User Required? On the Value of User Research in the Digital Humanities

User Required?On the Value of User Research in

the Digital Humanities

Max KemmanUniversity of Luxembourg

@MaxKemman

Martijn KleppeErasmus University

Rotterdam

@MartijnKleppe

The problem with tools

• Digital Humanities ♥ tools• But tools not always used

• Two approaches• Just publish the data• User-centred design

User-Centred DesignWhat is the added value of user research for developing tools aimed at digital research methods?Two problems:• Does the user know what she wants/needs?

People don’t know what they want until you show it to them

User-Centred DesignWhat is the added value of user research for developing tools aimed at digital research methods?Two problems:• Does the user know what she wants/needs?

First Rule of Usability? Don't Listen to Users

User-Centred DesignWhat is the added value of user research for developing tools aimed at digital research methods?Two problems:• Does the user know what she wants/needs?

The technology will come first, the products second, and then the needs will slowly appear

User-Centred DesignWhat is the added value of user research for developing tools aimed at digital research methods?Two problems:• Does the user know what she wants/needs?• Can user requirements be generalized? basic functions common to

scholarly activity across disciplines, over time, and independent of theoretical

orientation

User-Centred DesignWhat is the added value of user research for developing tools aimed at digital research methods?Two problems:• Does the user know what she wants/needs?• Can user requirements be generalized? This model of what researchers do

maps quite naturally onto a model of an infrastructure that supports these

research processes

User-Centred DesignWhat is the added value of user research for developing tools aimed at digital research methods?Two problems:• Does the user know what she wants/needs?• Can user requirements be generalized?

A 'one size fits all' approach would be a disastrous underestimation of the

specific needs of humanities research

User-Centred DesignWhat is the added value of user research for developing tools aimed at digital research methods?Two problems:• Does the user know what she wants/needs?• Can user requirements be generalized?

We will review user research of two case studies we conducted

Case studies

PoliMedia• Automatically creating links between

debates of the Dutch parliament to media items, made available in a search user interface in which debates of the Dutch parliament can be explored• www.polimedia.nl

PoliMedia - Search

PoliMedia – Debate page

PoliMedia – Link to Delpher

Case studies

Oral History Today• A search user interface similar to Google

but including advanced filter options, in which oral history interviews and collections can be searched and explored to discover topics across a multitude of collections• zoeken.verteldverleden.org

Oral History Today – Search

Oral History Today – Word cloud

Oral History Today – Visual facets

Oral History Today – Interview

Method

PoliMedia• User requirements elicitation: interviews

with 5 scholars (F2F)• Evaluation: eye tracking with 24 scholarsOral History Today• User requirements elicitation: interviews

with 15 scholars (Skype)• Evaluation: walkthrough with 5 scholars

per update (Skype)• Scrumming

Method

Processing• Categorizing statements• Feasibility discussion with developers

Results – PoliMedia

39 user requirements• 21 deemed within-scope• 18 deemed out-of-scope• 27 unique user requirementsMost common (3/5): • Media output from before the debate• Names of people involved• Location in the newspaper

Results – PoliMedia

Requirements with impact• Function of actors (e.g. minister,

member of parliament, but also show host, interviewer, etc.) (1)• Party of actors (e.g. VVD, PvdA, but

also Greenpeace or other lobby groups) (2)• Media output about subject before debate (3)

Results – Oral History Today

75 user requirements• 33 deemed within-scope• 42 deemed out-of-scope• 34 unique user requirements• Most common (10/15):• Filter for time period

Results – Oral History Today

Requirements with impact• Description of project (within which collection was created) and how collection came to be• Organization behind collection (management/creation)• Distinguishing search filters between content-related to general

Discussion

• Scholars are aware of what they want• But also limitations:

• Generalizing is difficult

Requirements

Project goals

Current practices

Discussion

• Within-scope requirements lead to usable tools• Out-of-scope requirements give hints of

wider research workflow & compatibility• Tool takes place in a wider research

workflow

DevelopingFor – With – By

Humanities

• User research!

• Support workflow demanding multiple tools• Separate tool and data

Conclusion

X

User Required!Thank you for listening

www.slideshare.net/maxkemman