From modelling to publishing:

Digital scholarship in practice

Tiziana Mancinelli

VeDPH

@tizmancinelli

 

 

Text

We are great at creating digital content but lousy at managing it

Aaron D. Purcell (2019)

Le Digital humanities hanno costruito progetti scientifici e materiali digitali nati nel corso degli anni.

Dove sono questi progetti?

DATA in HUMANITIES

An historical document

DATA

Christof Schöch’s “Big? Smart? Clean? Messy? Data in the Humanities” http://journalofdigitalhumanities.org/2-3/big-smart-clean-messy-data-in-the-humanities/

Goals and objectives

  • What are the aspects to consider when you start a digital scholarly project?
  • Planning data management
  • What are the short and long term goals
  • Identify potential data management problems
  • risks and possibilities
  • Sustanibility

Goals and objectives

Hermeneutical practices are described as “Digital hermeneutics” when digital sources and computational methods are involved in the epistemological process. ( Daquino, Pasqual, Tomasi)

Practices involve methods.

Debates about method are ultimately debates about our models. (Julia Flanders, 2018)

Components of a digital scholarly project

DH

Digital scholarly projects:  research data-driven

  • Are complex objects
  • are based on traditional research
  • take more time than a traditional research project
  • DHers need data management best practices training

Data in Humanities

But... What are data in Humanities?

1. Data in humanities are digital but most of them are not.

2. humanities data are a semiotic systems (Marras, Ciulia, 2016 )

 

Methods and methodologies

The methodologies we use are mostly a byproduct of our praxis. In other words: we make do with what we have. This situation is partly a result of what can be termed as the inheritance model of digital humanities inquiry.

The inheritance model means that the tools of digital humanists used in the early stages of the field (and partly to this day) have mostly been created for purposes very different than the study of the past. There are, of course, exceptions to the tools ‘inherited’ from other disciplines.

This situation has led to the creation of a mixed methodology, which tries to use tools not necessarily made to perform a particular task. (Mateusz Fafinski, 2021)

Data management

  • planning, organizing, storing, and sharing data and other research results and products

Data management

  • planning, collecting materials
  • Organise those data will allow to better manage it and to collaborate (and sustainability!)

Data modeling

Data structures that correspond to our intuitions about the intellectual organization of the data


Modelling is a creative process to gain new knowledge about material and immaterial objects by generating representations of them. It is widely understood and used as a heuristic strategy in the sciences (Frigg and Hartmann 2012, Mahr 2009) as well as in digital humanities (DH) research where it is considered a core practice (McCarty 2005, 20–72).

Data modeling

 

  • data elements whose level of abstraction and granularity matches that of our domain analysis, and an overall architecture in which making the information connections required for analysis is simple and direct rather than requiring elaborate or extremely indirect traversals.

  • Data modeled in this way is likely to be easier to document, to explain, and to program for. (Flanders, Jonnadis, 2018)

 

KDL Checklist for Digital Outputs Assessment

DIGITAL CURATION & STANDARDS

For digital humanities data, a standard has been defined as “codified rules and guidelines for the creation, description and management of digital resources” (Gill and Miller, 2002). Standards can be classified as a de jure standard, which may be mandated by law (or may be used to designate a formal standard), or de facto standards, such as the Text Encoding Initiative, which enjoys widespread use and acceptance.

 STANDARDS

  • Official standard
    • Encoding: utf-8
    • Data modelling: TEI, XML, RDF, etc
    • Protocols: HTTP, TCP, etc

 

Standard de-facto:

  • Development frameworks
  • Database technologies

BIFLOW PROJECT

BIFLOW PROJECT

  • Architecture and design project:
    • API
      • Role: expose the data and the model via a REST API
      • Database: MySQL
      • Language: PHP (Api-Platform/Symphony)

BIFLOW PROJECT

  • Architecture and design project:
    • Front-end (design, usability)
    • Role: the website, as static pages, hosted on github
    • Language: Jeckyll
    • Language: Javavascript (to retrieve and send data to the API component)

BIFLOW PROJECT

  • Architecture and design project:
    • Admin
      • Role: data management, through the REST API
      • Language: javascript and REACT
    • RDF/Ontology
      • Role: expose the data in RDF format
      • Language: PHP, using the REST API to retrieve data

BIFLOW PROJECT

Values and Open Data

  • Open source software
  • Open access license

Collaborations and communities of practices

  • One the high risk in the maintainability of the project:
    • Which technologies have been chosen? Will it be supported, developed and maintained in the next 15-20 years?

Collaborations and communities of practices

  • Do we relay on 3rd-party components?
    • Will they be maintained for the next 20 years?
    • Even small components such as CSS frameworks, often hosted in CDN could be an issue in the future.

LONG-TERM

PRESERVATION ISSUES

 

“one no longer preserves tangible physical objects per se, but views abstract representations of such objects that can be reconstructed in an unpredictable technological future.”

    J.P. Chanod, Will Your Data Still Be Around Tomorrow?, 2013, http://bit.ly/2tMRP3c)

Research Infrastructures for the Humanities

The costs of digital scholarship are frequently posed in terms of human labor and rights abuses, ecological and environmental damage, carbon footprints, waste of resources associated with obsolescence, and other issues that affect every stage of
the lifecycle of production.

Research Infrastructures for the Humanities

But sustainability also needs to be understood as an epistemic concept embedded in complex systems, not merely as a set of problems to be solved through instrumental means applied to operational logistics.

Interdisciplinary and intersectionality

DH scholars need to imagine ways of mitigating the impact of inequalities on project development, communication, training, collaboration and engagement.

Conclusions

 

  • DH research is not valued in the DH researchers are asked to deal with too many tasks on the large spectrum of digital scholarly project
  • Communities of practices (DH centres, tools)
  •  

 

Bibliographic References

Elena Pierazzo, How subjective is your model? in Shape of data in Digital Humanities ed. Julia Flanders and Fotis Jannidis, London, Imprint Routledge, 2018

 

Circling around texts and language: towards “pragmatic modelling” in Digital Humanities - http://www.digitalhumanities.org/dhq/vol/10/3/000258/000258.html

Images credits

https://en.wikipedia.org/wiki/Data#/media/File:Data_types_-_en.svg

https://www.ibs.it/two-cultures-scientific-revolution-libro-inglese-c-p-snow/e/9781684115334

https://www.katinamichael.com/seminars/2020/7/30/progressing-from-humanities-to-digital-humanities

Made with Slides.com