So you may have heard of Open Access or even open access, maybe the Open Access movement or open access publishing–but what about ‘Open Access Agriculture’?
When I first heard the term ‘Open Access Agriculture’ I imagined farms with gates wide open in which I could go visit, check out what was there, try my hands at agricultural activities, learn from what was being done there, and perhaps even take some of the products.
While this exact image that I formulated in my head is not correct, some of the principles are similar, as we will soon see.
‘Open Access Agriculture’ was the term coined for a new approach, a new culture even, to agricultural information sharing at the Multi-site agricultural trial database for climate change analysis: Planning and launch workshop, held in Lukenya, Kenya on the 12th and 13th October 2010. This workshop, organised by the challenge programme on Climate Change Agriculture and Food Security (CCAFS) theme 5 (Adaptation pathways under progressive climate change) was :
One of the first activities … to construct, and make publicly and widely available, a database of the performance of agricultural technologies at sites across the developing world. This will build on the existing Africa Trial Sites project which was funded by AgCommons as one of the QuickWin projects and developed by CIAT, ICRISAT, CIMMYT and IITA. Based on the experience gained during the Africa trial Sites project it was decided to take the trials sites concept a step further to include trial performance data and trial protocols in addition to the basic geographic characterisation of trial sites. (Ref: workshop report)
So in effect it is about allowing people ‘in’ to the trial sites via an online platform where they can look around, see what’s happening there, learn from it and also take some of the products–all in the form of valuable data generated in these sites that can now be found in one place. This project is trying to ‘open the gates’ to these trial sites through a specially designed platform which will give people access to the wide variety of data coming out of these sites.
According to one of the organisers, “This initiative will have multiple benefits, such as ensuring that research results are better disseminated, and would add value by bringing disparate data together”.
While the benefits for the wider community may be obvious, the question of how to get data to be shared in this platform is still uncertain. The next session was therefore aimed at spending time to investigate, learn and address “the incentives of sharing trial data”- and this is where my participation came in amongst a workshop of spatial analysts, crop breeders and climate scientists. (Can you pick me out in the photo below???)
I had been asked to join the workshop to share some information and ideas on the general trends of data sharing within the CGIAR and in the wider agricultural research for development community. I started by showing a short film which showed a scenario where valuable data was not shared from a project and the dire consequences it had. The film highlighted the limitations of scientific publications for sharing information and insights, and especially data, and demonstrated the relevance of access to and coherence of agricultural information, and data.
To follow this up I presented some of the initiatives that have been developed to help address this issue, namely the CGIAR Triple A approach, which seeks to ensure availability, accessibility and applicability of CGIAR research outputs(including data!), and the complimentary global initiative aimed at achieving coherence in information for agricultural research for development (CIARD).
But a big issue that reared its head again during this workshop is that researchers (not all but some) do not want to give up their data by making it publicly accessible, because they feel that they have worked hard on generating it and want to use it for analysis and writing papers; “it is my capital for career progression” I was told frankly by one researcher at the meeting. This raises the question of how to get researchers to share their data-something I started a discussion on in through a blog post a couple of months ago on How to motivate more knowledge sharing in research: using the carrot or the stick?.
Andy Jarvis, CCAFS Theme 5 Leader, suggested in his presentation that they will address this issue by using a carrot approach, by appealing to researchers with the following:
- This will stimulate greater use of promising agricultural technologies that researchers have developed
- This will enable retro-fitting of climate data to trials
- The project will add value to data shared through homologues, climate analysis, and enhanced visualisation
- Will allow comparability with other data and other trial sites
- The project will provide some money to rescue dataset on paper or on rotting hard drives
- Once available, experts can analyse in collaboration
But to hear from the researchers present what benefits and incentives they think there could and should be from sharing data, the participants broke up into groups to explore this further.
The benefits listed were interestingly very similar to the African ‘carrots’: Results of a consultation at the African Agriculture Science Week and included:
- Benefits to the INDIVIDUAL such as personal recognition; personal benefits and career progression
- Benefits to the INSTITUTE such as increased credibility, improving efficiency, public relations value, and more access to funding
- Benefits at LARGE/CROSS-CUTTING such as strengthening work of others, better collaboration, and increased impact
So with an extensive list of benefits that could be gained from sharing data identified-the question is HOW to share the data. Groupwork was used to discuss:
- What kind of trials are there?
- What sort of metadata is needed?
- How to collect, store and share the data?
As the type of data and various mechanisms that could be developed for sharing it were being discussed, the Triple A framework became very evident and useful in thinking about what the project wants to achieve and which ‘pathways’ can achieve which level of sharing:
- The data can be made Available by publishing the metadata on the datasets that exist and that will be shared on the platform–so people will know what exists and what they can expect to find at some point
- Data is Accessible when the datasets are stored in their current formats so that the full content can be obtained but cannot be disaggregated such as with repositories
- If the system set up to share the data allows for datasets to be taken apart and parts used and re-integrated such as with databases then this makes the data Applicable
There was also alot of interest in CIARD by the group, including:
- Learning from others who are trying to share data–what pathways has CIARD learnt and documented from others
- Documenting, sharing and promoting the trial sites data sharing ‘pathway’ through CIARD
- Registering the data sharing pathway with the CIARD Routemap to Information Nodes and Gateways (RING) which is a global registry of web-based services that will give access to any kind of information sources pertaining to agricultural research for development
So watch out for those open gates…
Also see blog post by Andy Jarvis on the workshop on Sharing data on agricultural trials: towards a global repository
Photos in the blog: Andrew Farrow, CIAT