Presented at the DCC Roadshow in Dundee on ‘Research data meets research information: the IRIOS, C4D, CIA and CRISPool projects’
ROS includes data sets as an output so would be good to add this to CiA … so crossover with C4D work.
See details of workshop and presentations at http://www.dcc.ac.uk/events/data-management-roadshows/dcc-roadshow-northeast-scotland
CERIF experts and newbies met at Bath on 9th and 10th Feb for an in depth tutorial on CERIF from euroCRIS’s CERIF Task Group leader, Brigitte Joerg.
This was followed by a data surgery where members of previous and current JISC RIM & CERIF-related projects discussed how CERIF mappings had been done – what had worked and where gaps in the model or their own understanding existed.
More info at http://www.ukoln.ac.uk/#cerif-tutorial-F2012-01-05-23-25-rr
The CIA mapping document was presented and several of the issues raised by this and other projects were discussed – in particular multiple identifiers e.g. for a single person, missing classifications schemes, names and how structured the model should be, and the proposed new CERIF-XML format.
We will assimilate the answers and follow-up outstanding questions over the next couple of weeks in order to finalise the CIA model before work starts at Atira, Symplectic, ePrints and RCUK on implementation
Below is a link to a collaborative spreadsheet which maps CERIF entities from the CERIF semantics and three previous projects: CRISPool, IRIOS and CERIFy. The aim of this mapping is to help agree a standard model for CERIF-XML.
The second tab lists the relevant entities together with areas for further discussion (i.e. where the CERIF semantics and one or more of the previous projects differed) and comments as applicable.
The CERIF Semantics are to the prefinal version of CERIF 1.3, but the mapping document will be updated on release of the final semantics from euroCRIS.
Any feedback is appreciated on this mapping and can be updated, provided you have a Windows Live ID (alternatively, emailed to Scott Brander, CiA Project Manager). We will be discussing the mapping document in more detail at the CERIF Tutorial and UK Data Surgery at Bath on 9th February, therefore any comments would be welcomed by Thursday 2nd February at the latest. Please highlight any changes you have made to the spreadsheet – this will make life a little easier for comments to be assimilated.
Note: This document will be updated when any new comments are received – last updated 06/02/12.
Excellent workshop today with presentations on CERIF mappings from
CRISPool, IRIOS, RCUK, R4R, CERIFy
Presentations will be posted shortly.
Main focus of discussions/debate :
- IDs – including issue of whether UUIDs should be used or not; multiple IDs describing the same entity
- Classification schemes in use … many overlaps and interesting ideas for scheme naming conventions referencing the ‘owner’ of the classification
- splitting of date into year, month and day components as some publications only have a publication year
Keith Jeffery and Brigitte Joerg from euroCRIS gave expert advice on best practice options and an preview of some of the new stuff in v2.0 of CERIF .. including improvements in handling external IDs e.g. HESA IDs and a new structure to the CERIF-XML to reduce fragmentation … Thomas Vestdam presented on the latter.
Much to ponder and we will continue discussions on the euroCRIS CERIF-TG forum [open to all members of euroCRIS] posting main outcomes to this blog.
A massive thanks to Sunderland for hosting the event … with many attendees both physically and virtually present it went very smoothly indeed 🙂