National Library of Australia © 2011

This work is licensed under a Creative Commons A t tribution 3.0 Australia License .
Contents

Contents               2

Version Control               4

About this Guide               5

Trove Identities Manager URL               6

Background               7

Section 1 – Guide to the Trove Identities Manager               8

Key concepts               8

Login               8

Overview of the home page               9

Searching unmatched records and identities               11

To display all UNMATCHED records from one collection               11

To perform an advanced search               13

To clear the search result               14

Anatomy of a record               15

Anatomy of an UNMATCHED record               15

Anatomy of an IDENTITY record               17

How to create a new identity               20

How to match a new record with an identity record               23

How to create a new identity from splitting an identity record               27

How to undo a split identity               31

How to move a record from one identity to another identity               32

How to delete a record               35

To delete an unmatched record               35

Deleting an identity record               36

To hide an IDENTITY record from public view               36

How to unhide a record within an IDENTITY container record               39

Bulk action to create new identities               42

Section 2 – Guide to searching records in the Trove Identities Manager               45

Search tips for personal names               45

Multiple given names               45

Spelling               45

Search tips for corporate names               45

Variant names, former and later names               45

Government names               45

Use of ‘the’               46

Acronyms               46

Search tips in general               46

Contributor ID               46

Section 3 – Guide to matching records               47

Identical names               47

Names with variant life dates               47

Names consisting only of a surname or given name               47

Personas               47

Alternative names               48

Variations of the same name (personal)               48

Variations of the same name (corporate)               48

Names and multiple roles (Personal and Corporate)               49

Names with m inimal information               49


Version Control

 

Date

Version no

Revision/ Comments

November 2010

0.1

Draft issued

3 December 2010

0.2

Guide issued 

26 July 2011

0.3

Revised  to reflect the:

        change of name from PAT to TIM and

        web screens and

        some functionality

15 August 2011

0.4

Change to the TIM web addresses 

15 September 2011

0.5

Added Creative Commons CC BY  Licence

 


Introduction

The Trove Identities Manager allows for the maintenance of records in the party infrastructure. It enables authorised staff from institutions and organisations who contribute party records to the infrastructure, and staff from the National Library, to manage these records. The Trove Identities Manager:

        is web-based and available via the internet;

        was built in-house by the NLA;

        enables authorised staff in the research sector and at the NLA to manage records and identities about people and organisations using the party infrastructure;

        is separate but linked to the NLA Harvester, the application used to get party records into the party infrastructure;

        includes  changes made to it through the Australian Research Data Commons Party Infrastructure Project.

For further information about the party infrastructure please see the ARDC Party Infrastructure Project wiki at https://wiki.nla.gov.au/display/ARDCPIP

About this Guide

This Guide is designed to provide users of the Trove Identities Manager with information on how to use the manager in addition to providing guidance on when to match one party record with another. It is divided into three sections:

Section 1 – Guide to the Trove Identities Manager

This section is essentially a manual to using the manager in order to review unmatched records, view and make matches between unmatched records and identity records to obtain an NLA persistent identifier (NLA party ID), hide records from public view, and disambiguate records to correct errors.

Section 2 – Guide to searching in the Trove Identities Manager

This section contains search tips to assist users of the Trove Identities Manager to find and display party records for the purpose of matching.

Section 3 – Guide to matching

This section provides guidance for users of the Trove Identities Manager on when to match one party record with another.

This Guide assumes knowledge of the NLA party infrastructure as described in:

        Australian Research Data Commons Party Infrastructure Project Draft Consultation Document – available from https://wiki.nla.gov.au/downl o ad/attachments/16515340/ARDCPartyInfrastructureConsultationDraftv1_0.pdf

        Australian Research Data Commons Party Infrastructure Project Wiki – available from https://wiki.nla.gov.au/display/ A RDCPIP/

Trove Identities Manager URL

 

TEST

The Trove Identities Manager BETA – the Test training instance is available at:

https://www.nla.gov.au/tim-beta/app

Changes made to records in the Trove Identities Manager BETA instance will appear in the Trove TEST instance – People and Organisations view

http://trove-test.nla.gov.au/people

 

PRODUCTION

The Trove Identities Manager PRODUCTION instance is available at:

https://www.nla.gov.au/tim/app

Changes made to records in the Trove Identities Manager PRODUCTION instance will appear in the Trove PRODUCTION instance – People and Organisations view

http://trove.nla.gov.au/people


Background

Party records contain data about people and organisations. This data can be constructed and displayed   in various ways by contributors of party records to the National Library’s Trove service.  One contributor may have implemented some form of name authority control when creating the record in their database while another may not. As a result, party records from different contributors to Trove may contain variations in:

        The number and form of personal names e.g.  ‘ Smith, Christian Brynhild Ochiltree Jollie’ vs. ‘Jollie Smith, Christian’ vs. ‘Jollie-Smith, Christian B.O.’

