All blog entries reflect the opinions of the author and have not been expressly endorsed by the Ivan Allen College of Liberal Arts or the Georgia Institute of Technology.
Perhaps one of the biggest challenges I've faced in assessing and planning the rebuilds of several Ivan Allen College (IAC) school websites so far is the need to provide faculty profiles. The challenge is that this is much more than just a simple biography page these days - faculty and administrators want a lot of intricate details about their work posted, including listings of publications and courses taught. You also have the issue of instructors with appointments in multiple schools, who traditionally end up with multiple profiles that never stay in sync.
The lesser seen but possibly biggest time sink is that there's been a tendency (at least in the sites in the college) to build the whole kit and caboodle into each school's website. So every website has some implementation of complex faculty profile data structures and editing mechanisms, which each have their own quirks and are usually not at all compatible with each other. Given that I knew I would be rebuilding all six IAC school sites over the course of my first year or two with the college, I did not want to go back down this same road and be continually reinventing the wheel.
Fortunately, there was a brief glimmer of a light at the end of the tunnel: a project called GT Scholar. Unfortunately, this system won't be fully ready for prime time use until 2017, leaving me with a dilemma: I can't wait for GT Scholar to be completely finished, but I still didn't want to implement complex profile systems on six different school websites (not to mention a couple of project center sites.) My solution was to abstract the profile data storage and management out into its own web application, which would be tailored from the ground up to pull data from outside sources, including GT Scholar. It would start out as an authoritative source for data that I couldn't pull from GTED (e.g. personal biographies, office hours, personal website link, CV files, etc.), but would be able to eventually pull some of that data from GT Scholar. It would also have its own database for courses taught throughout the IAC and publications produced by our instructors. Some of that publication data has been imported from our old websites, while other data is coming from an initial import process from GT Scholar, though we can't fully rely on GT Scholar at this time.
Much like with our News Translation Server, this system minimizes the complexity of each school site, allowing me to rebuild them at a much faster pace. Instead of having to setup and maintain custom content types, views, field modules, etc., those sites just utilize a single custom module that talks to the central college Faculty/Staff Directory server, which has a RESTful API that returns all needed data structures preprocessed and formatted in JSON for easy transport and parsing. The only thing the custom module has to do is implement different dynamic URL paths and generate the HTML formatted output of the data it receives along with UI controls for navigation. Being a completely self-contained module (as opposed to a content type / fields / views based implementation), it is very easy to push out code updates and ensure that all school sites have the latest features and bug fixes. The module is built around a standard formatting system that can use custom rendering plugins for schools that want their faculty pages to have a unique layout or a special field visible.
Like so many projects, this one is still a work in progress. I have it running in production on one project center site and have integrated it with three new school sites that are still in the late development / review phase. By the end of the summer, I hope that all three sites will be pushed out to production, allowing me to then focus on the remaining schools. I'll try to remember to post an update here when I get to working on those sites to report on how well the college's Faculty/Staff Directory system is working and just how much time and effort it saves me in rebuilding those sites.