I have used Archivematica on and off for several years, but my use was quite traditional. It was a desire to deepen my knowledge and how to get the most out of the system that I signed up for the virtual Archivematica camp.
Day 1 (Bootcamp)
After the welcome from our hosts at Concordia University and Artefactual bootcamp started by looking at the broader digital preservation landscape. We started with a few definitions of digital preservation and the importance of active management of content selected for preservation. We heard how Archivematica was built around the OAIS model before diving into the detail of the microservices which are central to Archivematica and the many third-party tools including BagIt, METS, PREMIS and PRONOM to name just a few.
Day 2
Advanced user training began with a look at the structure of the AIP and a deep dive on how to read (and understand) the METS file. The sheer size of the METS file had previously discouraged me from taking more than a quick look but the session took us through the structure and components of both and introduced new terms like dmdSec, techMD and structMap.
We then returned to the practical side as we returned to have a look at AIP and METS files we had created on Day 1 including using the METSFlask tool. Whilst recognising it does not replicate all of the details and content in the METS XML file it is certainly easier to ‘read’ and was visually more engaging.
The day ended with a case study from our virtual hosts. I was initially surprised that Concordia University Records Management and Archives and Concordia University Library did not share the same implementation but their presentations showed the two sections were at quite different stages of their ‘digital preservation journey’.
Day 3
Day 3 started with a look at workflows including manual normalisation, most typically used for digitisation content but equally applicable to scenarios for content with no normalisation path in Archivematica. We looked at the Format Policy Registry and the relationship between rules, commands and tools.
We spent some time on the crucial pre-ingest stage – the key advice being to think about the results you want and then work backwards! A checklist for this would include consideration of the structure of the transfer package, PREMIS metadata and normalisation pathways (for more on this see an article on Scholars Portal). The day also featured a session where four campers shared their experiences and talked through aspects of their work giving an interesting insight into the workflows and the particular challenges they faced.
Day 4
We started the final day with the Backlog and Appraisal tabs and AIP re-ingest again looking at a few scenario’s where they might be appropriate – we also finished the last of the 16 exercises. Sarah Romkey from Artefactual joined the camp to talk about the move towards a more open ecosystem and highlighted a few of the projects being developed in the Archivematica Labs on Github. In response to a question I had about reporting and being able to ‘see’ what content you had (eg by PUID, format group etc etc) we were given a demo of one Lab project AIPscan which addressed this very issue and which looks very promising.
Reflections
Personally I thought the virtual format worked very well. The team were so helpful and supportive, responding to questions via chat on Zoom and through the use of Slido which allowed us to ask questions anonymously and allowed others to vote on them – the most popular questions were then the first to be answered.
The practical exercises were excellent, well described and thought-out, often returning to files we had created in previous exercises. There were three of us in the breakout room with access to a virtual machine. With one person logged-on it was possible to watch or to take control via screen-share and do the task, we were encouraged to look around and be curious. The virtual machine was left open to us for a further week allowing me to go back and repeat the exercises to increase my familiarity with the software and the features. Personally I would have liked to have seen more on the integration with AtoM but recognise that this is a day conference in its own right!
A huge thanks to our hosts Kelly, Jenn and Sarah from Artefactual and to John, Sarah and Mirelle.
(Thanks to Artefactual for permission to use extracts from slides, CC BY-SA 2.0).
