I’m very interested in coming to Witzenhausen. I’m going to say a few things below about what we (Solidarity Economy Association) have done already in the world of data and maps, what we plan to do, and areas I’d like to work on in Witzenhausen.
We’d like to see maps that can consume data from many distributed sources. We also view maps as just one type of application that is powered by data describing the Solidarity Economy. This has several implications:
- Separation of the data from the mapping software that displays it.
- Use of standards for data.
- (jumping into detail here!)The ability to recognize when data from two (or more) sources is actually describing the same thing in the real world. The data can then be merged (e.g. on a map you can see merged information about a place which has been described in more than one dataset).
- many other things too!
The Web of Data
Background viewing: Tim Berners-Lee’s TED talk on the Web of Data
There’s plenty of data out there that is already available in a standard, machine-readable, form. We want to take advantage of this. For example, DBPedia:
The English version of the DBpedia knowledge base describes 4.58 million things, out of which 4.22 million are classified in a consistent ontology, including 1,445,000 persons, 735,000 places (including 478,000 populated places), 411,000 creative works (including 123,000 music albums, 87,000 films and 19,000 video games), 241,000 organizations (including 58,000 companies and 49,000 educational institutions), 251,000 species and 6,000 diseases.
So instead of just saying that we are based in “Oxford, UK” (the familiar, but quite limited string of text!), we link into a vast network of information: http://dbpedia.org/page/Oxford.
All our developments are on github.
We have chosen to use Linked Open Data as the technology for creating distributed data. It provides the hook into a vast world of self-describing data.
So far, we have created a few datasets. Recently, we created a ‘dotcoop’ dataset - containing the UK coops who have a .coop domain name. You can browse the data using a standard linked data browser. For example, this is a view of the Web of Data, starting with our dotcoop dataset as provided by Southampton University’s Graphite browser. If you want to play with it, just start clicking to see the machine-readable data! See if you can find the following:
- Data about the Agile Collective (hint search for ‘agile’!)
- Links to other datasets - e.g. ordnancesurvey (for geo data about a postcode), Companies House (for government info about a limited company)
- links to the same real-world thing being described in a different dataset (hint: look for “sameAs”)
We have created a simple Leaflet map that will display any of our datasets. For example:
We know Linked Data can look scary. We have developed tools to generate Linked Data from the lowest common denominator of data formats: the humble CSV (simple table, as in a spreadsheet).
I’m very happy to share my knowledge with others at Witzenhausen, but I’d really like to make progress with the community towards the next steps along this path. Below are some of the things on our list of what to do next.
Our linked data currently has very little in the way of taxonomic description. I am very interested to explore how to bring other taxonomies into this world. For example those used within TransforMap. Our linked data is based on the ESSGlobal vocabulary created by @mariana, and @jnardi at Ripess Europe. I have put the essglobal vocab into github, and we have permission (and encouragement!) to evolve it as required. I’d be very interested in doing some detailed work on this in Witzenhausen.
Create new Linked Open Data datasets. SUZY, Permaculture, TransforMap, others???
To date, we have been creating data from sources which are readily available. e.g. Coops UK and dotcoop. This has been great to provide data for experimenting with the Linked Open Data concepts and technologies. But the greater value comes from mapping the invisibles - those initiatives that are at the heart of the Solidarity Economy. We plan to run a pilot project to do this in our home city of Oxford. But we have no experience of this. We need help.
Our current Leaflet map is very simple. You can’t even search on it! It was designed partly as a platform for experimenting with, and demonstrating Linked Data. We want a better map, powered by distributed Linked Open Data, and designed around people’s needs. Should we use our map as a starting point? Should we look into adapting a different map to be powered by Linked Open Data?
Links to other datasets
- I mentioned DBPedia above, but we don’t yet link into it. An obvious starting point is to link to geographic data (e.g. Oxford, UK). How would the be best done, given an abitrary postal address?
- Are there other categories of data in DBPedia that we should link to?
- Are there other datasets we should link to?
I’d very much like to come to Witzenhausen, and will book tickets if what I’ve described above is going to be of interest to others.