        The use of abbreviation in personal names e.g. ‘OMalley, T’ vs ‘O’Malley, Tim’ vs. ‘O’Malley, Timothy’.

        The use of abbreviation in corporate names e.g. ‘Comm. Dept. of Environment’ vs. ‘Commonwealth Department of Environment’.

        Spelling e.g. ‘Helpmann’ vs. ‘Helpman’.

        Life dates or absence of life dates e.g. ‘1953-1987‘ vs. ‘1952-1987‘ vs. ‘b.1952’.

        The order and completeness of corporate names e.g. ‘Australian National University. Student Representative Council. Women’s Committee’ vs. ‘University Women’s Committee’.

        The use of acronyms for corporate names e.g. ‘Royal Australian Air Force (RAAF)’ vs. ‘Royal Australian Air Force’ vs. ‘RAAF’.

        The use of personal titles e.g. ‘Mawson, Douglas, Sir’ vs. ‘Mawson, Douglas’.

        The use of leading words in corporate names e.g. ‘The Wiggles’ vs. ‘Wiggles’.

        The use of the terms ‘incorporated’ or ‘proprietary limited’ in corporate names e.g. ‘Inglewood District Historical Society Inc’ vs. ‘Inglewood District Historical Society’.

These variations in the way a party record is created affect the success of automatic matching between a party record harvested from a contributor into Trove, and an identity record that already exists in Trove. Records which fail automatic matching are not assigned an NLA persistent identifier (also known as an NLA party ID) and are not made available through Trove or the NLA’s machine-to-machine interfaces (such as SRU or OAI). Instead, these records are made available for human review in the Trove Identities Manager . As the automatic matching rules are conservative and designed to avoid error, a record may fail automatic matching yet may still have a potential match that can be determined by human review of the records. This document is intended to provide guidelines for searching and determining possible matches between unmatched records and identity records in Trove using the Trove Identities Manager


Section 1 – Guide to the Trove Identities Manager

Key concepts

UNMATCHED RECORD refers to a party record that has failed automatic matching and requires human review. An unmatched record can be about a person or an organisation. An unmatched record does not have an NLA party identifier (persistent URL) and cannot be seen in Trove.

An IDENTITY RECORD is a container record that is made up of one or more party records that have passed automatic matching or human review. In an identity record, party records about the same person or organisation that have been contributed from various sources are co-located under the one NLA party identifier (persistent URL) and can be seen in Trove.

HUMAN REVIEW is the process of reviewing the UNMATCHED records for possible matches with identity records. Where a match is found, the UNMATCHED records can be matched with identity records resulting in that record being co-located under the identity container record and adopting the NLA party identifier of that identity. Where a match is not found, the UNMATCHED record can be used to create a new identity record which is automatically assigned an NLA party identifier and made available in Trove. The process of human review is achieved through the Trove Identities Manager.

Login

The Trove Identities Manager is restricted to authorised staff from contributors of party record collections and to staff at the National Library of Australia. Authorised staff are provided with a login. If your login is not entered, is incorrect or deactivated, the system will display an appropriate pop-up message.

 

Figure 1: Login screen


Overview of the home page

After successful login, the Trove Identities Manager home page will display.

Figure 2: Trove Identities Manager home page.

The L ogout button appears on the top right hand side of the home page screen beside “welcome [user name]”. Automatic logout will occur after 15 minutes of inactivity by the user.

To return to the home page at any time, click on Trove Identities Manager at the top centre of the screen.

The Unmatched records tab is the default display on the home page and lists party record collections by contributor. Only the collection(s) assigned to your Trove Identities Manager user account will be visible so if you only have one collection assigned (e.g. for your own institution’s records) then you will only see one collection. The records in the collection(s) have failed automatic matching and are waiting hand matching. The number beside each collection name indicates the number of records in the collection that are waiting hand matching. The Unmatched records tab also allows you to search the record collection(s) waiting for manual matching via simple or advanced search options. You will only be able to search and view unmatched records from the collection(s) assigned to your Trove Identities Manager user account.

There are two identities tabs on the home page. On the left hand side beside the Unmatched records tab there is a dark green Trove Identities tab and on the right hand side of the screen there is a purple Trove Identities tab. By selecting either of these tabs, you can search party identity records. These records:

        have passed automatic or hand matching

        have an NLA persistent identifier known as an NLA party ID

        act as container records for one or more party records from various contributors which are co-located within the container and share the NLA party ID.

You will be able to search and view all identity container records irrespective of the collection(s) assigned to your Trove Identities Manager user account. However, you can only action records from the collection(s) assigned to your Trove Identities Manager user account.

