Writing the Foundations of a Drupal/Islandora Partnership - Part One of an Interview with Susan Brown
Digital Echidna has been working with a number of partners on a project called the Canadian Writing Research Collaboratory (CWRC), a project that brings together researchers working with on-line technologies to investigate writing and related cultural practices relevant to Canada and to the digital turn.
CWRC was our first major Islandora-based project and, as Islandoracon 2017 is currently underway in Hamilton, ON (and Echidna's on site!), we thought we'd share a conversation with Susan Brown, the project leader for the Canadian Writing Research Collaboratory.
Today we start with Part One -- why CWRC initially selected Islandora, its strengths and weaknesses, and the challenges the community were facing.
ECHIDNA: What are the strengths of Islandora, why did the CWRC team initially select this platform for their Digital Asset Management (DAM)? What are its weaknesses, based on your experiences using the platform?
BROWN: “I actually just came from a workshop in Montreal where I was talking to Mark Leggott, who was one of [Islandora's] founders and he’s quite excited that Echidna and a couple of other companies have entered the market, and that a robust ecosystem has been developing around it. It’s open-source software, as you know, but in his view, it’s successful because it’s starting starting to attract a diverse set of players.
"We chose Islandora because we were looking for a really robust platform that would be consistent with best practices in the digital humanities, so we wanted to make sure that -- in the very first instance -- it would take care good care of the data, which is the most important thing. So having a Fedora base layer for the repository and knowing that it’s preservation-oriented and good for pushing out objects into long-term preservation systems. And, of course, that’s improved over the five or six years since we made the decision to go for Islandora. That was one of the really compelling reasons.
"Then, of course, there was the front end. We looked hard at Hydra, which was in its infancy then, and Islandora was, of course, not nearly as far along. We did a lot of talking to people and trying to suss out where things were going. I think it was clear that Islandora was further along -- and it gives you a lot out of the box, I guess you would say. Even then it meant that we didn’t have to deal with authentication and on some of the basics it gave you a GUI (graphical user interface) to work with that was customizable because it was using Drupal, which has become, even more than then, a very commonly used CMS for web development. It gives you the best of both worlds in terms of having the archive-oriented object store store and that flexible front-end."
ECHIDNA: Any key weaknesses?
BROWN: “I think that like anything that’s powerful it’s complicated to get your head around at first. I think that one of the biggest challenges for us as a team was to understand how the various components in the system interacted, like exactly what a content model was. It took us a long time to really figure out how things like content models and microservices interact.
"Basically It’s hard to get a handle on where there are areas of flexibility and where there are areas of constraint. I’d say that was the biggest challenge: the complexity of the way that Fedora interacts with Drupal and what should be happening at what level."
ECHIDNA: Prior to implementation of Islandora, what challenges were you facing and what caused you to look for a system of this nature?
BROWN: “We were looking for a system that would support distributed collaboration and the creation of content across multiple sites, multiple institutions, multiple users, multiple time zones... So we had a system that had been developed in 1990s and that had been built from ground up, which was pretty powerful in lots of ways, but it was composed of a bunch of different components cobbled together. Some of them involved interacting with a web interface. Some of them involved working with software that you would install on your desktop, so it was a very complicated, Byzantine workflow that took a lot of time to train people in, 'here’s where you go to do this, here where you go to do that.'
"We were looking for an integrated platform that would bring the various tools and services that we needed researchers to be able to use together and that would handle the workflow in a much more intuitive way within a single environment."
ECHIDNA: As a community as a whole, where there needs or difficulties that needed to be addressed by this solution?
"Yes. I think the biggest one was that Islandora was really oriented towards repositories and digital archives. And although it was designed to provide a flexible user front end, the back end was really oriented towards having a small number of people work on ingesting materials and preparing them for dissemination through the front end. But it wasn’t really designed with a view to having a set of dynamic research materials being worked on repeatedly over a period of time by a larger team of users. So the permissions and workflow tools that were initially there weren’t adequate for our purposes; they weren’t oriented towards dynamic, collaborative scholarship.
"We also wanted to have an integrated editor that would support editing XML documents and creating associated RDFs/semantic web annotations. There was kind of a rudimentary XML editor, but it wasn’t very robust. I think it was created by students as a summer project -- by CS students, or something like that. So we built the CWRC writer/editor, which is the centrepiece of CWRC, to fill that gap."
Up next, why the CWRC project team decided to go with a Drupal shop to build their site -- and the benefits (and challenges) that decision brought about.