ADT Future Paper for CAUL Executive

ADT Future Paper for CAUL Executive

Version of 8 March, 2010, updated 16 March 2010

Prepared by Kate Watson, CAIRSS Program Manager, with the CAIRSS Steering Committee

1. Recommendation

CAUL Executive is requested to approve the following recommendations:

1. NLA Trove be the new thesis portal.
2. No new content be added to the ADT site after the end of March 2010.
3. CAIRSS set up a temporary holding tank to assist CAULmembers to move their content from the ADT site into their institutional repository.
4. Institutions migrate all existing ADT content to either the institution’s repositoryor the CAIRSS holding tank by end September 2010.
5. The CAUL ADT service, branding and current webpage be decommissioned by October 2010.
6. CAUL create an ADT information webpage describing the ADT service and contribution details which would link to the NLA Trove thesis facet
7. The CAIRSS holding tank be decommissioned by February 2011. Institutions to migrate content from here to their repository by January 2011.
8. The actions and timeframes listed below in section 6 be accepted.

2. Background

In 2009the CAIRSS Steering Committee requested CAIRSS Central undertake an analysis of possible software options for the future of the Australasian Digital Theses Program (ADT).

This was done in the knowledge that:

• the existing ADT VT-ETD software and ADT Central software were no longer supported by a vendor and were unable to be moved from UNSW to another institution
• UNSW maintenance of ADT was scheduled to end in 2009
• many CAUL institutions were now including their theses content in harvested institutional repositories

Theanalysis was presented to the CAIRSS Steering Committee which requested further exploration and information on:Liaison between CAUL/NLA on possibility of ADT service being delivered as part of NLA services (such as Trove and Australian Research Online).CAIRSS subsequently approached the National Library to discuss a possible proposal.

3. Proposal

TheNLA hasoutlined how itcould provide a thesis “portal” (to replace the existing ADT) using the new NLA Trove product(which already harvests from Australian institutional repositories via Australian Research Online (ARO with minor modifications.

Thesis records will appear in both ARO and Trove until the ARO interface is migrated by the NLA to Trove (date not yet determined).

Within this proposal are some changes in how the service content, function and deliverywill differ to the existing ADT:

3.1 The NLA option would be more inclusive of all Australian theses content (as NLA systems currently define ‘theses’ only). Therefore content will include all theses content (and not just PHD and Masters by Research as displayed in the current ADT).
Individual institutions would need to consider local policy or user assistance implications of the more inclusive definition. They would also need to consider whether or not the more inclusive definition would lead to expectations of inclusion which the institution would or could not support. (For example, the inclusion of honours theses in Trove by some institutions may lead to pressure on other institutions to include such theses in Trove when the institution has chosen not to do so for policy or resourcing reasons)

3.2 NLA will only be able to harvest from institutional repositories, not from the unsupported ADT VT-ETD softwaresomeinstitutions are still using (16 in total)[1].These institutions will need to migrate their theses content from ADT VT-ETD software to their institutional repository(as 18 CAUL institutions have already done).

CAUL will need to make a recommendation to these 16 institutions to move their theses content from the VT-ETD software to their institutional repository by a set date.The CAIRSS Technical Officer can assist institutions with these migrations.

ARO harvests all but nine universities. Of the latter, four are currently in test for inclusion.
Given pressing ERA timelines, as an interim measure, the CAIRSS Technical Officer can offer a temporary store of institutions VT-ETD theses content in a CAIRSS ‘Fedora holding tank’ which can be harvested by the NLA while the institution makes the necessary changes and plans to include theses content in their repository. (Note: this content would then need to be moved into the institution’s repository before the end of the CAIRSS Project). It is quite possible that this approach may assist such institutions with their thesis migration plans.

3.3 The current ADT service is ‘Australasian’ and includes a harvest of New Zealand (NZ) theses content from 7 NZ institutions (5 from institutional repositories and 2 from VT-ETD software). With this proposal the NLA would prefer to harvest a single file of NZ thesis records directly from KRIS – the Kiwi Research Information Services hosted by the National Library of New Zealand. This would need to be communicated from CAUL to CONZUL.

Currently, Trove displays the ARO brand as a contributing source. Information about the originating institution is displayed with each record if provided by the institution.

3.4 The NLA Trove proposal would display the content as a ‘thesis’ facetin Trove. CAUL will need to make some decisions regarding ADT branding requirements. If a decision to keep the ADT branding is made, CAUL would need to host an ADT information webpage describing the ADT service and contribution details which would link to the NLA Trove thesis facetfor search/browse functionality of theses content (as no ‘ADT’ label will used by NLA in Trove). Alternatively the ADT brand could be retired, with the service marketedas a portal of Australasian theses hosted by the NLA.

3.5 Institutions will continue to undertake the creation of theses metadata and the digitisation of theses. That is, all the steps which make up the contribution pathway prior to OAI harvesting will not involve the NLA.At the point of harvesting, the ARO Technical Manual will apply. Reference queries arising from this new approach will need to be handled by CAUL (previously handled by Tom Ruthven at UNSW and currently handled by CAIRSS with current ADT software). CAIRSS will be able to continue to for the duration of the CAIRSS project (ending March 2011).

The National Library will require a CAUL email address for forwarding thesis enquiries it receives via Trove.

3.6 The National Library will expose thesis content to Google and other search engines via a Trove sitemap (already available) and open harvesting protocols (yet to be delivered).

4. Communications

A decision on the future of ADT in line with this proposal, will need to communicated tothe ADT Policy and Reference Group.

Further, any decision will need to becommunicated to all CAUL and CONZUL institutions with an outline of what the changes will mean to the institution, and to the end-user.