Records you are not able to action will appear in grey and you will be unable to select them for drag and drop.

Figure 3. Identity records tab. These unmatched records are from collections assigned to the Trove Identities Manager user: 

The Create new identity button allows a user to create a new identity container record in Trove. The identity is created by clicking and dragging a record from the user account list of unmatched records directly over the “create new identity” button.

Searching unmatched records and identities

To display all UNMATCHED records from one collection

To display all of the records from one collection that are awaiting human review you can either:

        Start from the home page and click on the collection name. The records will display under the unmatched records tab, or,

        Use the advanced search feature available under the unmatched records tab and search by contributor collection. The records will display under the unmatched records tab.

 

Figure 4: Displaying records from one collection under the Unmatched records tab.
To perform a basic search

You can perform a basic search of both UNMATCHED and IDENTITY records.

To perform a basic search of UNMATCHED records

1.      Open the light green “Unmatched records” tab on the left hand side of the screen

2.      Enter your search term(s) into the white search box

3.      Click “Search Records” or hit the enter button on your keyboard

4.      Search results will display under the “Unmatched records” tab on the left hand side of the screen.

To perform a basic search of IDENTITIES

1.      Open the purple “Trove Identities” tab on the right hand side of the screen

2.      Enter your search term(s) into the white search box

3.      Click “Search Identities” or hit the enter button on your keyboard

4.      Search results will display under the “Trove Identities” tab on the right hand side of the screen. You can expand an identity container record by clicking on “records” (number of records per identity container indicated in brackets) beneath the name entry. Identity records which do not belong to the collection(s) assigned to your user account appear with a grey bar and you will not be able to action these records, but you can still match to that Identity.

Figure 5: Displaying search results for identity records which contain the name ‘brown’.

To perform an advanced search

You can perform an advanced search of both UNMATCHED and IDENTITY records.

To perform an advanced search of UNMATCHED records

1.      Open the light green “Unmatched records” tab on the left hand side of the screen

2.      Click “advanced search” to the right of the orange “search records” button

3.      The advanced search screen will open and you can enter your search term(s)

4.      Click “search new records” or hit the enter button on your keyboard

5.      Search results will display under the “Unmatched records” tab on the left hand side of the screen.

Figure 6: Advanced seatch of Unmatched records.


To perform an advanced search of IDENTITY records

1.      Open the purple “Trove identities” tab on the right hand side of the screen

2.      Click “advanced search” to the right of the orange “search identities” button

3.      The advanced search screen will open and you can enter your search term(s)

4.      Click “search identities” or hit the enter button on your keyboard

5.      Search results will display under the “Trove identities” tab on the right hand side of the screen.

Figure 7: Advanced search of Trove identity records

To clear the search result

To clear the search result click on ‘Cancel’ or to return to the Trove Identities Manager home page, click on “Trove Identities Manager” at the top left of the screen.


Anatomy of a record

Anatomy of an UNMATCHED record

UNMATCHED records are only visible under the light green “Unmatched records” tab on the left hand side of the screen. The unmatched record consists of a number of data elements, some of which are mandatory and others optional and dependent on the information supplied in a contributor record. There are also some icons which enable you to perform various actions on the record.

 

Figure 8: Anatomy of an unmatched record

An unmatched  record may contain the following elements:

Name entry. This consists of the name parts supplied in the record by the contributor of the record. Clicking on the name entry executes a search to find identity records (those with an NLA party ID) that contain the same name. Search results will display under the purple Trove identities tab on the right hand side of the screen. 

Birth. Birth date associated with the party.

Death. Death date associated with the party.

URL .   The URL for the entry on the contributor’s website.

Contributor.   The name of the contributor

 

Details ( with drop down arrows). This may include the following information:

Biography .  Biographical information describing the party.

Occupation and subjects.  One or more occupations or subjects describing the party. Where this information is present in a record, it will display to the right hand side of the name entry.

Resources.  Details of the resources present in the record.

Alternative names .  Information on alternative names.

Identifier.  This is the local record number provided by the contributor of the record.

 

Log (#)  (with drop down arrows). This may include the following information:

Record ID. The contributor’s record id.

Contrib ID. The International Standard Identifier for Libraries (ISIL) of the contributor. This is made up of country (e.g. AU for Australia) and then National Union Catalogue (NUC) symbol (e.g. QUT for Queensland University of Technology).

Created .  Details, including date, of the creation of the record

Updated.   Date of update of the record.

 

Icons:

Magnifying glass  - “Search identities for : [the name]. Clicking on the name entry executes a search to find identity records (those with an NLA party ID) that contain the same name. Search results will display under the purple Trove  identities tab on the right hand side of the screen. 

