former presidents of the Southern Historical Association, of the American South, surveys of a graduate and an undergraduate history class on Southern History at Clemson University, as well as a non- random sample of black and white southerners who were not academics, we have constructed topic profiles for five different themes that might be used to segment contemporary discussion of the South online: Southern Culture (which includes concepts such as honor, hospitality, and accents), Southern Food, Discussions of Gender, Religion, and Southern History. This set of keyword groups results in approximately 300,000 unique hits from the social web each month. Figure 1 illustrates the relative volume of each keyword group during September 2012 and shows how the number of hits varies for each topic across the month. Similar keyword groups have also been constructed for specific events that occurred during the Civil War, such as the attack on Fort Sumter, the Emancipation Proclamation, and the Battle of Gettysburg.

Drilling into this data reveals numerous insights into how users of the social web conceptualize the South and what topics are of interest. Figure 2(a), for example, presents a word cloud of the major terms associated with civil war commemorations, again for the month of September 2012. This word cloud immediately reveals several potential avenues for further investigation, including the centrality of Gettysburg to the online discourse surrounding the Civil War, but also suggests other useful topics to explore. Indeed, although word clouds have frequently been criticized for divorcing words from their context (Harris, 2011), we are able to maintain the connection between the words represented in a cloud and the underlying data. Therefore, if we are interested in further exploring the appearance of the word “history”, we can, as shown in Figure 2(b), drill further into the data to reveal new levels of insight to our topics. At all times, the original posts remain available both to view at an individual level to ensure the relevancy of content and to download for further textual and statistical analysis with dedicated software packages such as R and Gephi.

We can also analyse demographic information, such as age, gender, and location, from the posts we have collected, allowing for a more nuanced understanding of our results. Figure 3 illustrates the location of users in the United States whose posts and tweets were collected from the Civil War topic profile over a seven-day period in mid- September 2012, while Figure 4 presents word clouds of the conversations that occurred in our Civil War topic profile on the 150th Anniversary of the Battle of Antietam by (a) people located in Maryland; (b) those in North Carolina; and (c) users aged between 36 and 45. These charts reveal not only the geographic distribution and volume of conversations, but also regional and demographic differences in the topic, tone, and scope of this online discourse.

Desde la década de los 80s hasta los 2000 el Perú vivió en medio de un conflicto armado entre dos movimientos terroristas, Sendero Luminoso y el Movimiento Revolucionario Túpac Amaru (MRTA), los cuales buscaban derrocar el sistema democrático peruano, desatando uno de los más violentos pasajes de la historia moderna.

Estos movimientos fueron vencidos militarmente, desarmados y encarcelados, pero a un alto costo en vidas humanas. A pesar de la desarticulación de estos grupos aún se reportan células armadas y organizadas en los lugares más alejados de la selva peruana. Al mismo tiempo, en las ciudades buscan impunidad criminal y política para sus líderes arrestados a través de mecanismos legales que les permitirían introducirse en la política activa.

Hoy, casi 10 años después del fin de este episodio de la historia, el museo que conmemora el recuerdo de las víctimas de la violencia, el Lugar de la Memoria del Perú, no ha sido terminado. Las nuevas generaciones no conocen claramente lo sucedido debido al deficiente programa educativo; tal que estos grupos subversivos aprovechan la desinformación para entregar una imagen distorsionada de esta parte de la historia. (Pisa, 2009). Sin embargo, somos la octava comunidad con mayor uso de las redes sociales (Comscore Inc. 2012) e Internet en el mundo. Asimismo, Perú ya ha sobrepasado el 100% de penetración móvil.(Budde.com 2012)

Estos índices muestran que Perú tiene una oportunidad única y sin precedentes de desplegar servicios que asistan al desarrollo y mantenimiento de una memoria colectiva usando tecnologías móviles.

La misión del proyecto evolucionó con el tiempo y se ha trabajado y conversado con expertos del tema.

Memoragram nace a partir de la idea de crear un museo geo-localizado de la memoria, accesible desde cualquier dispositivo sin importar la hora ni el lugar. Los usuarios podrían explorar eventos históricos cercanos a su ubicación actual o hacia la fecha que deseen consultar, revisar los personajes involucrados, los lugares (en su mayoría demolidos), organizaciones e incluso revisar qué libros, revistas, periódicos, películas, reportajes de TV, cómics, entre otras fuentes, hacen referencia al hecho.

Por el lado tecnológico, los usuarios podrían usar smartphones o tablets para que, al acceder a una web móvil, esta solicite permiso para usar el GPS del dispositivo y así mostrar contenido relevante. La tecnología que hace posible este acceso es HTML5, CS3 y Javascript. Asimismo, los usuarios podrían descargar una tercera aplicación móvil llamada Junaio, la cual, usando el GPS, giroscopios, brújula y cámara puede mostrar contenido en una vista de tiempo real.

Si bien esta parte del alcance no ha sido modificada, surgieron otros problemas fuera del campo tecnológico.

Con la idea y el alcance del proyecto se empezó por buscar fuentes de información fiables, entre ellas el Banco de Imágenes de la Comisión de la Verdad y Reconciliación Nacional del Perú (Comisión de la Verdad y Reconciliación del Perú), la cual se basa en los archivos fotográficos de los principales diarios y revistas locales peruanos. Este mismo banco se usó para formar la muestra Yuyanapaq (Programa de las Naciones Unidas para el Desarrollo — PNUD Perú) (término quechua que significa ‘Para recordar’). Este archivo posee alrededor de 1560 fotografías.

Luego de una evaluación de esta fuente y del mapeo de eventos posibles a registrar se concluyó que el número de eventos registrables sería mucho menor a 1560. De encontrarse otras fuentes de datos el número sería más limitado. Esta situación conllevaría a que la oferta de contenido fuera fija, lo cual podría evitar que los usuarios no se familiaricen con el sistema o no tengan una razón para regresar al servicio.

La memoria colectiva se da como un proceso social y natural en el que las generaciones mayores traspasan conocimiento a las generaciones más jóvenes a través de relatos, documentos, mapas, elementos multimedia como fotos y videos, entre otros.

Este acto puede tener un efecto terapéutico en las personas que vivieron hechos traumatizantes.

A la vez, en el entorno de marketing en Internet (campo en donde se desenvuelve el autor) se hace presente la tendencia llamada “LoSoMo” que son las siglas de tres elementos que se recomienda considerar en una campaña digital: “Local”, “Social” y “Mobile”. Este enfoque se puede encontrar en las más exitosas start-ups como Foursquare y Groupon. A este momento, el alcance de Memoragram sólo abarcaba el primer y último aspecto. El fenómeno de traspaso de la memoria entonces se postulaba como el elemento social para generar una comunidad activa de usuarios.

La inclusión de un fenómeno social en el alcance del proyecto abrió un nuevo espectro de posibilidades para la idea de formar un museo virtual. El sistema empezó a evolucionar: pasó de un servicio en el cual los usuarios sólo consumen contenido a una plataforma donde pueden aportar con sus propias memorias, sus propios recuerdos.

Este nuevo tipo de contenido obliga a generar nuevas acciones que se pueden expresar y medir con la plataforma, por ejemplo: los usuarios exploran el contenido y pueden registrar y medir el efecto que este genera en ellos, logrando que esta medida de rankings sea otro criterio de organización de los eventos listados.

Con las consideraciones listadas previamente se definió el alcance del proyecto, el cual se alcanzará a través de una serie de prototipos funcionales en proceso de perfeccionamiento.

Los usuarios podrán explorar los eventos históricos más cercanos a su ubicación actual, la fecha en que sucedieron, los personajes, organizaciones y las publicaciones que las mencionan como fotografías, video, audios, archivos periodísticos, libros, películas, entre otros. Asimismo podrán dejar sus propios recuerdos usando texto, audio, fotografías e incluso video, a manera de mensaje a las futuras generaciones (recuerdos públicos).

El contenido al inicio será creado con la asesoría de especialistas basados en fuentes confiables.

