The CREW team attended the JISC VRE Programme Meeting in Reading last week which usefully brought together all the current VRE projects in order to look for potential areas of cross-project collaboration and interest.
I gave a presentation on the CREW Project following presentations by the other 3 projects.
The VERA Project described their continued work on supporting archaelogical research with fieldwork tools for the recording of research findings and a central database (IADB – Integrated Archaelogical database) to which records can be uploaded automatically, and from which they become accessible to researchers via a portal environment. Their technical goal is to migrate from their VRE1 bespoke PHP application to using Gridsphere as a JSR168-compatible portlet container – it will bridge to the VRE 1 portal. They have more portlets planned – for 3D visualisation etc. They focus on portlets consuming web apps as opposed to shoe-horning applications into portlets, whilst respecting the advantages of portal frameworks (such as providing personalisation opportunities). Towards this goal they demo’d their RECYCLE bridge which can consume most web apps (Matt Grove’s work), addresses portal security, may require some CSS tweaks but on the whole is a working solution. The demo showed examples of wikipedia bridged, also wordpress bridged Their use of bridging occurs on the VERA project regarding the IADB for example. This bridging work is of relevance to the CREW project (we need to integrate the Events Application in a portal environment but don’t want to compromise the extensive functionality available within the Events Application user interface). A cross-project collaboration effort on this is being led by Mark Baker. Claire Fisher is coordinating their user engagement and they plan to also use portal stats analysis, and consult and workshop with specialists to see how they work with the IADB.
MY EXPERIMENT is like “Facebook for scientists”, said presenter Dave de Roure. And like a frontend to workflow systems, simple or complex, large or small. They use Taverna which came out of the MyGrid project and operates via a pooled set of web services. Their users are quite wide ranging. There are reuse and tutorial aspects to the system in terms of the ability for scientists to upload their own workflows as self describing MyExperiment design objects and to cite other people’s workflows. Reuse encourages users to document what they’re doing to contribute to the system. They are using Openid. Privacy, licensing and tagging are supported in similar modes as with facebook. These aspects overlap with CREW project issues – as does for example the tension they note arising from the use of informal tagging versus using ontological classifications. I think Dave de Roure said that use of their system shows tagging quality is good for discovery, ontological quality being better once an end-user had delved down and reached more specific resources. They changed their software license from GNU to BSD license recently.
OXFORD VRE Project: Mark Pybus presented on what a VRE would be for Humanties. They are engaging with disparate randomly funded projects to get them to work together. Regarding archaelogists they are doing quite a lot of work on tools for working with images. They have adopted Uportal and are using Shibboleth for authentication. Authorization is still open question on their project. In their portal they have an image upload, then search facility, links to chat services e.g. jabber, and a link to the access grid. They have an image and annotation store. Christian Fernau (who we at ILRT collaborated on previously in the JISC funded SPIE project) was there with some helpful comments on the PERMIS decision engine supporting annotations and security aspects to CREW software. He suggests we look at SimplePermis as a much more light-weight and suitable component in the access policy controller we want to implement for the CREW project. Sounds more like what we need behind the Authorization API planned for CREW Events application Architecture. I agreed to lead collaborative, cross-project efforts on annotations and provenance tracking/security in social softwares in the VRE programme.