Red Cross – “Delete this  record”. Clicking on this icon will delete an unmatched record. After deletion, a record may only be restored via a re-harvest of the record from the contributor of the record.

Tick Box “Drag to an identity  or to “Create a new identity” above. Ticking this box allows for a number of unmatched records to have identities created in batches. 


Anatomy of an IDENTITY record

Identity container records can be viewed from the purple Trove identities tab on the right hand side of the screen AND from the dark green Trove identities tab on the left hand side of the screen. The identity record may be made up on one or more records and consists of a number of data elements, some of which are mandatory and others optional and dependent on the information supplied in a contributor record. There are also some icons which enable the data administrator to take various actions on the record.

Figure 9: Anatomy of a Trove identity record

An identity container record consists of the name entry (derived from the first party record used to create the identity – in most cases the Libraries Australia Name Authority File) and below the name entry, one or more party records from one or more contributors that are co-located within the container and share the same NLA party ID.

An Identity Record may contain the following elements:

ID. The NLA party ID. This is the persistent identifier assigned to the identity when it was created, either automatically or manually.

Name entry. The subject of the identity record.  Birth / death dates may appear beside the name entry where this data is available.

Alternative names.   Other names that the name entry is known by. These are derived from all party records contributed to the identity container record.

Records (#). Clicking on this will open all records co-located under the one identity record. For example, one record from the ANBD, one from ADBO and one from EOAS. The number in brackets beside “records” indicates the number of records that are in the identity container record.

Log (#) (with drop down arrows). The log for the Identity. This appears at the bottom of the identity record and below all co-located records. Clicking on this will open the log of the audit trail of the identity record. This records all actions related to the identity from all records co-located within it.

Each record may have:

Name entry. This consists of the name parts supplied in the record by the contributor of the record. Clicking on the name entry executes a search to find identity records (those with an NLA party ID) that contain the same name. Search results will display under the purple Trove identities tab on the right hand side of the screen. 

Birth. Birth date associated with the party.

Death. Death date associated with the party.

URL .   The URL for the entry on the contributor’s website.

Contributor.   The name of the contributor

 

Details ( with drop down arrows). This may include the following information:

Biography .  Biographical information describing the party.

Occupation and subjects.  One or more occupations or subjects describing the party. Where this information is present in a record, it will display to the right hand side of the name entry.

Resources.  Details of the resources present in the record.

Alternative names .  Information on alternative names.

Identifier.  This is the local record number provided by the contributor of the record.

 

Log (#)  (with drop down arrows). This may include the following information:

Record ID. The contributor’s record id.

Contrib ID. The International Standard Identifier for Libraries (ISIL) of the contributor. This is made up of country (e.g. AU for Australia) and then National Union Catalogue (NUC) symbol (e.g. QUT for Queensland University of Technology).

Created .  Details, including date, of the creation of the record

Updated.   Date of update of the record.

 

Icons:

Magnifying glass  - “Search identities for : [the name]. Clicking on the name entry executes a search to find identity records (those with an NLA party ID) that contain the same name. Search results will display under the purple Trove identities tab on the right hand side of the screen. 

Eye – “Hide this record”. Clicking on this will hide the identity record from public view. A record hidden from public view is indicated by an exclamation mark within a red circle in place of the hide icon. Hidden records may also be unhidden.

 


How to create a new identity

Once you have ascertained that there are no suitable matches between an unmatched record and an existing identity container record, you can choose to create a new identity from the unmatched record. This will mint a new NLA party ID and associate it with the record. To create a new identity:

1.      Click and drag the record from the unmatched records tab on the left hand side of the screen and place it directly over the top of the bar “Drag record here to create new identity ” button at the top of the screen. When you have placed the unmatched record correctly, the “create new identity” button will change colour from a dark colour to orange.  You can now let go of the mouse.

Figure10: click and drag to create a new identity record.


2.      The system will display a pop-up message asking you to confirm or cancel the creation of a new identity. The pop-up message includes an option to add a note by typing it into the text box. If you choose to add a note, it will display in the audit trail of the new identity beside your username. The note will not be visible in Trove.

Figure 11: Pop-up message to confirm or cancel creation of new identity record.

3.      Click “Create new identity”.   Or click “Cancel” if you do not want to proceed to create a new identity.

4.      The system will create the new identity in the party infrastructure and assign it an NLA party ID. The new identity will appear on the right hand side of the screen under the purple Trove identities tab. The record will disappear from the unmatched records tab as it is no longer a record requiring human review.

Figure 12: The new identity is created.


How to match a new record with an identity record

Once you have ascertained (via searching etc) that there is a match between an unmatched record and an identity record, you can match the unmatched record with the identity record. Through this process, the unmatched record will be moved into the identity container record. The record will be co-located with records from other contributing collections within the identity container record and will retain its local identifier in addition to being assigned the NLA party ID of the container record. 