Para construir la plataforma se requiere de un paquete de software web modular, preferentemente gratuito para ahorrar costos y open-source para aprovechar el avance de la comunidad que lo mantiene. Se eligió entonces el CMS (Sistema de gestión de contenido) Drupal 7 basado en el lenguaje PHP, el cual se conecta a una base de datos MySQL (Metaio Inc.) Este sistema ya incorpora funciones de manejo de registro de usuarios, creación de perfiles de usuario, tipos de contenido y una robusta estructura de etiquetas así como soporte de distintos tipos de archivos multimedia.

En la versión 7 de Drupal ya se maneja HTML versión 5, llamadas asíncronas AJAX e incluso soporte de diseño responsivo, el cual permite crear una sola interfaz que adapta el contenido mostrado según el dispositivo desde el que se accede a él.

Asimismo, se decidió experimentar con una interfaz de realidad aumentada para móviles; para ello se eligió la aplicación móvil Junaio www.junaio.com, disponible para iOS y Android, a la cual basta alimentar con contenido desde una interfaz entre servidores usando XML.

A continuación pueden observarse unas capturas de pantalla. Todas las imágenes son de elaboración propia, se realizaron desde una PC con una navegador web y en un Apple iPhone 3GS.

Es técnicamente viable crear una plataforma que llegue al alcance propuesto, sin embargo, a la fecha sólo se ha implementado el acceso web y realidad aumentada móvil sin la interfaz de creación de memorias por parte de usuarios. En el futuro cercano se irán extendiendo las funcionalidades y luego se pasará a realizar pruebas de usuario para diseñar una interfaz adecuada.

A key challenge facing professional researchers in the domain of cultural heritage across Europe and worldwide is the interrogation of growing digital humanities collections. However, the full value of these heritage treasures is not being realised. After digitisation, these collections are typically monolithic, difficult to navigate and can contain text which is highly variable in terms of language, spelling, punctuation, and consistency of terminology. These difficulties are compounded by a lack of normalised spelling in most European languages before the eighteenth century. This means that search across these digital collections tends to return sub-par results as multiple spellings for many common words are treated as independent document keywords. CULTURA is a corpus agnostic environment with a suite of services, including personalisation, annotation, and recommendation, providing necessary supports and features for a diverse range of professional researchers.

In order to empower communities of researchers with personalised mechanisms which support the collaborative exploration, interrogation and interpretation of complex digital cultural artefacts, the adaptivity provided in CULTURA is required to be integrated and intelligent. The next generation adaptivity provided by CULTURA supports the dynamic composition and presentation of digital cultural heritage resources. Automated adaptivity however, is not enough on its own. Ensuring that the user is in control of the personalisation process is essential. Such user-centred control is enhanced through correlating usage patterns with self-expressed user goals; pre-defined strategies (e.g. research strategies, investigation strategies, discovery strategies, explanatory strategies etc.); and the provision of appropriate tools for users to explore and navigate large cultural heritage information spaces.

A central aspect of the CULTURA environment is its use of rich metadata (user generated, computer generated and expert generated) coupled with natural language processing, entity extraction and social network analysis techniques, in order to support collaborative exploration, interrogation and interpretation of the underlying cultural resources. The Qviz (http://www.qviz.eu) project has some similarities in approach to the CULTURA project in that it makes explicit recognition of the value of users as members of communities, and as contributors to digital cultural heritage collections, however Qviz, however, does not incorporate a personalised or adaptive aspect.

The manual determination of descriptive metadata across a large corpus is too time-consuming to be practical. For example, the process of metadata identification for the 8,000 1641 depositions took a research assistant 12 months (http://1641.tcd.ie). An automated entity extraction process is used by the CULTURA project (Carmel et al. 2012). This process interprets words and combinations of words to identify entities in the corpus such as people, places, events, and dates. Complex entities are then constructed from these entities and the corpus allowing the identification of events, such as WHO did WHAT on such a DATE at a specific LOCATION.

Entity extraction is most powerful on a corpus in which all the content has been normalised as entities can be matched across multiple documents. To enable normalisation a ground-truth is manually generated across a proportion of the corpus, approximately 10% in the case of the 1641 depositions (Hampson et al. 2012). The ground-truth assigns non-normalised terms in the corpus to normalised terms. The ground-truth is applied in the generation of a statistical model that is then utilised on the entire corpus outputting a normalised corpus. It is this normalised content from which entities are identified and extracted. The linguistic model used in the normalisation of the 1641 Depositions has proved highly reusable, and provides robust results when applied to a range of other material contemporary to the corpus used to generate the model, enabling the re-use of the model on additional content collections.

The output of these processes ensures each individual piece of content from the corpus has descriptive metadata in the form of individual entities such as a person or place, and complex entities such as where a person lived (i.e. compound entities) as well as a normalised variant. This enables a simple keyword search to provide results that cover the entire corpus and all variants of spelling of the terms entered. Additionally, and more significantly, individual pieces of content from within the corpus can be linked to other content that contains information on the same event or similar events, based on the date, location, people, and type of event. These links enable professional researchers to quickly identify content that is related to their current research topic, by way of visualisations of these links (Hampson et.al. 2012).

While the automated tools are providing useful results for professional researchers, these state-of-the-art tools cannot replace the insight and experience of a professional researcher. These insights can be captured via an annotation tool that can be used to annotate specific aspects of any piece of content, both textual and visual. Annotations can be at any level, from a single word up to the entire document, or any user-identified region of an image. Annotations also allow the researcher to link the identified content to any other document within the corpus. Links such as this feed a professional researcher’s knowledge into the system, and can be used to aid in the adaptivity and personalisation of the system for all users.

Collaboration between professional researchers occurs in an implicit manner via the use of annotations as link generation between content, which feeds into recommendations and personalisation. Explicit collaboration is also present in CULTURA. Researchers can share annotations with other users, enabling the propagation of insights and discoveries in the content. As annotations are anchored to specific elements of a document they provide a powerful mechanism for inserting detailed and relevant knowledge. This knowledge can be made available to either groups of researchers or all researchers thus enabling a greater collaboration across all users of the environment.

The CULTURA environment has already been used by professional researchers in the course of their research. Services including annotations, document level notes, and multi-dimensional recommendations of content were enabled within the environment.

A number of the professional researchers involved in trials of the CULTURA environment were initially wary of the potential impact of adaptivity on their research. They expressed concern that the recommendation system would create a ‘filter bubble’, distorting the appearance of the collection. As they engaged with the environment, however, these concerns were abated. Research into the user requirements of these end-users had identified this concern, and a concomitant need for a high level of scrutability. In line with this, recommendations made by the CULTURA engine were very explicitly presented in a way that made it easy for researchers to decide whether to use them, or to ignore their suggestions. In addition, the interface made it clear why a given user was being recommended a particular piece of content.

After some initial caution, professional researchers agreed that the adaptive recommendations provided by CULTURA were genuinely useful. This was especially true of researchers who were previously unfamiliar with the CULTURA corpora. These users reported that the recommendations facilitated their mastery of the collection. Researchers who had greater familiarity with the collection rated the utility of the recommendations less highly, but still expressed an appreciation of their potential usefulness. In particular, they suggested that the recommendations were valuable in encouraging them to look at the collection in new ways.

The user model for the professional researchers was evaluated and determined to be an accurate reflection of the researcher’s interests and topics under research. While some anomalies were identified in the user model, the areas of interest that were weighted most highly correlated with the topics of the research.

The CULTURA environment is currently engaged in additional evaluations with trainee researchers utilising a broader set of documents, tools, and adaptivity.

In the following paper a new research project undertaken within the framework of the AAC–Austrian Academy Corpus operated by the Institute for Corpus Linguistics and Text Technology at the Austrian Academy of Sciences in Vienna will be presented. The topic of this research project is focused on the questions of developing a diachronic text corpus of historical significance and establishing a corpus based research environment for language studies of the interwar period with particular emphasis on the year 1933, the year when the Nazis came to power in Germany. In the presentation of this project an overview of the necessary methodological considerations and an outline of the research perspectives based upon the principles of corpus linguistics will be given. Corpus-based approaches for analyzing the language of the historical periods before, during and after National-Socialism have been rare, despite the numerous works in the fields of historical studies as well as in German language studies. A research group of the corpus research framework of the AAC has decided to start a special investigation into this challenging question.

