GLAM Workbench

Recently there were some invite-only discussions about National Digital Research Infrastructure, including one on #HASS & #GLAM. The questions and summary from the HASS/GLAM discussions are here: https://lnkd.in/gVgtAe5x Most of it's pretty obvious, but I'm encouraged by the concluding para: "Overall, rather than a single, all encompassing facility, the focus should be on developing an interconnected ecology of systems and tools, with emphasis on collaboration, accessibility, and adaptability to various research needs." So glad we're finally moving away from the idea of a single platform...

The GLAM Workbench includes a list of Australian GLAM (Galleries, Libraries, Archives & Museums) data sources with details of APIs, downloadable datasets, data portals, and Zotero compatibility. I've been updating the list over the last couple of days. More contributions are always welcome! https://lnkd.in/g3sFUVqQ There's also another long list of datasets shared by Australian GLAM organisations though government open data portals. I re-harvested all this data yesterday and it currently includes details of 22 organisations, sharing 443 datasets, containing 1,251 data files! https://lnkd.in/gvSWN3rQ You can also search the harvested metadata using Datasette-Lite: https://lnkd.in/gAV2RS-H #GLAM #HASS #openData

Just finished processing another volume of the Tasmanian Post Office Directories. The volume for 1940-41 is now OCRd and searchable, along with the other 50 volumes from 1890 to 1948. There's still a few more from the 1940s to process, but I'm hoping to add one a week until they're done. https://lnkd.in/gVecFEiA

Visualisation is a great way to find problems in your data. I was processing records from the Tasmanian Post Office Directories the other day when I realised that 1920 was missing from my online, searchable database. Eeek! I've spent the last couple of days fixing this, so if you want to explore *all 49* volumes from 1890 to 1948, including 1920, you now can: https://lnkd.in/gVecFEiA It wasn't a straightforward process because the 1920 PDF was quite different to all the others. If you're interested in the gory details, there's some information here: https://lnkd.in/gSjygzdE #GLAM #familyHistory #localHistory #digitalHumanities Libraries Tasmania

The work of #maintenance is never done – some details of last week's updates to the Trove newspapers section of the GLAM Workbench: https://lnkd.in/gZSvquCt The Trove newspapers section includes 23 Jupyter notebooks and 6 datasets, so it's not a small job! #HASS #GLAM #researchInfrastructure

If we don’t document the shape and structure of online GLAM collections now, how will future historians understand their impact? This post includes a list of datasets I've created from GLAM collections (mostly Trove & RecordSearch), and some notes about why I think it's important to do this sort of work – it's my love letter to future historians... https://lnkd.in/gPCa8pHV #GLAM #HASS #openData #digitalPreservation

It took a week, but the Trove newspapers section of the GLAM Workbench is now fully updated. That's 25 notebooks and 6 datasets. https://lnkd.in/gmPjj9zZ For those who are interested (probably nobody), as well as updating the code to work with v3 of the Trove API, I pulled all the datasets out of the code repo and put them in their own dedicated repositories, added metadata to the notebooks, and created an RO-Crate metadata file for the repo. I also made some changes to the documentation site, adjusting the notebook previews, and moving things around. I still need to adjust some things relating to the datasets, but it's not major. #HASS #GLAM #maintenance

I've added another dataset to the 'Trove historical' collection on Zenodo. This dataset includes information on approximately when newspaper titles were added to Trove between 2009 and 2021, extracted from web archives. (The dataset itself isn't new, I'm just giving it a permanent home.) https://lnkd.in/g_nH8ESR Combined with the weekly newspaper snapshots which have been running since early 2022 (https://lnkd.in/gqWQbPUX) this should allow future historians to explore the historiographical implications of when particular titles became available online. #HASS #openData #newspapers #history