These wiki pages are a working area for the SWAF 2012 workshop. For the main workshop pages, please go here.
Hashtag for the event is #swaf12
One of the key challenges in the agri-food area is the integration of data across the various stages of the supply chain. Within the SmartAgriFood project we refer to this as the “super-scenario” i.e. our vision is one where all actors along the supply chain are able to provide semantically marked up data in a manner that is “ridiculously easy” for all events or actions. This data needs to be made available (published) in some form that can be consumed by other actors, queried and reasoned over.
In this hackathon, we (Daniel and Christopher) would like push as far as possible towards this vision. However, participants should feel free to self-organise and identify other agri-food challenegs they wish to undertake.
We would like to build a demo/prototype of something like what we have called the “super-sceanario” in SmartAgriFood, i.e. the linking of several stages of the supply chain and using several vocabularies.
As a set of tasks, this ambition may involve the following:
- Vocabularies people know of or turn up with/bring to the day
- Create a small set of information items that might be of use in the agrifood supply chain for each of them ←— We need this data before the day so as not to spend time data wrangling
- Give people some time to put together some samples, ideally maybe already with some references to other vocabularies (taking care for the linking with the linked data)
- serve them via http
- either by peoples own tools, which we should invite them to bring in
- or via some prepared server, into which we can throw the prepared instances
- Think of things people might want to ask to this linked data network ←- @cbrewster will provide examples via Tim Verwaart
- Implement this (probably using SPARQLs named graphs or version 1.1 federated queries to pull together distributed data)
Ideas presented here are based on the abstracts handed in. A large part of the ontologies presented are based in agriculture. There seems to be a certain lack of ontologies in upper layers of the supply chain. Nevertheless, a first step would be to start linking up what is there in primary production and roll it out to upper levels later on.
- Grain drying as a process to be conducted after harvesting: see what might be done about deriving parameters necessary for grain drying process control from documentation and merging/linking them up for next supply chain stages (effectively making the ontology presented by Kluza et al. and Martini et al. cooperate)
- Integration of available semantic technology based pesticide data portals (data.igreen-services.com and/or psmid.de) in Germany into the agrocompanion semantic web app
- supply chain integration of geospatial data: roll-out of technologies proposed by Norremark et al. into the supply chain
- Building a generic input/output flow ontology for agricultural processes based upon the algal biomass ontology framework by Solanki. This could be useful e. g. for calculating sustainability indicators like nutrient balances, economic parameters etc.
- Crop disease monitoring + agrocompanion: Any possibility to improve by integrating the systems proposed by Damasio et al. and Jackman et al.?
- Leveraging unstructured data sources: any chance of integrating information from the annotation work presented by Jaques et al. into the more strictly structured ontologies presented (e. g. Jackman et al., Kluza et al., Norremark et al., Martini et al.)
- Relating food stuff information to databases provided by the EC: using the annotation mechanisms presented by Jaques et al. to link distributed food stuff information - provided by producers, consumers etc. - e. g. to databases of authorised food additives
- Using methods provided by Scholten to integrate EPCIS into a semantic web supply chain infrastructure (specialist topic)
- The workshop
- Followed by the hackathon from 16:00 - 20:00
- Snacks to fuel brain power
For the afternoon VoCamp and hackathon session the following things might be useful:
- motivation to edit and work on small (or larger) semantic web datasets, vocabularies or tools
- favourite text editor
- ontology development tools like IDEs/IDE plugins, visualization tools etc.
- favourite programming language tools and accompanying semantic web libraries
- linked data/ontology server installations
- virtual machine software with prepared machines to be able to simulate networks
So one or more should be brought by each attendant. Internet access is available at the conference site.
- Protege 3.4.8 and 4.1 installation
- redland rdflib installation + python bindings
- VM with webserver instances
- access to several iGreen project developments