The National Library could provide a brief flyer on how to find theses in Trove if required.

5. Costs

There is no cost if ADT branding is restricted to the information page hosted by CAUL. ARO should continue to be used to quantify individual institutional metrics until it is migrated to Trove. For metrics specific to theses, please see the note on encodings in the technical details section.

6. Actions & Timelines

Actions / Responsibility / Date
Decision on proposal made by CAUL Executive.
If yes – continue below.
If no – CAUL make alternative plans for ADT future. / CAUL Executive. / March2010
CAUL Executive communicateADT future decision to ADT Policy and Reference Group. / CAUL Executive. / March 2010
CAUL Executive communicate ADT future decision to all CAUL institutions. / CAUL Executive. / March2010
CAUL Executive communicate ADT future decision to CONZUL. / CAUL Executive. / March2010
NLA provides information for searching Trove and ARO to select theses / NLA. / March – August 2010
CAUL Executive communicate the planned withdrawal of ADT VT-ETD software to the 19 institutions that do not yet have their theses content in their repository.
Advise:
(a) Date VT-ETD software will no longer be harvested for ADT (i.e.date for institutions to either have theses content migrated to repository or added temporarily to CAIRSS holding tank).
(b) Date CAIRSS holding tank will be turned off (and therefore date theses content must be in repository to be included inARO/Trove). / CAUL Executive. / March 2010
CAIRSS set-up CAIRSS holding tank. / CAIRSS. / March-April 2010
CAIRSS test feed from CAIRSS holding tank into ARO for Trove theses view. / CAIRSS.
NLA. / March-April 2010
CAIRSS contact 19 institutions who do not yet have their theses content in their repository and:
(a) assist them to migrate their theses content directly into their repository
or
(b)move their theses content into the CAIRSS holding tank (temporary). / CAIRSS.
CAUL institutions. / March – August 2010
CAIRSS and NLA communicate with NZ contacts to test harvest from KRIS into ARO for display in ARO and Trove. / NLA.
CAIRSS.
NZ. / March – April 2010
CAUL create ADT information webpage describing the ADT service and contribution details to link to the NLA Trove thesis facet for search/browse functionality of theses content. / CAUL. / April-May 2010
Cut-off date for institutions to either have theses content migrated to repository or added temporarily to CAIRSS holding tank.
(I.e. Date VT-ETD software no longer harvested). / Cut-off date for CAUL institutions. / 31 September 2010
New NLA Trove-based ADT service launched.
Existing ADT service, branding and webpages decommissioned. / NLA.
CAUL.
CAIRSS. / October 2010
Cut-off date for institutions to move their theses content from the CAIRSS holding tank (temporary) into their repository. / CAIRSS.
CAUL institutions. / January 2011
CAIRSS holding tank turned off. / CAIRSS. / February 2011

* These timelines have been developed keeping in mind the busy ERA preparation and submission periods for repository managers from Jan 2010 – June 2010.

CAIRSS sites harvested from IR: 18
- BOND
- SCU
- WOOL
- CURTIN
- UMELB
- UADELAIDE
- USYD
- UTS
- JCU
- QUT
- UTAS
- VU
- MACQ
- MONASH
- BALLARAT
- UNSW
- UNISA
- UWS
CAIRSS sites not yet in ADT: 5
- NOTRE DAME
- CSU
- CDU
- USC
- UNE / CAIRSS sites still harvested from VTEDT (and will need to move harvest and possibly contentto IR): 16
(incl newcastle which has a harvest from both IR and VTEDT)
- ECU
- UWA
- ANU
- FLINDERS
- GRIFFITH
- MURDOCH
- USQ
- ACU
- UCANB
- DEAKIN
- RMIT
- UQ
- CQU
- LaTROBE
- SWIN
- NEWC (has both VTEDT and IR)
=39 sites

7. Technical Details of Proposal

Behind the scenes, Trove would be harvesting records from institutional repositories, and mapping them to the thesis facet in 'Format’ from CAUL contributors.The idea is to let Trove do the work and encourage users to see theses as just one of the resource types that are discoverable at the NLA.

ADT pages at CAUL would link directly into Trove using a URL similar to this: The thesis view can also be reached by drilling-down from Trove to via aforthcoming Research tab/view (yet to be scheduled).

University libraries also have the option of submitting thesis records to Trove via the Record Import Service in Libraries Australia. These records will not appear in ARO.

Note that the encodings supported by Trove will allow a thesis to be found only if the word ‘thesis’ is included in the Dublin Core resource type or the MARC 008/24-27 field or Notes fields contain the words ‘thesis’ or ‘dissertation’. The majority of ADT records are not encoded uniformly.In order to ensure reliable facet behaviour and metrics in Trove and ARO, the NLA would be able to add the word ‘thesis’ during ingest if it does not appear in the resource type.

All content will be sourced via OAI-PMH from institutional repositories (online) or via the ANBD (offline). Radio buttons to toggle showing online, offline and Australian only contentare already available in Trove. Ability to limit by contributor (institution) will be built in when ARO is migrated to Trove.Until this occurs, the ability to limit by institution is available in the advanced search option in ARO (see screen two below).

The following screenshot shows some options:

Screen 1 – thesis facet in Trove

Screen 2 – search for thesis by institution

1

ADT Future Paper for CAUL Executive

[1] * Of the 39 CAUL institutions, 18 are already being harvested by ADT from their institutional repository, 16 are still being harvested from the unsupported VT-ETD software, and 5 are yet to submit to ADT (although when they decide to a harvest direct from their institutional repository can be easily set up).