To merge an unmatched record with an identity:

1.      Ensure the unmatched record you want to match is displayed under the green unmatched records tab on the left hand side of the screen.

2.      Bring up the candidate identity container record under the purple Trove identities tab on the right hand side of the screen, by either clicking on the name in the unmatched record or searching in the Trove Identities.  In the Identity record click on “records” to expand the view to all records within the identity container record.

Figure 13: Displaying the candidate identity record.


3.      Click on the top area of the unmatched record and drag it over the top of the identity container record to match it to the existing identity.

Figure 14: Drag and drop to match the unmatched record with the identity record.


4.      The system will display a pop-up message asking you to confirm or cancel the “Move record”. The pop-up message includes an option to add a note. If you choose to add a note, it will display in the audit trail of the identity container record beside your username. The note will not be visible in Trove.

Figure 15: Pop-up message to confirm or cancel match of unmatched record with identity record.

5.      Click “Move record”.

 


6.      The system will move the unmatched record to the exiting identity. This will co-locate the unmatched record within the existing identity and its party identifier. The unmatched record will disappear from the unmatched records tab as it is no longer a record requiring human review.

Figure 16: The matched record is co-located within the identity record.


How to create a new identity from splitting an identity record

You may need to split a record from an identity container record and create a new identity record with it. This is called disambiguation and allows you to correct a mistake whereby a party record has been incorrectly matched with an identity record. This process will:

        create a new identity from the split record

        create a new identity from the record(s) remaining in the old identity

        retain and notate the old identity for viewing in the Trove Identities Manager and via the persistent identifier but it will no longer be indexed and searchable via Trove

To split a record from an identity:

1.      Search for the identity and display it under the purple Trove identities tab on the right hand side of the screen.

Figure 17: Displaying the identity you need to splt.


2.      Expand the identity record to view all records within the container and select the record you wish to split from the identity. Note: you will only be able to action records that belong to the collection(s) assigned to your Trove Identities Manager user account.  Records outside of your collection are indicated by a grey bar and may not be actioned.

3.      Click and drag the record up to the “Drag record here to Create new identity” button on the top of the screen. The “Create new identity” button will turn orange when you have placed the record directly over the top of it. Now let go of your mouse.

Figure 18: Drag and drop to split the identity record and create a new identity.


4.      The system will display a message asking you to confirm or cancel the creation of the new identity. There is also an option to add a note by typing into the blank text box. Notes are visible in the audit trail and will not display in Trove.

Figure 19: Pop-up message to confirm or cancel creation of new identity record.

5.      Click “create new identity”.

6.      The system will:

        Create a new identity record from the record you split from the old identity. This new identity record will have a new NLA party ID and be available in Trove. The record will have a related heading containing information about the identity split and an audit trail containing information about actions on the identity. Information under the “related” and “audit” headings will not appear in Trove.

        Create a new identity from the remaining record(s) in the old identity. This new identity will also have a new NLA party ID and be available in Trove. It will have a related heading containing information about the identity split and an audit trail containing information about actions on the identity. Information under the “related” and “audit” headings will not appear in Trove.

        Alter the old identity record so that the only information contained within it will be a related heading containing information about the identity split and an audit trail containing information about actions on the identity. This identity will NOT be available in Trove.

In the example below:

The original party id (with the 2 names) was nla.party-1016827

The ‘split’ record (Woodward, Guy) is now nla.party-1476795

The remaining record (Woodward, Judy (1916-1971)) is now nla.party-1476796

The original party id nla.party-1016827 is now the record with audit information for the split.

Figure 20: The identiite records  after the split.


How to undo a split identity

If you have split a record from an identity in error, you can correct the mistake by merging the identities back together. To do this:

1.      Search for the new identity and display it under the green Trove identities tab on the left hand side of the screen.

2.      Execute a search to display the new identity created from the old identity under the purple Trove identities tab on the right hand side of the screen.

3.      Click on the record you wish to move from under the identity on the left hand side of the screen and drag it over the identity you wish to merge it with on the right hand side of the screen. Let go of the mouse.

4.      The system will display a message asking you to confirm or cancel the record move. Click “move record”.

5.      The system will merge the record from the left of the screen with the identity on the right hand side of the screen.

6.      The system will annotate the old identity on the left hand side of the screen with “related” and “audit” information. This record will not be available in Trove.


How to move a record from one identity to another identity

You may need to move a record from one identity container record to match it with another identity container record. This process will:

        create a new identity from the split record

        create a new identity from the record(s) remaining in the old identity

        notate the old identity and make it unavailable for viewing via Trove

To split a record from an identity container record:

1.      Search for the identity record you wish to split and display it under the purple Trove identities tab on the right hand side of the screen. Search for the identity record you wish to match the split record with and bring it up under the green Trove identities tab on the left hand side of the screen. You may also bring both records up on each side of the screen as per the screenshot below.

