Looked at the BM survey on improving their collections online. The answers they'll get will do little to improve it. Here's my view on where it needs to go. 1/n
The whole system needs rewriting using a modern web framework, using up to date open source search software such as solr or elastic. It needs dumps of data to be made available under a proper open license, not restricted by nc-sa clauses.
It needs to feed into large scale initiatives such as Europeana and have an API built as the foundation for the Museum to eat its own dog food. (We researched this in 2016). It needs multiple ways into data - thematic, unified and advanced search, facets, data export.
It needs to be available 99% of the time, not frequently borked and *must* be opensource to allow other museums without their funding privilege to learn. It has to be iteratively designed, built and improved. None of this monolithic project vision museums are fond of.
It needs rich media interfaces, high quality image led with IIIF viewer capability, maps, embedded 3d models to complement the text and make for a total record. It needs cool URIs, content negotiation for JSON, XML or LOD to be exposed.
It needs the vocabularies used to be freely made available along the lines of the Getty. The system should allow for personalised research to be generated, annotations, sharing etc.
It needs staff working on it with a collections background, who understand the complexity of archaeology, anthropology and history of art.
The collection is the one tool of the museum, which has greatest reach for soft power, for spreading the core role of the museum and enabling narratives, change and many other processes. It needs investing in.
We talked lots about the FAIR principles at #reachdialogue. It fully applies to the BM COL. Findable, accessible, interoperable and reusable. At present it is mainly the first, but mainly because of brand power, not through tech wizardry.
Anything that they build needs to give full attribution to the authors, show versioning and be transparent about collecting histories.
Now I'm on the outside, it feels like the BM sees its collection online as a private silo, not one to link across resources *note I'm excluding researchspace*, or to share knowledge freely via well documented standards.
Going back to opensource, collaboration on code would be great. Imagine the missing features a 3rd party might build that could benefit the research community as a whole (just set up contributor licenses.)
I also hope to see linkedart LOD being implemented more. Accessible design (for low bandwidth communities and for users who needs access tools.) Yes there's a lot to cover, but building this stuff enables research grants inside and outside of WC1B.
I hope they are also looking around at sector leaders, at third party collection browsers and also learning that people often deliver fantastic collections front ends and then leave. Value the staff and retain them, pay proper wages.
Damn, idealistic Dan has appeared...
Should probably add, this was all stuff that I worked up when I was technically product owner at the BM, but couldn't do anything as the internal collection system was so behind in its upgrade cycle. Big aversion to open source there.
