RECOMMENDATIONS

PARTNERSHIPS

ENGAGING IN A PARTNERSHIP

It is important for collaborators to be on the same page about timeline and capacity. Ideally the partnership should begin during the project conception phase to ensure the required level of commitment and availability. The Linking Lost Jazz Shrines recommendation for others looking to engage in joint developmental activities is to jointly develop a project timeline once a collaboration is established and ensure that all involved parties are aware of and consent to the timeline. It is equally important to periodically update the timeline jointly and to promptly notify the other party when there are delays or necessary adjustments.

COLLABORATOR AGREEMENTS

The Linking Lost Jazz Shrines team also recommends a formal memorandum of agreement or understanding between collaborators that details the terms and conditions and the roles and responsibilities for each collaborator and individual team members, outlines the benefits of the collaboration for each collaborator, and discusses expectations for communication, publication, publicity, extension, etc.

RENEWING COMMUNITY RELATIONSHIPS

The Linking Lost Jazz Shrines team recommends renewing relationships with community stakeholders earlier on in the project or even in the planning/application stages as they are integral to how the project reaches the intended audiences and communities and may have insight into ways the project can be more impactful. Community stakeholder engagement can be written into the project timeline and stakeholder representatives could serve on an advisory board throughout the project.

PROGRAMMING AND OUTREACH

ENGAGING COLLABORATORS

Engage potential programming and educational collaborators and/or participants well in advance of the implementation of programs so that they have the opportunity to help inform the approach to programming and provide insight into how to best reach new audiences through effective marketing, advertising, and promotional activities.

PLANNING EDUCATIONAL ASPECTS

Conceptualize, plan and incorporate educational components of the project early on in order to ensure the development and timely implementation of outreach initiatives.

WORKING WITH EDUCATION OR INSTRUCTION STAFF

Familiarize the education department or instruction staff with the project so that they can articulate its goals and assist in engaging with educational institutions. The educational staff may also be able to create resources, curriculum, etc. that can be used by an instructor when workshops are not an option.

ADMINISTRATIVE FLUX AND ADAPTABILITY

CONTINGENCY PLANS

Ensure that there is at least one full-time staff member on the project team throughout the duration of a project. There should also be a contingency plan developed with leadership that addresses possible changes to project personnel and how project responsibilities would be properly absorbed or managed by appropriate permanent staff.

STAFF INVESTMENT

Incorporate staff training and project orientation into the project objectives and timeline. It is imperative that other staff members understand the importance of and become invested in the success of the project. It is just as necessary to execute agreements with staff members that outline their responsibilities to the project as it is to get partnership agreements. A formal written document will help to hold staff members accountable.

PROJECT TEAM ROLES

SEPARATING PROJECT COMPONENTS

The Linking Lost Jazz Shrines team’s recommendation is to have individuals or teams dedicated to the primary components of the projects: technical (data creation), archival (data sustainability and use), programming, and administrative. The tasks and responsibilities related to each component should be outlined in the application and/or in a post-acceptance agreement. We also recommend a shared notes document and a shared Drive folder to ensure that communication between all components is transparent and readily accessible by all team members.

TECHNICAL

DOCUMENT ALL DECISIONS

While there are times to expand the network of what can be described and how, it is important to keep the main thing the main thing, and not be too distracted by pressure to fully represent real life. The end goal should be a workable dataset that focuses on the most pertinent content to your purposes. There will inevitably be generalities in this process. It is important to be clear and transparent when this occurs, documenting why a certain decision was made. Referring back to these decision logs will help keep the work on track when subsequent variants of the project parameters and “what if” questions arise.

LOCATING AN ONTOLOGY

If you desire to create linked data out of oral histories or other textual documents, first research if there is an existing ontology that could be used to express your data. Linked Open Vocabularies (https://lov.linkeddata.es/dataset/lov/) is a good place to start, with schema.org, the Relationship Ontology (https://vocab.org/relationship/) , and Friend of a Friend (http://xmlns.com/foaf/spec/) being common “go to” ontologies. Depending on the type of project, it may also be necessary to seek out ontologies that are domain specific, for example, The Music Ontology (http://musicontology.com). An additional method for deciding which ontology or ontologies to use for a project can be to identify existing linked data projects that deal with the same or similar field of study and look to see which ontologies are used by that project.

HAVE A BACKUP PLAN

If your project depends upon the completion of a tool that is in development, it is smart to anticipate that the completion of the tool will likely be delayed. Depending on how critically important the tool is to the project, it is beneficial to reevaluate often and have alternatives in case the tool does not become available within a given timeline or grant period. For example, a question the LLJS team was forced to ask and answer was: “Are there components of the workflow that were originally going to be conducted within the tool interface that might be pulled out and worked on independently while the tool is being completed?”

MULTIPLE ACCESS POINTS

The Linking Lost Jazz Shrines recommendation for others looking to do a similar project is to allow for users to be able to access the data in multiple ways. We took our cue from the Linked Jazz project, which originally allowed users to access the data through a SPARQL endpoint, a data dump, and a network visualization.

KEEP IT SIMPLE

As was the case with the Linking Lost Jazz Shrines project, it may be necessary to stay within the “nuts and bolts” format of data work and just work with CSVs. If you’re at the level of using a spreadsheet or CSV format, that is perfectly acceptable. While use of more complex data management formats or tools may seem more appropriate, it is preferable to use the application or format that best optimizes the team members’ abilities while still fulfilling the project needs. A bulk of linked data work is simply preparing to make linked data, so it is fully acceptable to do that organizational work in CSV or JSON files.

LIMIT THE SCOPE

For those who may replicate this type of project, it is important to know that frustration will result if too many competing objectives are attempted at the same time. Those who might be new to linked data may find that there can often be a blurring of the lines between the data itself and the presentation or visualization of that data. A huge strength of the Linking Lost Jazz Shrines project was that, from the outset, all of the team members were cognizant of that distinction. We all entered into the project knowing that, although integration into the Linked Jazz network visualization was an ultimate goal, before we got remotely close to achieving it, first we had to create the data. The Linking Lost Jazz Shrines project recommendation is to limit the initial project scope to data creation, and to compartmentalize linked data presentations and visualizations as separate components or sub-projects.