The AAC is an established German language text corpus of more than 500 million tokens and represents a considerably large diachronic digital text corpus comprising several thousands of German language texts of important historical and cultural significance. The core of the AAC is from the first half of the twentieth century, so that the research issue of an analysis of the German language of 1933 is not only highly appropriate, but can be addressed on a comprehensive basis. Among the sources of the AAC a large number of texts of the historical period in question have already been collected, digitized, converted into machine-readable text and fully annotated as well as been provided with metadata. Structural and thematic mark-up has been applied according to annotation and mark-up schemes based upon XML related standards. The AAC has been committed to the research field of text technology since its official foundation in 2001.

Building a diachronic digital text corpus for historical German language studies of this particular kind is a particularly challenging task for various reasons. First, the technical difficulties of corpus building in dealing with a large historical variety of different text types and genres have to be taken into consideration. Second, the specific historical parameters and the methodological scope of such an investigation has to be taken into account. The German language of the year 1933 is being considered as a historical focal point for which an exemplary corpus-based research methodology for the study of the German language could be developed. The sources of a first exemplary study will cover manifold domains and genres, not only newspapers and political journals and magazines, which will be at the core, but also several other text types representing the historical communicative strategies will be included. Among them are pamphlets, flyers, advertisements, radio programs, political speeches, but also essays and literary texts as well as administrative, scientific or legal texts, just to name a few examples, which are all difficult to collect. The AAC has started to build up a small collection of ephemera in this field.

In the overall project a special emphasis will be given to the "Dritte Walpurgisnacht" (Third Walpurgis Night), written by the satirist Karl Kraus, which will be taken (in digital format), among other sources, as a starting point for text selection for the corpus. This text is the most important contemporary text of German literature dealing with National Socialism. In the "Third Walpurgis Night" Karl Kraus has documented the murderous reality of the Nazi regime as early as May 1933 and documented and commented upon the murderous language of that time in numerous examples. Because of this text no one can claim not to have been able to know from the very start where Nazi rule would lead. However, Karl Kraus, the editor and author of his journal "Die Fackel" (The Torch), who died in 1936, did not publish his text, a text which begins with the famous line "Mir fällt zu Hitler nichts ein", because in the face of violence the deed of the word was considered inappropriate by him.

The historical period covered by the AAC is ranging from the 1848 revolution to the fall of the iron curtain in 1989. In this period significant historical changes with remarkable influences on the language and the language use in the German speaking areas can be observed. The year 1933 and the years preceding as well as following the "Machtergreifung" (seizure of power) of the National socialists is a historical period of particular interest for language studies. In this case not primarily the well-known documents and the evident language of the Nazis will be included in the analysis, but systematically the less easily visible documents and less significant lexical items will be taken into consideration as well. This methodological approach is considered as particularly fruitful by means of applying methods of corpus linguistics and by testing new strategies of the application of these methods in the context of historical language studies. For this historical period the AAC corpus holdings provide a great number of reliable resources and interesting corpus based approaches for investigations into the linguistic and textual properties of the texts in question. The digital text is going to be enriched by additional data and will be lemmatized and provided with POS data thereby making use of the tag set for this purpose of the STTS (Stuttgart-Tübingen Tagset). "Quantitative corpus linguistics has proofed to be a valuable technique in many domains of philological, sociological and historical research. The digitized and linguistically annotated corpus is therefore an interesting source for studies in many fields and facilitates the investigation of changing patterns of language use, and how these reflect underlying cultural shifts." (M. Volk). The question is, whether corpus research methods based upon a multidisciplinary combination of corpus linguistics, lexicography, historical studies and cultural studies can be applied in order to gain insights into the textual representations of historical collections of this importance.

The AAC research group will go beyond a quantitative approach and integrate text studies into its research of the German language of 1933. The methodology of corpus based text research is determined by corpus linguistic, lexicographic and analytical procedures. The historical condition of Germany and Austria with their cultural and linguistic diversities and in particular the situation at the time of National Socialism have to be taken into consideration as historical changes with significant influences on the language. In contrast to other corpus-oriented projects, the working group proceeds from literary studies and text lexicographic premises. Corpus research and the creation of large electronic text collections have traditionally been the domain of corpus linguists. Literary digitization initiatives were quite often restricted to particular writers and many of these projects did neither produce large amounts of data nor pursue research on methods of how to tackle the problems involved in working with such data. Our perspectives parallel those formulated in the European project CLARIN which has been set up to "create, coordinate and make lan-guage resources and technology available and readily usable" (Call for Proposal), in this case also for text historians and for those interested in ideologically determined language change.

The AAC has already developed methods and tools to allow scholars to access these texts and other comparable resources. For this purpose the tools provided in order to access the corpus holdings will enable the researcher to input queries and to get a display of the results in forms provided with the necessary metadata such as information on sources, authors, date of publication etc., and with a display of the related pages of the results as digital text, also allowing access to the XML source of the texts and to define custom style sheets, alongside a feature to view facsimiles of the texts offering simultaneous access to text and facsimiles of pages. In addition to that a sophisticated navigational control tool will be provided, offering random access to the variety of different documents, journals, books, etc. to do linguistic, literary or historical research. Access to the text corpus will be given not only through query result lists but also through a structuring tool, which allows readers to navigate to any desired part of the corpus and results of queries are delivered by the server in XML format which makes it fairly easy to adjust the representation of the output, where XSLT style sheets can be used. Using such style sheet transformations also allows creating statistical analyses of the data. And it has been pointed out before (Smith, 2008) that available standard tools provide only limited support when processing query results. Building a diachronic digital text corpus for historical German language studies of this particular kind is a particularly challenging task which demands also the development of new tools and new approaches of text technology. This special research environment would be especially useful for corpus-based analyses of the language of critical historical periods such as the case of the German language in the year 1933.

Collective Biographies of Women (CBW) is a collaborative1, open-access2 literary and prosopographical project focusing on published collections of biographies of women 3. The literary focus of this interdisciplinary project (which also centers in women's history in Britain and the U.S.) concentrates on a popular genre that has received little critical attention. We study the narrative structure of short biographies of diverse women’s lives delineated through interpretive analysis captured in a stand-aside XML mark-up using the BESS schema4. Having presented the theoretical and literary aspects in multiple venues5, in this short paper we will present emerging prosopographical interpretations of a database of over 1200 volumes, comprising more than 13,000 biographies of more than 8000 women. The women in this corpus come from all walks of life (not simply one occupation or nation, such as the women writers considered by Orlando or Brown's Women Writers Project6), and range from ancient and biblical figures to living contemporaries of the biographers.

In our proposed paper we introduce the phrase, “documentary social network,” as a label for our prosopographical interpretations because we are interested in social networks as discovered in biographical documents7. Yet these networks do not always resemble those that can be found in an investigation of Twitter feeds or Facebook “friends” or even of archival materials such as provided by SNAC8. The social networks evidenced through our collection of collections are those of documentary grouping and reference. In specialized collections, Christian missionaries in Africa or nurses in World War I may have interacted in historical time and place, but assorted tables of contents commonly link some individuals who never actually shared a "live" event or a relationship or communication. Thus, the relationships are as perceived and presented through printed volumes collecting short versions of biographies.

Let us consider one Mrs. John Livingstone. She appears in only three CBW collections and by way of those collections her immediate documentary social network has 37 other women9. To visualize this network, Figure 1 is centered on Mrs. Livingstone with the three collections positioned on the innermost circle and the other women of those collections around the second circle. Of her 37 collection “siblings,” nine also appear in these same three collections and nowhere else in CBW, confirming some consensus on the names that represent "Notable Women of the Scottish Reformation," a specific historical episode in one location. Other members of Livingstone's documentary “siblings” range into other collections and in Figure 2 those additional collections are positioned on the third circle. In more eclectic lists, some of the Scottish heroines of religious conflict intersect with a multi-national set of women widely recognized today, including writers such as Harriet Beecher Stowe and heroines of war such as Joan of Arc or the Countess of Montfort. Mrs. John Livingstone’s 267 documentary “cousins” are displayed on the fourth circle in Figure 310.

