I had the opportunity to visit the ECSA conference in Berlin. The event attracted a 200-ish citizen scientists and related organisations from around the globe, with contributions from biodiversity monitoring programs, ecological monitoring (air/water quality, light/noise, flooding & weather monitoring) and various other topics such as astronomy, health & emergency response.
One of the main drivers in this topic, Muki Haklay, made some detailed blogs about the event.
On the first day I got distracted by the several definitions people seem to use for citizen science, which led to an interesting discussion on twitter, but later was resolved by Lucy Robinson, who presented 10 principles of citizen science.
On the third conference day I presented some of our work in the scope of the COBWEB research project, on which some of us worked in the last 4 years. In the scope of the COBWEB project GeoCat has been improving GeoNetwork and GeoCat Live to better facilitate Citizen Science.
It was interesting to notice that while we were working on the matter, the citizen science associations have been working on similar matters. Two challenges are identified on which the COBWEB best practices do offer potential solutions.
The first challenge is discovery of project information, citizen science portals (such as those based on GeoNetwork) should be able to store, visualise and exchange project metadata with the goal that citizens will easily discover citizen science campaigns running in their area, irrelevant of the portal they are looking at.
The citizen science association of america has actually already proposed an exchange format for project metadata: PPSR-CORE. Having such a standard in place will increase the wide discovery of citizen science projects. However there is a note to make here. Since the association hasn’t based their work on any of the existing more general standards, I think they’re missing an option to benefit from existing infrastructures.
An existing infrastructure that could be benefited from is the schema.org ontology. A similar initiative, that of the “web observatories”, which is based on the schema.org ontology, shows how one can set up a model for participatory projects as an extension to schema.org. The benefit of schema.org is that it is already well accepted in a lot of domains, such as the search engines. Schema.org already offers ontology for other objects relevant in the citizen science domain, such as websites, images, datasets, scientific papers, locations etc.
The second challenge is the discovery, assessment and conflation of CitSci observations by other (citizen) scientists that may want to use the raw data in their research. Currently much of the citizen science frameworks either don’t offer an endpoint to the data, hardly advertise the endpoint and provide it in uncommon formats. Resolving this issue is not an easy task, since it requires adoption of standards for each of the relevant domains. During the conference some even doubted if adoption of standards would restrict the freedom of citizen science. Also in this case extensions to schema.org (combined with json-ld or csv-ld) may provide an easy step to get started with standardisation.
Related to this another thing stroke me. Quite some of the citizen science communities themselves are hardly aware of related datasets to their work, which could facilitate their work considerably, if they would be able to embed that data in their work processes (for example maps with citizen air quality measurements didn’t include measurements done by the officiel agencies, which are available as open data). My expectation is that they struggle with similar open data challenges as the general public; existing data portals are quite hard to use, data formats are complicated, the open data license doesn’t match or the aggregation level of the open data is limiting. For this reason I decided to focus my 10 minute presentation (or would you call that an elevator pitch) on the recent work we did in the geo4web testbed organised by GeoNovum. I presented the OGC proxy approach, which exposes CSW and WFS services as schema.org (RDFa and json-ld), so it’s easy to crawl by search engines and easy to use by web developers. The presentation was warmly welcomed, considering the amount of questions I got just after the session. I was surprised to see one of the next presentations in the same session presented a very similar approach to expose iNaturalist api’s as SOS. Apparently there’s a lot of movement in this area.
It was an interesting conference. More and more tooling and experiences become available, which allows this community to empower society. I expected a bit more attention for privacy aspects of citizen science data. “Privacy by Design in Citizen Science” will probably be the topic I will submit for the next issue of this conference.
Group picture by Florian Pappert.