Figure 21: Displaying the identity record you wish to split.

2.      From the green Trove identities tab on the left hand side of the screen, click on the record you wish to split from the identity. Even if there is only one record within the identity record, you must click on the record not the identity container.

3.      Drag the record and place it on top of the identity container record you wish to match it to and let go of your mouse. The system will display a pop-up message asking you to confirm “move record”. The pop-up message includes an option to add a note by typing into the text box. Notes will appear in the audit trail of the identity record. Notes will not appear in Trove.

Figure 22: Moving the Identity to another Identity.

4.      Click “move record”.

5.      The system will:

        Co-locate the moved record within the identity record you moved it to. This record will be assigned the NLA party ID of the identity container record.

        Add a notation to the identity record the moved record was split from referring to the new identity record (see screenshot below).

Figure 23: Result of split identity.


How to delete a record

An UNMATCHED record may be deleted. Reasons for deleting an unmatched record may include that it was not intended to have an NLA party ID allocated for that record, or that information in the record needs to be corrected by the contributor and then re-harvested by the NLA before it can be matched with an identity record (e.g. the birth date on the unmatched record is incorrect). 

Unmatched records can only be deleted individually. Once deleted, there is no option to undelete the record. However, the record may be re-harvested into the party infrastructure. 

To delete an unmatched record

1.      Search for the record and display it under the green unmatched records  tab on the left hand side of the screen.

2.      Click the “delete record” icon (red cross) on the record you wish to delete.

Figure 24: Deleting a record.

3.      The system will display a message asking you to confirm or cancel the record delete.

Figure 25: Pop-up message to confirm or cancel record delete.

4.      Click the “delete record” button.

5.      The system will delete the record. It will no longer display on the screen and will not be searchable in the Trove Identities Manager.

Deleting an identity record

Records within an identity container record may not be deleted. However, these records can be hidden from public view so that they do not appear in Trove or its machine-or-machine interfaces.

To hide an IDENTITY record from public view

1.      Search for the identity container record and display it under the purple Trove identities tab on the right hand side of the screen.

2.      Select the individual record from within the identity container record that you wish to hide from public view. Click the “hide from public view” icon (eye).

Figure 26: Hiding an identity from public view.

3.      The system will display a message asking you to confirm or cancel hide record. The message includes an option to add a note. Notes added with the hide record action will appear as “record deleted” in the audit trail of the identity container record and not in the audit trail of the record within the identity that you are hiding. Notes will not appear in Trove.

Figure 27: Pop-up message to confirm or cancel hide record from public view.

4.      Click “hide record”.

5.      The system will hide the record from public view. The record will not be visible in Trove or its machine-to-machine interfaces. Where the hidden record was the only record in the identity container record, the identity record itself will not be visible in Trove. Where the hidden record was one of several records within an identity container record, all information from the hidden record will be removed from the identity container record.

6.      The hidden record will still display in the Trove Identities Manager and remains searchable within this interface. However in place of the hide icon (eye), you will see a red exclamation mark within a circle. If you put your mouse over this icon a pop-up message will indicate that if you click that icon, you can unhide the record from public view.

Figure 28: Records hidden from public view have an exclamation mark symbol.

How to unhide a record within an IDENTITY container record

Once hidden from public view, there is an option to unhide a record from an identity container record. To unhide a record from an identity container record:

1.      Search for the record and display it under the purple Trove identities tab on the right hand side of the screen.

2.      Select the hidden record from within the identity container record that you wish to unhide and click the red exclamation mark within the circle.

3.      The system will display a message asking you to confirm or cancel unhide record. The message includes an option to include a note. Notes will appear in the audit trail of the identity container record and not the audit trail of the record within the identity container record. Notes will not appear in Trove.

Figure 29: Pop-up message to confirm or cancel unhide record.

4.      Click “unhide record”.

5.      The system will restore the record to public view in Trove and its machine-to-machine interfaces. In the Trove Identities Manager, the hide icon (eye) will replace the exclamation mark on the unhidden record. The “unhide” action will be recorded in the audit trail of the identity container record. This audit trail is visible in the Trove Identities Manager but not in Trove.

Figure 30: Records unhidden from public view now have the  hide (eye) icon and details recorded in the audit trail.


Bulk action to create new identities

The bulk action feature allows you to select multiple unmatched records and create new identity records from them. This feature should be used with caution and only when you are certain there are no possible matched between the unmatched records and identity records.