CBW's first experiments in digital analysis of narrative structure using the BESS schema have focused on two sample archives that are also productive for this paper's claims about documentary social networks. (Each sample archive is a set of all female collective biographies that include a certain woman, as in Fig. 1.) The two sample archives are "Noble Workers," 20 books that include a biography of Sister Dora, who ran hospitals in the industrial Midlands, and "Women of the World," 14 books that include a biography of Lola Montez, a celebrity in Europe, New York, California, and Australia. These Victorian women never met and never appear in the same book. The mediated interconnections between Sister Dora and Lola Montez appear in Figure 4. Only two women, Marie Antoinette (presented as a victimized queen) and Jenny Lind (the celebrated, virtuous singer), appear with Sister Dora in one collection and with Lola Montez in another (i.e., within one degree of separation of both Dora and Lola). The multiple versions of the lives of 141 women who "network" with Sister Dora in Noble Workers collections, compared to the versions of the 133 persons linked to Lola Montez in Women of the World books, will demonstrate the utility of our approach to analyze historical social networks and nonfiction narratives of many kinds. In spite of the vast differences between two Victorian women — one a saintly nurse, the other a notorious courtesan and performer — we discover patterns among their associates and their proximity in the CBW documentary social network.

These prosopographical networks reveal unexpected affiliations among different types of women, life stories, and collections; with the CBW tools for searching and visualization, we are now able to calibrate not only frequencies and proximities of persons and publications over time, but gradations of rhetorical assessment of women's roles and deeds, according to the perspectives of these publications11. Such multivalent interconnections not only cross categories but also historical periods in significant ways. An example is the connection between Sister Dora and Anne Boleyn, shown in Figure 5. This diagram reveals that the middle-class, now-forgotten Englishwoman, Sister Dora, appears with her contemporary queen, Victoria (in seven books), and in two collections with the martyred queen, Jane Grey (one of the latter, a186 12, is a book that Victoria also shares), but not with Henry VIII's second wife13. At sufficient scale, and integrated with other elements in our analysis, data on frequency or "distance" in networks can lead to productive interpretations. Anne Boleyn, unlike good Lady Jane in ambition and sexual notoriety , but like her in being executed because of the politics of English monarchy, features in 24 collections, including 11 dedicated to English queens and three focused on the English Reformation; the latter three collections include Lady Jane Grey but not one of the obscure Scotswomen of Mrs. Livingstone's ilk. To put it simply, Boleyn serves English and Protestant history, but not the promotion of feminine virtue, heroism, or social service.

More complex documentary interrelationships can be seen in Figure 6, associating the Gothic novelist Ann Radcliffe (dead before Sister Dora was born) with Sister Dora through a range of widely respected eighteenth-and nineteenth-century women writers; these lists also commend Jeanne d'Albret, the French queen regnant who championed the Calvinist Huguenots, as well as Lady Russell, Mme. Roland, and Lady Jane Grey, represented as good, highly educated women who played indelible parts in the history of their countries in revolutionary times.

Our proposed paper will present the methods and implications of prosopographical analyses afforded by the CBW documentary social network. Our demonstration of documentary social networks has implications for any "personography," as well as for historical studies of women and studies of biographical narrative. We argue, in this paper, that digital exploration of the persons, narratives, collections, and documentary social networks in the CBW genre opens a prolific field of ramifying versions of female biography impossible to decipher through an approach to individuals or single full-length biography, and invisible through the customary lenses of historiography, time and period, place and nationality.

The transcription of primary textual sources is at the center of digital editing projects. The capacity offered by the digital medium allows for the inclusion of much more detailed transcriptions than is possible within the frame of printed editions. But this capacity tempted scholars into attempts to capture every possible aspect of the document in their transcription. Unless we are very careful about the choices that we make in relation to the transcription of a document, we risk making an enormous effort and spending a great deal of money on transcriptions that end up not being very useful. This paper describes our use of an encoding system that allows scholars to present a transcription of the text of the document alongside the variant states of the text, making it possible to go beyond the encoding of documents.

Some important projects in digital humanities are focused on the transcription of documents. Notable examples of this are Transcribe Bentham (http://www.ucl.ac.uk/transcribe-bentham/), the Jane Austen’s Fiction Manuscripts Digital Edition (http://www.janeausten.ac.uk) or Nietzsche Source (http://www.nietzschesource.org/). Moreover, editors such as Hans Walter Gabler who have made the transition from print to digital and who advocate critical editions are shifting their focus towards the transcription of primary sources. For example, Gabler, in his article “The Primacy of the Document in Editing,” asserts that:

This re-definition of text as construct from a document develops in Gabler’s argument to become “…a set of document functions comprehensively deriving from the continuous manuscript posited…” (207). This represents a shift from what has been considered the traditional role of the editor as the scholar that establishes the text. Indeed, Peter Robinson points out that:

Robinson’s interpretation of Gabler’s text is informed by my own concept of the text of the document, as developed for the article “The Commedia Project Encoding System.” While working on the electronic edition of the Divine Comedy, I developed a complex encoding system to represent, within a single file, the text of the document and the variant states of the text. In this article, I explain that:

I am not the first in referring to text of the document, G. Thomas Tanselle has contrasted the text of the document (1978; 1996) with “the text of the work” (Tanselle, 1996).

The text of the document, then, can be recorded as a sequence of meaningful marks present on a particular document. MS. Riccardianna 1005, a witness of Dante’s Divine Comedy, presents the following example:

This could ordinarily be represented by: <del>dura</del><add>duro</add>

In our transcriptions, however, we represent this as: <app>
 <rdg type="orig">dura</rdg>
 <rdg type="c1">duro</rdg> <rdg type="lit"> dur<hi rend="ud">a</hi>o</rdg> </app>

Here, the last reading, <rdg type="lit">, represents the text of the document literally. Unlike traditional TEI recommended encoding we do not use <add> or <del>, on the premise that both of these confound a statement of editorial interpretation (this is a deletion or an addition, so we should read the text with or without these characters) with a statement of what appears on the document (there is a mark under these characters; they are placed interlineally). In our view, the statement of what we see on the page is a different order of declaration from a statement of how we think this should be read. The aim of <rdg type="lit"> is to present the meaningful marks we see on the document. The statement of how we think these marks should be read, as a sequence of readings, should be left for the other elements within <app>: thus, <rdg type="orig"> and <rdg type="c1">, representing the different perceived stages a particular text has gone through during a process of revision, from its original (“orig”) to first corrected state (“c1”). All these statements — <rdg type="lit">, <rdg type="orig"> and <rdg type="c1"> — are interpretive. However, they are interpretive in different ways and serve different purposes.

The TEI, in “An Encoding Model for Genetic Editions” (http://users.ox.ac.uk/~lou/wip/geneticTEI.doc.html#index.xml-body.1_div.1_div.1) includes some similar elements to the ones developed for our edition of the Commedia. Consider this example from “An Encoding Model for Genetic Editions”:

In our proposed system, not originally designed for genetic encoding but easily adaptable for this purposes, we would encode it as: <app> <rdg type="orig">Now, if we could but get </rdg>
 <rdg type="c1">Now, if we could get</rdg>> <rdg type="lit"> <seg type="overwritten"> <seg type="orig" rend="pencil">Now, if we could but get </seg> <seg type="c1" rend="ink">Now, if we could get</seg> </seg> </rdg> </app> Both systems explicitly state that there was an original text, written by Jane Austen and that the text was rewritten by herself. The “lit” reading in my proposed transcription specifies that this was originally written in pencil and that the second version of the text was written in ink. In my transcription, it is clearly stated that the original form of the text was “Now if we could but get ” (written in pencil) and that this was altered to “Now if we could get” (written in ink). In contrast, in the proposed TEI encoding it is not at all clear from the use of the various attributes (stages “0” and “#s1”) and from the <ge:rewrite> and <del> elements what was originally written, what it was revised to, and how this was done.

