National eResearch Architecture Taskforce

The challenge around NeAT is to make a strategic investment with limited funds that:

• Builds services of continuing importance
• Delivers early value and leads to widening adoption
• Integrates technologists and users in teams
• Produces generalisable solutions from specific interactions with users

Overall one could imagine a matrix made up of:

• service components down the rows; and
• a user community and a specific need per column.

The identification of NeAT projects could be achieved by diagonalising the matrix; that is, bringing the columns (needs) that are similar together, and bringing the services related to the same need together, to create larger projects. This kind of analysis would work, if as we suspect there are indeed ‘hot spots’ where some common needs require similar packages of services, and where there are early adopters that can be engaged in initial projects.

The NeAT process is intended to bring together a critical mass of the right people to effectively target the obvious hot spots. Therefore the proposed project definition process is to develop a package of activity for identified communities and a broadly stated need, and then adjust those packages

• in terms of best fit for overall service development (coherent and not too complex) and resource availability; and
• to ensure that an envisaged future service landscape is covered by the combined projects.

Consequently the development of a ‘services blueprint’ as requested by AeRIC is a key input in refining the projects, but not essential in outlining them.

This is also an argument that the 80/20 rule applies, and as the error bars on everything are so large, it is better to start and evolve than plan to the last detail.

The envisaged iterative and negotiated process also argues against an open ‘call for proposals’, leaving the challenge of inclusion as a task to be managed.