To bulk action the creation of new identity records from unmatched records:

1.      Open the unmatched records you wish to bulk action under the “unmatched records” tab on the left hand side of the screen.

2.      Click in the small white square box beside  the name entry in each record to select the records (note: you can click in this box again to de-select a record).

Figure 31: Selecting records for bulk create new identities

3.      Click the orange “Create new identities” button towards the top of the green “unmatched records” tab.

4.      The system will display a “Create identities” message asking you to confirm or cancel the creation of identities for the number of  unmatched records you ticked. There is an option to add a note. Information in the notes field will only be visible in the Trove Identities Manager and not in Trove.

Figure 32: Pop-up message to confirm or cancel creation of new identity records.

5.      Click “create identities”.

6.      The system will create a new identity for each of the records you have selected. These records will be displayed under the purple identities tab on the right hand side of the screen. The new records will have disappeared from under the green “new records” tab.

Figure 33: Bulk identity records created.

 

 


Section 2 – Guide to searching records in the Trove Identities Manager

Search tips for personal names

Multiple given names

Some personal names may include more than one given name. When searching this type of record via a ‘find similar’ search (i.e. clicking on the name entry of the unmatched record to generate a search result), a zero result may be returned. This is because the system is looking for an identity record which contains all of these names and contributors create name entries differently. Try narrowing the search to surname and one given name or surname and initial.

e.g. A find similar search for ‘Lindsay, Norman Alfred Williams’ could be narrowed to ‘Lindsay, Norman’ to maximise the search result.

e.g. ‘Melba, Nellie, Dame’ could be narrowed to ‘Melba, Nellie’. 

Spelling

Some names can be spelt differently and this may result in a zero search result. If you think the name may have been spelt differently, try searching for it using a different spelling.

e.g. for ‘Helpmann, Robert’ try searching for ‘ Helpman, Robert’.

e.g. for ‘Cook, Ann’ try searching for ‘Cook, Anne’.

You could also try searching for the given name initial e.g. ‘Cook, A’. This is likely to broaden the search results significantly but the name entry may have been recorded in this way by a contributor.

Search tips for corporate names

Variant names, former and later names

Corporate name entries can be constructed and displayed   in various ways. Examples include using abbreviated or different spellings of names, using acronyms, placing the name parts in a different order. Try a range of search techniques to cover the variations.

e.g. ‘Human Nature Musical Group’ could have a variant name of ‘Human Nature’.

e.g. ‘ Inglewood District Historical Society’ could have variant names of ‘Inglewood District Historical Society Inc.’ and ‘Inglewood & District Historical Society Inc.’

However be aware that variant names may not be just a different form used by a contributor, but could be a Former and/or Later name. These names are bodies in their own right and should have a separate identity record in Trove.

Government names

The form of names used for Government Departments and bodies can vary considerably from contributor to contributor. Some may use abbreviations such as ‘gov’ or ‘govt’ or ‘dept’ and others may spell the name out in full such as ‘government’ or ‘department’. Ordering of name parts may vary considerably. 

When reviewing and matching government departments search for names using variant terms. However be careful with former and later names, as both a former and later names should have an identity in its own right.

e.g.  ‘ Queensland Department of Employment, Economic Development and Innovation ’ (later name) and ‘Queensland Department of Primary Industries and Fisheries’ (former name). In Trove, each of these names would be a separate identity record with its own NLA party ID.

Use of ‘the’

Some corporate name entries start with an article such as ‘The’. However, where the contributor of a record has not described the name entry with ‘the’, a zero search result will be returned. Try searching for the corporate name again but without the word ‘the’ at the start.

e.g. try searching for ‘Wiggles’ rather than ‘The Wiggles’.

Acronyms

If a corporate name includes an acronym and a search by double clicking returns a zero result, try a search of the name, or part of the name, without the acronym. Also try searching for just the acronym.

e.g. for ‘Royal Melbourne Institute of Technology (RMIT)’ try searching for ‘Royal Melbourne Institute of Technology’ or ‘Royal Melbourne Institute’ or ‘RMIT’.

Search tips in general

Contributor ID

Record ID (the ID used by the contributor of the record for that record) can be used to search for an identity. This can be a useful search if a contributor’s data has been re-harvested and has not resulted in an automatic match. This can happen if a contributor has made a change to the name for that record ID in their system.


Section 3 – Guide to matching records

Identical names

Where an unmatched record contains a name entry that matches an identity record exactly and there is not enough information to determine with certainty if they are the same or a different person or organisation:
              1. Do not match the records. Instead, create a new identity from the unmatched record.
              2. Add a note if we have some useful information to help distinguish between the identities.