For our purposes, we find that by stating explicitly what occurred to the text, that is whether it was stroke through, underdotted, underlined, erased, scrapped or rewritten, we avoid the ambiguity of the <del> element. The <add> element has also been set aside, to be replaced with markup that indicates position or mode of insertion.

The encoding system being used in our projects is TEI compliant, but it attempts to serve the purposes of various types of scholarly editions. For the purposes of textual scholarship, it is necessary to have a clear idea of what is meant by “representing the text of the document.” In any case, it is not enough to represent the text of the document, we must also represent variant states of the text and give editorial interpretation a more explicit place in the transcription of primary sources.

The emergence of formal ontologies into the World Wide Web has had a significant effect on research in certain fields. In parts of the Life Sciences, for example, key research information has been captured in formal domain ontologies, like those mentioned in the Open Biological and Biomedical Ontologies website (OBOFoundary 2012). In parallel with this has been the development of the AO annotation ontology framework (AO 2012) which formalises the act of annotation as a way to connect ontologies such as those in the OBOFoundary to references to them in the scientific literature: an act sometimes referred to as "semantic annotation", and tools such as the SWAN annotation system (SWAN 2008) have emerged to support this. We will call the activity of linking references in a domain literature directly to entities in one or more domain ontologies "direct semantic annotation", and show it in schematic form in figure I. The annotations — shown as heavier lines connecting spots in the literature (to the left) to the ontologies (to the right) could be expressed in the AO annotation ontology, or something similar to it.

Can direct semantic annotation like this be applied to research in the Humanities? For it to work as it does in the Life Sciences, formal models of humanities materials, built using an ontology framework such as CIDOC-CRM, would need to exist and be already accepted as a useful representation of material of interest to the humanities. Not much of this has happened at present, although perhaps Linked Data initiatives (Heath and Bizer 2011) show some promise in that general direction, and we in DDH are both exploring making data from our projects available in the Semantic Web sense by, for example, setting up SPARQL endpoints, and thinking about the open-data significance of our structured prosopography model (Pasin and Bradley 2012).

Although open, linked data provides a context for exploring direct semantic annotation and even though there is evidence of this being thought useful in the Life Sciences, by itself the mere act of linking a spot in a published online journal to a relevant bit of an online ontology does not represent anything other than rudimentary research activity in either the Life Sciences or in the Humanities. Instead, almost all humanities scholars want to spend their time, not connecting things they read only to an existing, shared, understanding of things, but instead developing their own original interpretation of the materials they study, and they aim to subsequently explore these new concepts and paradigms in articles and books that they write. (see Brockman et al 2001 and in Palmer et al 2009) For traditional humanists, their scholarship does not start out only with predefined formal structures such as those provided through their community's shared concepts, but begins with a set of vague notions and insights that emerge more clearly over time in the scholar's mind as they read, and that only over time becomes clear enough to be described in original published work. For most humanists scholarship (a) is normally personal, (b) is meant to produce original ideas that must first emerge and then mature over time, and (c) even when the ideas are mature enough for publication, represents a structure that is at least "pre-ontological", and perhaps at best only partly compatible with the clarity of ontological modelling.

Surprisingly, however, although products of humanities scholarship do not seem currently to match the formalisms of computer ontologies as perhaps some of the Life Sciences do, there is evidence of some degree of inherent structure in the process of creating them. Many researchers, including Brockman and Palmer mentioned earlier, have noted the importance of notetaking and the management of those notes in humanities research. We can see a significantly structured approach around the process of developing new scholarly ideas when we look at traditional strategies for taking and managing notes as described in books like Altick and Fenstermaker's The Art of Literary Research (1992). Altick and Fenstermaker describe paper-based procedures aimed to provide the new researcher with a methodology to organise notes taken while reading into a structure of topics and concepts that will eventually contribute into the writing of articles that represent publishable thinking. There is structure in Altick and Fenstermaker's approach: figure 2 shows this process in schematic form. We see the original notes created during the reading of articles and books on the left, these notes contributing thoughts in the mind of the scholar that eventually allows him to create new concepts in the middle (only 2 "concepts" are shown here, but a real user would have many more), and (towards the right) these new emerging concepts fitting with references to original sources and supporting an argument for new ideas to be presented through the writing of papers. Note the difference from direct semantic tagging in figure I: the annotations do not link directly to preexisting formal ontological entities, but first appear as informal prose notes that may, as the researcher's understanding grows, contribute to a more formal set of ideas and then emerge as entity-like objects in the form of personally developed new concepts, themes, ideas, etc.

If we wish to explore how traditional personal scholarship could connect to the formal world of RDF and ontology schemes such as the Open Annotations Collaboration (OAC 2011), we need a model that reflects some of these aspects of the act of doing personal scholarship. The models behind the Pliny project (Bradley 2008 and 2012) fit this bill, since Pliny was launched precisely to explore how computing could facilitate exactly this traditional scholarly practice. Pliny tried to be "Englebartian"— referring to Douglas Englebart's H-LAM/T paradigm (Englebart 1962) that successful software integrates with the human way of doing intellectual things so well as to almost disappear, and that this disappearing software can, paradoxically, sometimes allow its users to do entirely new things that they had been previously incapable of doing. Out of this work came two models: the interface which developed a particular view of how users might usefully interact with a note-taking and note management tool to help them develop their own interpretation, and the data model that stored the information. Bradley 2008 describes Pliny's user interface in terms of affordances: 2-dimensional space, containment and hierarchy, naming and labelling and multiple reference of notes material in different contexts, including typing of a reference.

It turns out that Pliny's data model, as well as being designed to represent aspects of traditional scholarship in its three phases, is strongly suggestive of RDF and broader ontological technologies. Like RDF, the structure is a network and the links between the network nodes can be typed in a way similar to a RDF predicate. Pliny from its first release had the ability to export its structure into a Topic Web format, and some preliminary work has been done (see Jackson 2010) to map Pliny data into RDF through the OAC ontology. Further work has been carried out by us to take Jackson's approach further and better map an interpretation as stored in Pliny into an RDF representation.

The resulting paradigm is one that, unlike direct semantic annotation (as in figure I), separates the annotation of the domain literature from the highly formal world of shared domain ontologies by injecting a personal interpretative component in-between. One introduces, in ways compatible with semantic web technologies, a personal, more informal, and emerging representation of the scholarship into the picture.

Figure 3 is similar to the direct semantic annotation model shown in figure 1, but adds a structured area, representing aspects of the personal interpretative work of an individual, between a scholar's reading material to the left, and any shared public domain ontologies and linked data to the top right. This interjected personal interpretation "cloud" might well never be as clear-cut as formal ontologies must be, but its presence here recognises and enables the process towards formality that is a central part of interpretation in humanities scholarship. By interposing this somewhat-informal semantic "cloud" between the texts and the formal ontologies of the semantic web, we see a way of thinking about this central personal interpretive work that fits with the larger, more formal, semantic web picture. Although it would seem that the nature of traditional humanities research does not suit the standard direct semantic annotation model currently active in parts of the Life Sciences, we propose here an approach that, over time, encourages the researcher to turn their clouds of personal interpretation into material that might become more and more compatible with computer ontologies and the semantic web.

This presentation will describe work that was first shown in a preliminary fashion in the NeDiMaH Ontology workshop at DH2012 (Bradley and Pasin 2012), but that has continued since then and reaching a significant stage of development.

Along with open data about sources and places, open data about historical people provides one of the three legs for an open framework for historical information, since it is one of the three most obvious "entry points" into a potentially shared world of historical data.

