Pure release 4.9.0

A number of new features are included in 4.9.0. The most important ones are outlined below. For full details, please obtain the release notes.

Pure's REF2014 module allows a complete return to be made to HEFCE's 2014 Research Excellence Framework programme. The module introduces special, REF-specific content types, relations, classifications, user-roles, reporting capabilities, functionalities, special security measures, and data export options, which together allows institutions to assess and select Staff, Outputs, and other content to be submitted and to make the final submission.

This new module is available for the first time with this release of 4.9.0. Pre-release versions have however been available to the members of the UK Pure User Group for a longer period of time leading up to this release.

The early availability of the REF module reflects our policy to design and develop Pure's REF functionality continuously although HEFCE's requirements are not yet final, thereby accepting additional costs related to several and frequent refactors over time as HEFCE's guidances have been made available. One benefit is that Pure-owners will have fully working REF capability in their Pure systems from a very early point, which offers reassurance that it will indeed be available on time as well as the opportunity to do mock return exercises several years before census date. Another benefit is that staff with RAE experience, REF panel staff members, and other domain experts from the UK Pure User Group have been participating in every refactor of Pure's REF module since its first inception. Pure's REF module is the result of a joint, user-driven effort among owners as much as a product from Atira's design team, which should guarantee easy adoption by REF-responsible users in new Pure-using institutions as well as in older ones.

Users can now use a new improved version of the report module. The functionality in the previous and new version is almost the same, but the user interface in the new version has changed a lot and the user friendliness has improved radically. Since we still work on a number of user interface improvements, both the previous and the new interface will be available in 4.9.0, but we do plan to deprecate the old interface from the release of version 4.10.0.

In addition to the immediate gain in usability, the new user-interface will allow new reporting functionality to be added, which would not easily have been incorporable in the old interface. The new user-interface will better serve as platform for the planned growth in reporting functionality over the coming years, and therefore was a necessary improvement at this time.

Users are now presented with both the name and a description of an online import source when they import publications. A help text for available search operators is displayed on the import search page, and a spinner now indicates when a search is being submitted to the online source. Results with duplicates are indicated more clearly, and a link is provided to jump straight to an editor of the duplicate. Also, the final import screen which displays the finished result has been redesigned to more clearly present the matches, and it includes more information about each match.

Scopus, WorldCat, and JournalTOCs. Scopus and WorldCat requires subscriptions while JournalTOCs is a free source.

The top navigation now presents a "big menu" when hovering over a link. This big menu provides direct access to all types of content and a button to create that content. The left hand menu of the content dashboards has also been updated to present the selected filter setup (e.g. editable content) more clearly.

The filter bar below the search field has been re-designed to a breadcrumb layout that indicates which filters have been added instead of presenting all available filters at once. Filters can be added as needed, and easily removed individually from the filter bar. Also, the content listing below the filter bar will fade out slightly and present a spinner when content is being retrieved from the server as a result of changes to the filter settings, which leaves the user aware of currently ongoing activity.

RoMEO information is now displayed directly above the list of documents in the publication editor. The information is initially displayed in a compact form, but can be expanded to provide full details. Publications that may be considered Open Access but yet have no full text, are presented as a personal task in the right hand side of the workspace. RoMEO colours are configurable from the Administrator dashboard under "System Settings", "RoMEO".

A new "Publications with documents" view is available for relevant editors in Pure. It presents statistics such as the number of publications with and without documents, RoMEO colours, and document visibility. This new view is accessible from the left hand menu in the workspace when publications are listed.

One organisation can now take another. This new feature allows an administrator to let one organisation take over another. When doing that, all content (e.g. publications) formerly associated with the overtaken organisation will be associated with the overtaking organisation but without loss of the information, that the overtaken organisation once existed and that certain content was produced there. If Pure did not preserve this information, a publication written in 2002 might belong to an organisation founded in 2010, for example, without any clue to how that could logically have happened. Another argument for preserving historic information is the need for retrospective statistics and other reports.

Information about overtaken organisations is displayed correctly in Pure, on PurePortals, and on websites retrieving information from Pure's Web Service API. It is also possible to report correctly and in full on overtaken organisations and their content.

A new "technical administrator" role is now available in Pure as standard. The "Administrator" role available in Pure until now granted access to all technical administration as well as global content management. The addition of a purely technical admin role makes it possible to grant access to IT teams and other technical type staff with access to Pure without allowing content management, reporting, or similar data access privileges.

The Web Service element "OrganisationType.owner" has been deprecated from version 4.9.0 and should no longer be used as the Web Service schema will not hold the element.

In previous versions, Pure used ActiveMQ JMS (Java messaging service) for passing indexing and file synchronisation messages. From version 4.7.0, file synchronisation was handled entirely by a cron job, and from 4.9.0 indexing is no longer handled in JMS and JMS has been removed from Pure from 4.9.0. To that, Pure no longer used Atomikos for JTA transaction management. Standard JDBC database transaction management is used instead. This means that all installations will need to alter their data source configurations to use Apache commons datasource instead of the Atomikos aware datasources.

Configuration of the datasources for each supported database (Microsoft SQL Server, Oracle, and PostgreSQL) is described in the updated installation manual.

Microsoft released Internet Explorer 9 on 14 March this year. It was very close to our release of Pure 4.9.0, which means that Internet Explorer 9 has not been fully tested with Pure. This version of Explorer is therefore not officially supported by Pure. We expect add official support in one of the following 4.9.X releases.