Names with variant life dates

There may be a conflict with life dates between an unmatched record and an identity record and where the data matches in other respects e.g. Sutherland, Jane, 1855-1928 (identity record) and Sutherland, Jane, 1853-1928 (unmatched record). In this case:

        Conduct some research to see if the correct life dates can be determined.

        If the correct date cannot be determined: match the unmatched record with the identity record in the usual way and include relevant information in the note field.

        If the correct date can be determined, and the unmatched record is incorrect: do not match the record and instead refer the record back to the record contributor for correction and re-harvest.

        If the correct date can be determined, and the identity record is incorrect: match the unmatched record with the identity record in the usual way and include relevant information in the note field. Then advise NLA staff of the party ID.

Names consisting only of a surname or given name

There may be unmatched party records which consist on only of a surname or   a given name. In this case:

        Check the contributor’s   source data to determine if additional names   should be   included in the   record.   There may have been a technical problem in which a name part was ‘lost’ in the transformation to display it in the party infrastructure. Or, additional name parts may not exist in the contributor’s data.

        Where the contributor’s source data includes additional name parts, contact the NLA for resolution and re-harvesting of the record.

        Where the additional name parts do not exist in the contributor’s data, refer this back to the contributor for correction or annotation and delete the new record from the Trove Identities Manager.

        For   names genuinely consisting only of a single name (eg. MumShirl - http://nla.gov.au/nla.party-783718 ) match or create new identity records   in the usual way.  

Personas

Where there is a record for an identity and another record for a distinct and separate persona of that identity, an identity record should be created for each and each identity can link to the other. For example Pamela Teresa is a persona of Jack O’Hagan. In the party infrastructure, there is an identity record for Jack O’Hagan - http://nla.gov.au/nla.party-557483 - and an identity record for Pamela Teresa - http://nla.gov.au/nla.party-556440 Each identity record links to the other under ‘related people and organisations’. Ideally, each would be linked to the other under ‘also known as’. Another example is Barry Humphries - http://nla.gov.au/nla.party-466079 - whose personas include Everage, Edna (Dame) and Patterson, Les (Sir). If a separate record for Patterson, Les (Sir) was provided, we would create a new identity from this record and the new identity would link back to the Barry Humphries identity.

Alternative names

The name entry for some records may be an alternative name for an identity. Persons with alternative names include married women and   names changed by deed poll. Where the alternative name is not being used concurrently, match the record in the usual way.

Example: Anderson, Maybanke (1845-1927) - http://nla.gov.au/nla.party-466159

Maiden name: Maybanke Susannah Selfe

1st married name Maybanke Wolstenholme, or Mrs Edmund Wolstenholme

2nd married name Maybanke Anderson or Mrs. Francis Anderson

Exception: Alternative names may also be used for various reasons, eg. one name used in a specific field and another one used as a private person or in another field of interest. These "alternative" names would be treated as different identities, same as Pseudonyms, and hopefully with alternative names links etc.

Variations of the same name (personal)

Records for the same person with variations of spelling, initials etc should be matched, as long as it is certain they are the same person.

Records for the same person that are displayed in a different order   should be matched, as long as it is certain they are the same person.

Examples of records with name variations for the same person that should be matched:

‘Don Walker’ with ‘Donald Hugh Walker’

‘OKeefe, John’ with ‘ O'Keefe, Johnny’ and ‘O'Keefe, Johnnie’

‘Barnes, James’ with ‘ Jimmy Barnes ‘

‘Gill, Samual Thomas’ with ‘Gill, S. T. -   G, S. T.’

‘Robert Helpmann’ with ‘Sir Robert Helpmann’

‘Young, John Paul’ with ‘John Paul Young’

Variations of the same name (corporate)

Records for the same corporate body with variations of spelling, acronym etc should be matched, as long as it is certain they are the same corporate body, and not former and later names of that body.

Records for the same corporate body that are displayed in a different order   should be matched, as long as it is certain they are the same corporate body, and not former and later names of that body.

Records for former names and records for later names of a corporate body should not be merged. Where a corporate body has changed its name,   there should be an identity record created for each name.

For example:

‘James Cook University of North Queensland’ changed its name in 1997 to ‘James Cook University’.   Both names   should appear as distinct identities in Trove.

Names and multiple roles (Personal and Corporate)

Some persons will have a name entry for them as a person, but their name may also feature in an entry for a corporate body or group. These should be added to Trove as separate identities. The identities may be linked under ‘related people and organisations’ where a record provides this data.

For example:

Olive Pink - http://nla.gov.au/nla.party-473797

Olive Pink Society - http://nla.gov.au/nla.party-546368


Names with minimal information

Some records contain very minimal information. The minimal requirements for a record to be harvested into the party infrastructure are that it has a name entry and a local record identifier (plus some record maintenance information). These records can be processed (i.e. matched or new identity created) in the usual way.