DDH has been involved in a number of prosopographical projects (PBE, PBW, PASE, CCEd, PoMS, BoB1 and most recently the Making Charlemagne's Europe project), and we have become convinced that prosopography is particularly well suited to the world of open and linked data, since its purpose is to establish unique digital identities for people that could be used by other researchers. Furthermore, although we here at DDH have done quite a bit of work on structured prosopography we have found that "one size does not fit all" when it comes to structuring the prosopographical materials. Several of our projects have, at their core, a common approach we are calling "factoid prosopography". However, several of them (Clergy Church of England, Early Modern London Theatres) 2 do not. Finally, of course, many other people take on prosopographical work in their research without taking up a model of their data like ours: one thinks of the "personography" model work within the TEI, for example, and we know of other colleagues with prosopographical components that have developed a different model for their data.

The issue of prosopography as a open, shared resource goes beyond the provision merely of an identity for historical persons. In the same way that CIDOC-CRM (CIDOC-CRM 2011) was developed to record information about cultural heritage objects and therefore not only help uniquely to identify them but also to allow different systems to share a much broader range of data about these objects, a prosopography for historical persons would work to find common ground among the different ways in which different projects collect data about historical persons so that semantic links between them could more effectively be exploited. Indeed, it would seem that much of such a prosopographical ontology would usefully be based on CIDOC-CRM, since there is often significant overlap between the two domains: prosopography often comes into the work of people working with heritage objects, and there are thus substantial elements of prosopography in CIDOC-CRM. We have done some exploring of this idea in a recently created paper by Michele Pasin and myself (Pasin and Bradley, 2012) that takes the "factoid" in our prosopographical model and explores how it might be represented primarily in CIDOC-CRM terms.

We propose calling this shared ontological model for prosopography an Ontology for Historical Persons. By the very nature of such a thing, it would need to be developed out of as broad a shared perspective on the different ways that prosopographical data is being collected as possible. As an early step in thinking about this ontology, we have already been in touch with not only our project partners for the prosopographies we have already done, but also with a range of colleagues who have worked on the question of structured prosopography outside of our own projects, and with colleagues who have worked on the development of CIDOC-CRM. In our talk we would also describe the current state of our ontology, and explain how we think it would link with CIDOC-CRM. However, we'd be most interested in hearing of other people who are working on structured prosopography, and who would be interested in working with us towards enabling the open data potential that an ontology for historical persons might provide.

The “preliminaries” section of a 17th-century book encompasses the pages appearing in the printed text before the beginning of the work itself. This information is divided into seven different types of documents: details of publication, documentation of censorship (both civil and ecclesiastical), licensing, selling price, dedications, letters, and errors. The importance of the preliminaries for this project lies in the information present in these sections: the names of the officials signing the documents, their governmental/institutional affiliation, dates, place of issue, and literary circles that appear in the form of dedications and poetry written by various authors and published in their friend’s or associate’s books. In a few pages, the preliminaries give a complete image of the formal process required for the publication of each work of literature. By compiling all this information into a graph database and performing queries specific to various research questions, we have at hand a valuable source of information about the historical networks that influenced the publication of Early Modern Spanish literature.

To get a comprehensive look at this information, we generated lists of every edition of what we consider literary texts (fiction in prose, theatre, poetry, chronicles) published during the 17th -century in the Spanish empire (Jiménez et al. 1980)(Calvo et al. 2003). As shown by the following screen shot, we have focused on acquiring every available edition of each literary work.

We then divided the 17th Century into periods corresponding to the different “validos” —royal favorites that served as head of government or “prime ministers” — of the various kings in order to address the changing power structures of the time and their influence in literary production (Hernán et al. 2002). Through interlibrary loans and, in some cases, trips to the libraries that hold the edition, we acquired copies of the pages of each book that make up the preliminaries section. Then, we manually built a graph database using sylvadb.com, an open source software and free graph database management service developed in the CulturePlex Lab. Within Sylva, data was stored and organized using a custom designed system of schemas based on a node/edge relationship system. Finally, we exported the database to Gephi (https://gephi.org/), a software package that allows for visualization and statistical/metric analysis of the network using built-in algorithms and Python based scripting (Bastian et al. 2009). This allows us to detect important communities within the network, key players, important objects, and hubs of production.

For this study, we have unearthed the social networks of publishing and literary creation in 17th-century Spanish literature, focusing particularly on the period during the rule of the Duke of Lerma (1598-1618). Currently the first of our editions lists (Duke of Lerma) consists of 330 editions, out of which we have successfully obtained 228 scanned copies of preliminaries sections: approximately 70% of the total number. Of these scans, 121 have been entered into the database, producing a graph with 1612 nodes and 3472 relationships. Rendered in Gephi using the built-in OpenOrd algorithm, the graph looks like this:

Using the algorithms, metric analysis tools, and filters built into Gephi we pinpointed the individuals, governmental and ecclesiastical bodies that influenced publication in this period. Also, by using the concept of “ego network” from social network analysis, we established what we call the “publication network” of some of the authors that interest us (Carrington et al. 2011). A publication network includes the editors, censors, and other individuals important in the formal process of publication, as well as any other individuals that are more directly connected to the author: friends, family, patrons, literary colleagues, etc. We determined the range of the publication network based on the internal data structure of the Preliminaries database as follows. Due to bibliographic concerns (Bowers et al. 1962) and organizational aspects of our data schema, in order to establish a connection between the author and those involved in the approval, licensing, and publication of an edition there are four steps e.g., Author->Work, Work->Edition, Edition->Approval, Approval->Censor. Therefore, to establish an author’s publication network we needed to find neighbors for up to four degrees of separation. Although Gephi does not include ego network filters that extend to four degrees, using its Python based scripting console we were able to code functions that allowed us to isolate subsets within the graph to generate ego networks for any node to n degrees of separation. For instance, in the graph below we can see the publication networks of two authors associated with Mexico; Bernardo de Balbuena, author of Grandeza Mexicana; Juan de Torquemada, author of Monarquía Indiana; and the intersecting nodes in their publication networks:

Using the above techniques, we set out to find and isolate the main nodes of this social network that made possible the creation and sustainability of a transatlantic network of cultural agents. The first thing that stands out in the graph is Lope de Vega and his powerful, Madrid based publication network (Martínez et al. 2011). Using the Python scripting console, we determined that Lope’s publication network consists of 1083 nodes, or 67% of the nodes in the graph. This information is not new, based on the extremely prolific nature of his literary production we can assume that he was very well connected. However, we can also determine who wasn’t in his publication network. Departing from Lope’s publication network, we were able to locate the successful political and institutional connections that help us explain the central position of institutions such as the House of Zúñiga in the cultural fabric of the period.

To do this we used the scripting console to remove the subset of nodes representing Lope’s publication network from the other nodes that make up the graph, and returned a list of the names of all of the people who are not in Lope’s publication network. A quick review of this list produced some interesting results: we found several authors based in Spain including Gonzalo de Céspedes y Meneses and El Inca Garcilaso; and two authors active in Peru, Diego Dávalos y Figueroa and Pedro de Oña. While a quick look at both Céspedes and El Inca produced interesting results, the two Lima based authors attracted our attention. In this period social circles were highly influenced by geography, and it is logical that these authors find themselves at the periphery of a network centered geographically in Madrid. However, despite geographic concerns both authors remain connected to Lope de Vega’s network. We found that both Oña and Dávalos y Figueroa are connected to Lope’s network at 3 degrees of separation through their dedications to the Viceroy of Peru, Luis de Velasco y Castilla; and at four degrees through Juan de Zúñiga, Diego de Ojeda, and the Order of Santiago:

In order to contextualize the Peruvian network we compared the aforementioned “Mexican” authors with the “Peruvian” authors. Combining the four social networks into two based on geographic constraints, we found that at 4 degrees of separation there was no direct overlap, so we upped the parameter to 5 degrees of separation and produced the following image:

As shown here, even at five degrees of separation there are few overlaps between the networks. However, in the above image we begin to notice the importance of the House of Zúñiga. It is well known that the House of Zúñiga was powerful in both Spain and the Americas, and also that certain members of this house were important patrons of the arts and literature (Cátedra 2003; Díez Fernández et al. 2005). Nonetheless, we don’t think that their role in transatlantic literary production has been adequately explored. The political importance of this family in New Spain is obviously important (an Archbishop and a Viceroy); however, the Preliminaries graph illustrates not only the political role this house played in America, but also the importance of political figures/nobility in publication circles and how the members of one house can spread their cultural influence throughout geographic space. To take this concept one step further, we followed the Zúñigas back the Spain. Here we find the Duke of Béjar, Alonso López de Zúñiga y Pérez de Guzmán, and the first part of Don Quixote. It turns out that American authors were not the only artists soliciting support from the House of Zúñiga: Miguel de Cervantes dedicated part 1 of Don Quixote to the famous Duke of Béjar(Rico 2005).

