OCKHAM Registry Experiment Workshop

16-17 January 2007

Seattle Public Library, Seattle, USA

Ann Apps and Amanda Hill attended this meeting, which was organised by Jeremy Frumkin, who leads the OCKHAM project. The meeting was held in the futuristic building of the Seattle Public Library where the meeting rooms were to be found down red, 'womb-like' corridors. The purpose of the workshop was to inaugurate and plan a Registry Experiment for the US National Science Digital Library (NSDL) based on the OCKHAM Service Registry. The people invited to the meeting represented various projects that it is hoped will participate in the experiment. The project plan is under development on the Registry Experiment Wiki.

The General Recommendation Engine being developed at New Jersey Institute of Technology is a 'more like this' service that aims to suggest documents and services that other users with similar tasks or interests have used. This project is investigating three methods for providing recommendations:

• collaborative filtering, which uses statistical techniques to process the browsing history of like-minded users;
• content based filtering, based on the content of a selected document; and
• knowledge based filtering, which considers a user’s level of knowledge, domain expertise and tasks (this last being leading-edge research).

Currently the Recommendation Engine operates on five collections that are mirrored locally. They would like to provide access to a wider range of collections, hopefully NSDL-endorsed collections in OCKHAM. But there are issues to resolve. The second two filtering methods need to process documents in advance, and the first would need access to a user’s click-stream from remote sites. Possibly a service registry could assist by recording collections that provide logging statistics services. As part of the registry experiment, the Recommendation service intends to use OCKHAM to discover new suitable collections with which to work, and also to advertise itself as a service within OCKHAM.

The National Science Digital Library contains resources within the areas of science, technology, engineering and mathematics, along with metadata and contextual information. The resources include 'Expert Voices', which are blogs about science and events. NSDL Pathways are audience specific views of appropriate resources and services (i.e. portals). NSDL intend to describe these Pathways collections in OCKHAM. The Core Integration Group of NSDL gives endorsement to quality resources. They intend to experiment with using the registry as part of their aggregation activities, as well as advertising their available services. They would like a mechanism within the registry and the metadata to mark a resource as endorsed by them, preferably in some 'signed' way.

The OpenCourseWare Repositories project aims to improve discovery of course materials by providing federated searching over repositories. They are investigating innovative Web 2.0 technologies. They have a delicio.us-like tool to describe relations between web sites called Scrumdidilyumptio.us, a social tagging tool, Ozmozr, and have implemented microformats on items such as dates in the web pages of blogs. They intend to experiment with discovering and configuring access to collections using the registry, and also to advertise their services.

It is also hoped that the Skolr research repository project will participate in the experiment. Their representative was unable to attend the meeting because of snow and ice storms.

There was some discussion about the possibility and practicalities of implementing a service registry at the DNS level. This could be based on the technology used by Zeroconf / Bonjour. If this is enabled with a service discovery mechanism, it allows for dynamic connection to services. Zeroconf is utilised by iTunes to automate music sharing by users who are on the same sub-net. RFC 2782 specifies fields in DNS records, which if set can enable automatic discovery of suitable services of a particular type. At the meeting Dan Chudnov was quickly able to set up a demo of this in action, using some service data records he’d gathered from IESR via OAI-PMH. Developments at the DNS level are complementary to initiatives like IESR and OCKHAM. It allows for discovery of services that implement a particular protocol or serve a particular function. IESR’s 'transactional services' probably include those that would lend themselves to 'low level' service discovery. But this method doesn’t lend itself to implementing use scenarios that are based on discovery via IESR’s and OCKHAM’s rich collection descriptions, i.e. discovery of 'informational services'.

IESR has a stake in this experiment because the OCKHAM registry uses the IESR metadata schema. OCKHAM intend to harvest some of IESR’s records to increase the range of collection and service descriptions available. Conversely IESR intends to develop ingest of XML records via OAI-PMH and import OCKHAM descriptions. These will include the NSDL Pathways resources when they become available in OCKHAM.