A couple of months ago I introduced a project to work on a combined resource of cultural collection resources across Tasmania. Well that project is well and truly underway and mostly complete. It’s now going through review, so I can’t send out any links quite yet, but I thought it would be an opportune time to talk about one of the major deliverables that we built for the project. And that is, using Microsoft Excel as the aggregation tool for all cultural objects and linking them to locations.
Microsoft tools often get a bad reputation, and often for some good reasons (my Mac bias kicking in here), but there’s no doubt the Office suite is ubiquitous and something most people have access to and can use. On the flip side, inter-operable formats like CSV are not something everyone is comfortable using and they also don’t allow for features such as pick-lists or validation.
For this project, we wanted to enable as many people as possible to contribute cultural collection pieces, and link them against places around the state. It became clear there was no common collection system, and getting everyone to create accounts on a web system and duplicate their records and images was never going to work. That’s where our Excel process kicks in.
As the project is still under development, the following is a description at a very high-level; we will wait for the resource to be publicly announced before going into any serious detail.
But the publishing process starts with an Excel workbook with two sheets: one has a set of core metadata to describe objects, and another with media references. The Excel sheets have a pick-list that allow publishers to create relationships between places and objects, and additionally there is mechanism to create tags and categories for the objects. Publishers can then put all of the images alongside the Excel workbook into a folder, zip that up, and we developed a pretty nifty system to upload the folder, and then unpack those resources and publish them in a consistent way to a common web resource. So effectively, no-one needs to learn a new platform, duplicate their content, or work CSVs or text files.
Our publishing system was created in Amazon Web Services (AWS) as a serverless build, which means it doesn’t cost anything to run unless it is ingesting and publishing materials, to help minimise operational costs. There are a series of Application Programming Interfaces (APIs) that allow the AWS publishing system uses to directly push content to the host website, and it can also ingest feeds from other 3rd party systems from across government and the wider sector.
That’s unfortunately all I can really say at this point, but once it becomes more public, I look forward to describing the process in a lot more detail.
Lastly, I can’t write a post about Tasmania without sharing the inspiring beauty of the place. Here is sunset over Mt Wellington – an amazing, daily occurrence in Hobart.