Tools and Data

This section uses the (inter)action system as a framework for identifying and describing the tool used, and data produced, by #DataCreatives. The tables below are a record of specific tools and data produced. For a more high level overview of how the (inter)action system can be applied to the understanding of #DataCreatives actions, see Appendix A.

Jump to 1. Communication, collaboration and process 2. Data analysis and visualiation 3. Sharing research and outputs

Communication, collaboration and process

Tool Usage Data Alternatives
Zoom We are holding fortnightly team meetings using Zoom, alternating with fortnightly discussion meetings. Smaller sub-groups of team members also meet via Zoom as needed. The use of videoconferencing technology for meetings was necessitated by COVID-19, but has also allowed us to record meetings, both for the benefit of those who are unable to attend, for later analysis, and to incorporate into research outputs. We share screens within Zoom in order to demonstrate visualisations and work through data structures. We share links in the Zoom chat, both to our own documents, and to potential data sources, examples of visualisations, and useful tools. The Zoom white board and annotation features have provided us with a mechanism for conducting visual retrospectives to reflect on our progress to date and brainstorm where to go next. Affordances: talking in real time, sharing screen (showing), chat. Audio and video recordings; Transactional time series data (call history); Chat text Teams, Skype
Microsoft Teams Outside of meetings, we communicate with one another through chat in Teams. We have a collaboration Teams site, with a number of different channels for things like links and resources and upskilling. Most the chat takes place in the general channel. The Teams site also services as the repository for many of the documents, data files, and research records associated with the collaboration. In the Files section of Teams we have a folder structure for storing all of the collaboration documentation. The files stored here are living documents and can be collaboratively edited by all members of the team. They are often created on the fly, shared, and edited during Zoom meetings. The downside to Teams is that there is little transparency over how it displays certain file types (eg. whether it plays animated gifs posted in chat) and its tendency to automatically rename plain text files to remove the .txt file extension. It is also not clear how or whether it is possible to export chat text from Teams. We’ve used Team’s chat for data exploration, with quick data visualisations posted for discussion that leads to questions that lead further data visualisations. Affordances: channels, posts, comments, upload files, reactions Number of posts; Number of posts in channels; Number of comments to posts (through all modes, vid, text etc); Kinds of files uploaded (vid, text, etc); Like buttons Slack channels has features for keeping track of different conversation threads but without the integration into the MS tools it would have been harder to collaborate.
Microsoft SharePoint Microsoft SharePoint integrates with the Teams site we use for collaboration and is used for file storage. Files stored in SharePoint can be edited using versions of Microsoft products that integrate with the Teams Desktop application or with the browser or can be edited in the authoring application. Affordances: file upload, file structure, collaborative writing, version control, track changes Audit log; Files; Folder structure Share drive, Mediaflux
Microsoft Word Notes of team meetings and drafts of documents have been written in Microsoft Word, stored in SharePoint, and edited collaboratively through Teams. Text (formatted) In some cases, for documents that are intended to become research outputs that will be shared openly, the team considered using markdown with Git to author the document, but ended up opting for Word, primarily out of familiarity.
Microsoft Power Point
Microsoft Outlook Affordances: emails, calendar invites. Calendars – meeting attendance; Communication
Microsoft Excel Excel was chosen for collecting the metadata about the surveys of interest. We decided on the data dictionary as a team, which we documented in the second sheet, and this data dictionary formed the column headings of the first sheet, which is where the survey metadata was entered. We could have used survey tools such as Qualtrics or Google Forms, connected to Google Sheets, to gather this data, but it was felt that it would take too much effort to set these up, and that Excel was located with and fit the working practices we already had in place for this collaboration, as well as being a familiar tool.
GitLab GitLab Issues are being used to manage tasks and activities. Commit history (git log) GitHub. We chose to use GitLab because it is hosted and managed by the University of Melbourne, whereas GitHub is hosted on public servers that are not located in Australia.

Data analysis and visualisation

Tool Usage Data Alternatives
Voyant We are using Voyant (https://voyant.tinker.edu.au/) for exploratory visualisations of text corpora with a graphical user interface and no coding required. This is an instance of the open source Voyant Tools environment (https://voyant-tools.org/) that is locally hosted on Australian servers. Python and R give control over where the data is hosted, a more private, but slower, option for visualising data.
Jupyter Notebooks Some data exploration is using python in a Jupyter notebook. This allows us to meet our obligation to protect the sensitive data that is being shared with the project team. R has similar capabilities for data exploration but we don’t have the same familiarity with the language.
Unity Unity is a 3D development platform and game engine that we are using to create an environment in which to develop 3D data visualisations. These will be then be transferred over into Mozilla Hubs for presentation and exploration.

Sharing research and outputs

Tool Usage Data Alternatives
Omeka An Omeka site (https://omeka.cloud.unimelb.edu.au/datacreative/) is being used to showcase research data and outputs as an online gallery. We intend to integrate Omeka with figshare, so that research outputs that are hosted in figshare can be displayed as part of the Omeka gallery.
Figshare Research outputs such as curated datasets and data visualisations will be deposited in the University of Melbourne’s institutional figshare repository (https://melbourne.figshare.com/). This enables them to receive a DOI so that they are citable, and an embed code, so that they can be embedded into an Omeka gallery or a GitLab Page. Affordances: upload files (data, images, etc), mint a DOI. Views; Downloads; AltMetrics Zenodo, OSF, Dryad, figshare.com, etc. We chose to use the data repository that is hosted and supported by the University. A member of the collaboration team is a systems administrator for figshare and can assist with working out how to use it most effectively, including raising support tickets with the vendors and thinking about integrations with other platforms, such as Omeka.
GitLab GitLab Pages are being used to display static data visualisations. Commit history (git log) GitHub. We chose to use GitLab because it is hosted and managed by the University of Melbourne, whereas GitHub is hosted on public servers that are not located in Australia.
Wordpress Selected members of the collaboration have access to post on the SoTEL blog. This is a Wordpress blog, hosted on the University of Melbourne website (https://blogs.unimelb.edu.au/sotel/datacreativities/datacreativities-about/). Blogger, GitLab Pages. WordPress is familiar and easy to use, is supported by the University, and was already in use by one member of the research collaboration for another project, so it was relatively easy to start using it for this collaboration too.
Twitter Individual team members who have Twitter accounts tweet about their research, both while it is in progress and when research outputs are shared publicly, using the hashtag #DataCreatives. This allows members of the research collaboration to connect with the broader research and creative arts community, discuss ideas as they are developing, and generate data feedback loops, as well as feeding the results of the research back to the wider research and arts communities. Affordances: hashtag, retweets, comments. Number of retweets; Text from comments; Location of tweets (kind of) Mastodon. The creative and research audience for this collaboration use the Twitter platform.
Mozilla Hubs We are planning to use this as a space for sharing and exploring the 3D data visualisations created in Unity.