The above samples show the potential of a research model that combines network-based analysis with quantitative and qualitative studies of cultural production, providing evidence of the interaction between political structures and cultural production in the Spanish Empire (Martínez et al 2008). By repurposing bibliographic data, the Preliminaries Project allows us to explore the concept of cultural networks within the framework of transatlantic studies and complexity theory (Wood 2010; Suárez 2007). Furthermore, this study demonstrates the effectiveness of digital humanities methods as a tool to locate previously overlooked areas for further study using a more traditional humanistic approach.

‘If it is drawn it must be true.’ Most people have the tendency to treat visualized data as facts and are much less prone to question visualizations than written words. Incorporating ways of conveying the notions of uncertainty and imprecision within visualizations is therefore a critical issue. This presentation will address this problem and consider what possible steps can be taken to visualize uncertainty and imprecision within the constraints of diagrams whose shapes and appearance are necessarily dictated by those of the objects that they represent.

Amongst the disciplines devoted to the study of books, bookbinding history is young, lacking a well-established vocabulary and a stable description system (Szirmai 1999). Bookbinding descriptions are often inaccurate, and difficult to communicate especially to international audiences.

Verbal descriptions are inadequate and are usually accompanied by drawings, which do not comply with any standard. Different authors employ different styles and conventions, and sometimes a number of conventions are used within one publication.

Recent work at the Ligatus Research Centre addresses the bookbinding vocabulary and classification problem. We have developed a descriptive schema and glossary for bookbinding structures utilizing eXtensible Markup Language (XML) technologies. This has been used to survey the bookbinding structures of the books from the St. Catherine’s Monastery Library, Mount Sinai, Egypt. The records are combined with freehand drawings of the binding structures (Velios 2008; Velios and Pickwoad 2005). More recently, we have been working towards a methodology to automatically transform the XML descriptions of bookbinding structures into Scalable Vector Graphics (SVG) diagrams. The advantages of the automated transformations are: (i) standardized output; (ii) production speed as they save significant time during the survey, and (iii) better accuracy as they function as verification of the surveyed data during the survey. However, the data produced during the survey is often uncertain (when features are obscured), imprecise (when surveyors have limited capacity to interpret evidence), and incomplete (when the available time is limited). Uncertain data is so flagged within the XML records, and also the imprecision and the absence of certain elements can be inferred from the records. Therefore, it is possible and appropriate to communicate imprecise and uncertain data within the visualizations. These should be drawn clearly and without affecting the overall meaning of the diagram.

When dealing with historical bookbinding structures, one has to accept that a degree of uncertainty, imprecision and incompleteness is inevitable for the following reasons.

As in the case of quantitative and geographical data visualizations (Wilkinson 2005; Wainer 2009; Gethin Powell 2012), factors of uncertainty can be found in (i) inherent problems with the definition of the object of study, i.e. the limitations of current knowledge regarding bookbinding structures and their description; (ii) the sources of information and their interpretation, i.e. the books being described and the interpretation of their binding structures; (iii) issues with the categorization and representation of certain complex features, i.e. undetermined complex shapes for highly decorative and non-functional elements. Uncertainty can arise in all of these areas. One of the main problems faced by bookbinding historians lays in the young age of the discipline. Goldschmidt, in 1928, pointed out that ‘far fewer people [could] give a reasoned opinion on the country of origin and the approximate date of an old bookbinding, than a piece of pottery or furniture’ (Goldschmidt 1928). Unfortunately, almost a century later, the situation has not changed and often bookbinding descriptions are still inaccurate. Moreover, a book is sometimes too damaged to show clear evidence of its origin, or else, elements of the binding, especially decorative ones like metal furniture, are so complex that it is difficult to categorize and describe them.

The automated visualizations of binding structures, based upon their XML verbal descriptions, do not aim at a naturalistic representation of the item being described. But rather at conveying its general form in a highly prototypified representation, thus making it possible and easier for it to be compared with similar structures found in other books. Similarity of shapes depends on the preservation of certain particular features, referred to as prägnant features, e.g. symmetry, orientation in space, etc., while non-prägnant features, e.g. precise measurements, can be substantially changed without psychologically affecting the overall impression for the observer (Goldmeier 1972). Precise measurements are therefore not essential for the prototypification of shape, and for this reason only few precise measurements are required from the bookbinding record. Nonetheless, we feel that imprecisions in measurement-dependent elements should still be identifiable because they indicate scale, which may be a critical aspect of the description.

When inaccuracy is objectively determinable, it can be expressed as error (MacEachren, et al. 2005). As it is often the case with human generated data, input problems and errors do occur. However, these are not easily identifiable through automated means and, considering that our transformations are also thought of as a possible step in ensuring data correctness by part of the surveyors, ambiguities due to human error are not being considered here.

Uncertainty and its representation have been studied in the scientific and geographical visualization communities (Wilkinson 2005). Bertin's (Bertin 1981; Bertin 1983) basic graphic variables — location/position, size, value, texture, colour, orientation, shape/form — with the addition of colour saturation, transparency/opacity, and sharpness/blurring (MacEachren 1992) have been used to build graphic representations and depict uncertainty. Interestingly, all these graphical features correspond to the basic feature channels in our primary visual cortex (V1), being thus perceptually distinct features (Ware 2012).

Cultural heritage visualizations are constrained by the shapes, spatial relationships, and relative sizes of the objects being portrayed. Therefore, only a limited set of graphic variables can be used to depict uncertainty. While uncertainty depiction needs to be perceptually salient, understanding the overall meaning of the diagram should not be precluded. Transparency has been used in archaeological reconstructions (Zuk and Carpendale 2006); however, lacking a distinct background, the use of transparency on its own can be problematic. Further, automated visualizations need a flexible system to apply a higher or lower degree of uncertainty to its elements.

Out of the possible graphic variables, we have identified blurring as the best option to express uncertainty in our diagrams. People can easily interpret less sharp details as uncertainty (MacEachren 1992), while blurring can be applied to single lines without affecting the diagram's overall appearance. SVG provides a Gaussian-blur filter applicable to any shape and allows for different standard deviation values for the x and y axes. This grants the preservation of the overall shape of uncertain elements (see fig. 1) and permits to visualize different degrees of uncertainty through a flexible and universally applicable system. Uncertain elements are drawn in the configuration deemed most probable, while the degree of blurring can be determined by the number of possible variations.

Once identified, uncertainty can be resolved by providing the user with a set of 'small multiples' (Tufte 1990), i.e. alternative possible visualizations, allowing for immediate comparison within the scope of the eye span (see fig. 2).

To indicate imprecision we need a graphical variable that could make the element identifiable without affecting the overall outline of the diagram, thus allowing for uninterrupted contour realization by part of the user. While colour labelling could be an option, we need a V1 feature channel that could make the element perceptually distinct leaving the contour unaffected. We find that this could be achieved with a surround colour (see fig. 3).

A coloured halo is a perceptual distinction feature (Ware 2012) that can be universally applied to graphics through SVG filters. The reduced contrast with the background makes the feature perceptually less salient, while the shape contour is left unaltered.

Visualizations within scholarly research projects should convey uncertainty and imprecision where needed. Depicting uncertainty has been the object of study of many scholars, however, its application to diagrams whose shapes are dictated by those of the objects being represented, like in the case of cultural heritage objects, poses particular issues and constraints. Even more so in the case of automatically generated diagrams within a large-scale project as our bookbinding structure visualizations.

This presentation is intended to serve as an example of the kind of perceptually salient graphical variables that can be used to successfully depict uncertainty and imprecision within cultural heritage prototypical diagrams.

Introduction: This paper presents the results of the Modernist Versions Project’s (MVP) survey of existing tools for digital collation, comparison, and versioning. The MVP’s primary mission is to enable interpretations of modernist texts that are difficult without computational approaches. We understand versioning as the process through which scholars examine multiple witnesses of a text in order to gain critical insight into its creation and transmission1 and then make those witnesses available for critical engagement with the work. Collation is the act of identifying different versions of a text and noting changes between texts. Versioning is the editorial practice of presenting a critical apparatus for those changes. To this end, the MVP requires tools that: (1) identify variants in TXT and XML files, (2) export those results in a format or formats conducive to visualization, (3) visualize them in ways that allow readers to identify critically meaningful variations, and (4) aid in the visual presentation of versions.

The MVP surveyed and assessed an array of tools created specifically for aiding scholars in collating texts, versioning them, and visualizing changes between them. These tools include: (1) JuxtaCommons, (2) DV Coll, (3) TEI Comparator, (4) Text::TEI::Collate, (5) Collate2, (6) TUSTEP, (7) TXSTEP, (8) CollateX, (9) SimpleTCT, (10) Versioning Machine, and (11) HRIT Tools (nMerge). We also examined version control systems such as Git and Subversion in order to better understand how they might inform our understanding of collation in textual scholarship. This paper presents the methodologies of the survey and assessment as well as the MVP’s initial findings.

Problem: Part of the MVP’s mandate is to find new ways of harnessing computers to find differences between witnesses and to then identify the differences that make a difference (Bateson). In modernist studies, the most famous example of computer-assisted collation is Hans Walter Gabler’s use of the Tübingen System of Text Processing tools (TUSTEP) to collate and print James Joyce’s Ulysses in the 1970s and 1980s. Yet some constraints, such as those identified by Wilhelm Ott in 2000, still remain in the field of textual scholarship, especially where collation and versioning applications are concerned. Ott writes, “scholars whose profession is not computing but a more or less specialized field in the humanities have to be provided with tools that allow them to solve any problems that occur without having to write programs themselves. This leads to a concept of relatively independent programs for the elementary functions of text handling” (97). Indeed the number of programs available for collation work have proliferated since 2000, including additions to TUSTEP (TXSTEP) as well as the newest web-based collation program, JuxtaCommons.

Accordingly, the MVP has reviewed tools currently available for collation work in order to provide an overview of the field and to identify software that might be further developed in order to create a collating, versioning, and visualization environment. Most of these tools were developed for specific projects, and thus do what they were designed to do quite well. Our question is whether we can modify existing tools to fit the needs of our project or whether a suite of collation and visualization tools needs to be developed from scratch. This survey is thus an attempt to chart the tools that may be useful for the kinds of collation and versioning workflows our team is developing specifically for modernist studies, so we can then test methods based on previous tools and envision future developments to meet emerging needs. Our initial research with Versioning Machine and JuxtaCommons suggests that there is potential for bringing tools together to create a more robust versioning system. Tools such as the Versioning Machine work well if one is working with TEI P5 documents; however, we are equally interested in developing workflows that do not rely upon the TEI, or do not require substantive markup. Finally, we are examining whether the development of version control systems such as Git present viable alternatives to versioning methods now prevalent in textual studies.

Method: Our method adapts the rubric Hans Walter Gabler devised for surveying collation and versioning tools in his 2008 white paper, “Remarks on Collation.” We first assessed the code and algorithms underlying each tool on our list, and we then tested each tool using a literary text. In this particular case, we used two text files and two TEI XML files from chapter three of Joseph Conrad’s Nostromo, which we have in OCR-corrected and TEI-Lite marked-up states from the 1904 serial edition and the 1904 first book edition. During each test, we used a tool assessment rubric (available upon request) to maintain consistent results across each instance. All tests were accompanied by research logs for additional commentary and observations made by our research team.

Preliminary Findings: Our preliminary findings suggest that: Many existing collation tools are anchored in obsolete technologies (e.g., TUSTEP, which was originally written in Fortran, despite having undergone major upgrades, still relies on its own scripting language and front end to operate; also, DV-Coll was written for DOS, but has been updated for use with Windows 7). Many of the tools present accessibility obstacles because they are desktop-only entities, making large-scale collaborative work on shared materials difficult and prone to duplication and/or loss of work. Of the tools that offer web-based options, JuxtaCommons is the most robust. The “commons” approach to scholarly collaboration is among the most promising direction for future development. We suggest the metaphor of the commons is useful for tool development in versioning and collation as well as for building scholarly community (e.g., MLA Commons). We note the particular usefulness in this regard of the Juxta Commons collation tool and the Modernist Commons environment for preparing digital texts for processing. The latter, under development by Editing Modernism in Canada, is currently working to integrate collation and versioning functions into its environment. Version control alternatives to traditional textual studies-based versioning and visualization presents an exciting set of possibilities. Although the use of Git, Github, and Gist for collating, versioning, and visualizing literary texts has not gained much traction, we see great potential in this line of inquiry. Developers and projects should have APIs in mind when designing tools for agility and robustness across time. Web-based frameworks allow for this type of collaborative development, and we are pleased to see that Juxta has released a web service API for its users. During tool development, greater attention must be given to extensibility, interoperability, and flexibility of functionality. Because many projects are purposebuilt, they are often difficult to adapt to non-native corpora and divergent workflows.

Renear (2000) introduces the term "transcriptional markup" with a loose sense of pertaining to the reproduction of an existing text, and he does not attempt to define transcription per se. More recently Huitfeldt and Sperberg-McQueen (2008), and Huitfeldt, Marcoux, and Sperberg-McQueen (2009, 2010) have developed a model of transcription (hereafter the HMS model) that they believe "provides a sort of greatest common denominator for markup systems" (2010, 15). But an awkward gap exists between the HMS model's abstract components and the realities of, for example, the TEI markup language and its typical usage. Here I make the case for a kind of markup — perfectly possible though in practice improbable — which bridges that gap. I call it pure transcriptional markup because it refines the sense of Renear's term and grounds it in an actual, formally specified model of transcription.1

Transcription and text encoding are clearly connected, especially where markup replaces the work of presentation.2 However, not everything that typical scholarly encoding practice might assert regarding the logical domain of a pre-existing text can be considered proper to transcription (if "transcription" is to have any specific meaning of its own). In the initial HMS model Huitfeldt and Sperberg-McQueen (2008) avoid the trap of trying to decide what visible details of an exemplar document E should be reproduced in a transcription T by asking instead what overall relation must hold between the two such that document T is a successful transcription of document E (expressed as t_similar). They say that under a set of reading conditions R, marks of E can be seen as a sequence of tokens each instantiating a type. So, abstractly, a document E is a sequence of types, and if under the same set of reading conditions R a document T can be seen as a token sequence whose corresponding type sequence matches E's type sequence, then E and T are t_similar.

The initial HMS model considers a simple sequence of basic, indivisible tokens at grapheme level; later versions (2009, 2010) introduce components that allow the modeling of a complex structure of token groups (and related types). These groups, known as compound tokens, may have as constituents either compound tokens or basic (atomic) tokens. The resulting view of a document comprising a structured, multi-level token sequence immediately suggests similarities with the well-known 'ordered hierarchy of content objects' (OHCO). As noted above, its creators explicitly connect the HMS model to markup, saying some aspects "serve purposes analogous to the generic identifiers and attribute-value pairs of SGML and related markup languages" (2010, 11), that "element types are types", and "element instances are tokens" (2010, 15).

However, some features of common encoding practice are problematic for the model. I briefly outline two of them here; a single example will illustrate both.3 Suppose an original printed text contains the sentence "Joe stinks!" Describing this sentence in terms of the HMS model we have:

A typical TEI encoding of our example might be: