Frequently Asked Questions
Cataloging and Metadata Standards
- Does CollectiveAccess support DublinCore?
Yes. An installation profile configuring CollectiveAccess as a DC-compliant cataloguing system is provided in the software distribution.
- Is CollectiveAccess SPECTRUM compliant?
Nick Poole of the CollectionsTrust has supplied the following answer, which complies with the requirements of the Trust:‘SPECTRUM Compliance’ is a trademark of the Collections Trust, as part of the paid SPECTRUM Partners Scheme. CollectiveAccess is an open-source platform and is not able to participate in the Scheme. CollectiveAccess provides a SPECTRUM installation profile that should be usable ‘out of the box’ for most SPECTRUM-seekers, but is not able to state that the system is ‘SPECTRUM Compliant’.
- I want CollectiveAccess to support <MY REQUIRED STANDARD HERE>. How do I do this?
Implementing support for a data standard – either a recognized standard or one of your own design – is an involved undertaking. You have to consider many details and perform a lot of iterative testing. Fortunately, CA ships with a selection of profiles that can be used as-is or as a starting point for your own setup. For the more popular standards (like DublinCore and PBCore) you should be able to use pre-built installation profiles included with the software. For data standards of your own design, or for projects to which no standard clearly applies you should be able to use one of the included pre-built installation profiles as a starting point. In general you should never' have to start a implementation profile from scratch.
- I need to write a profile to support <MY REQUIRED STANDARD HERE>. How do I get started?
Your first step should be to locate the existing profile that best matches your requirements. Then make a copy of it and start editing. To create the largest and broadest collection of profiles possible, we are encouraging our users to share their profiles via the Configuration Library section of the CollectiveAccess web site. Be sure to look both at the profiles included with the CA installer (in install/profiles) as well as those in the Configuration Library. Next read the Building System Installation Profiles manual.
Once you've found the best match for your project and read the manual, the next steps are:
- Carefully consider the data requirements of your project. Enumerate the data you need to record - the "fields" - and define the required limits on each. By limits we mean things like type ("this field contains a date"), value boundaries ("the date should never be before 1900"), value enumeration ("the value should be one from this proscribed list"), specificity ("this field only applies to objects that are postcards") and repeatability ("each record can take up to 10 date fields").
- Decide where these fields (or "attributes" in CA parlance) live in the CA data model. The CA model includes structures for collection objects and entity (aka "people"), geographic place, and collection (arbitrary groups of objects) authorities. It also includes a generic authority structure - what we call occurrences - that can be used to model any number of additional authorities. Occurrences are typically used to maintain authorities for events, exhibitions and productions (anything that's not an entity, place or collection). The process of linking your attributes to the data model should make it obvious what sorts of occurrences you need (if any).
- Figure out what types of objects, entities, places and collections will be needed. Determining the type lists can be a little tricky: attributes can be restricted to a specific type (implementing the specificity limits mentioned above), and your type lists need to take this into account.
- Define the types or relationships that are possible between objects and the various authorities.
- Specify how user interfaces to edit all of these types and fields work. This includes breaking interfaces for each type of database item into screens and laying out fields on each screen.
- Encode all of this information as an installation profile.
It is possible use a pre-built profile. And feel free to ask questions of us at email@example.com. We've seen a whole lot of schemas and may be able use this experience to save you some time.
- Can I use the Art and Architecture Thesaurus with CollectiveAccess?
Yes, you can. The instructions for doing so are here.
- Can I use the Dutch translation of the Art and Architecture Thesaurus with CollectiveAccess?
Yes, you can. The instructions for doing so are here.
- Can I use Library of Congress Subject Headings with CollectiveAccess?
Yes, you can. The instructions for doing so are here.
- Can CollectiveAccess automatically generate accession numbers? Can I enforce standards and specific formats for accession numbers?
The multipart ID numbering system in CollectiveAccess provides flexible support for the types of numbering systems typically seen in museum and archival work. It is capable of auto-generating unique numbers at both the lot and object (item) level. And if CA's standard numbering system does fall short in some way, you can always develop your own numbering plug-in.
Interoperability with other systems and services
- Can CollectiveAccess function as an OAI data consumer or provider?
See this page for more information.
- Does CollectiveAccess work with Fedora/IRODS/<SOME OTHER DIGITAL OBJECT REPOSITORY>?
There was support for storage of CA-hosted media and object item-level metadata in Fedora and IRODS-based repositories in a branch of pre-1.0. This never made it into the 1.0 mainline code. We are working to re-implement this functionality for Fedora for version 1.5. There are no plans to restore irods support at this time.
- Can CollectiveAccess take advantage of web services?
Yes. We are implementing plugins to support various services based upon users' needs. If you want support for a particular service let us know and we'll try to accommodate it.
- What web services are supported "out of the box"?
As of October 2013:
- the Google Maps service is supported for both generation of maps and translation of addresses into coordinates
- the Amazon S3 storage service is supported experimentally as a replicated storage target
- the Apache SOLR search service is supported as a back-end search engine (can be used in place of the default MySQL-based search engine)
- the GeoNames geographic place name service is supported. Integration includes searching of the GeoNames database and linking of names to any type of record
- the uBio taxonomic name service
- the Library of Congress Subject Heading web service is supported. Integration includes searching of LCSH terms and linking of terms to any type of record
- Library of Congress Subject Headings doesn’t appear to be working. Can you fix it?
The Library of Congress Subject Headings datatype actually uses the LOC’s Linked Data Service: Authorities and Vocabularies facility to perform searches. If the search function appears to be broken in CollectiveAccess, it is possible that the search is broken at LOC. In order to troubleshoot this, try replicating the search at LOC directly.If the search does in fact work at LOC, but not in CollectiveAccess, it may have to do with the number of results generated by your query. Some keywords in LOC return thousands of results. To keep response times reasonable, there is a limiter in CollectiveAccess that truncates the number of results generated to 150. We’ve found that the function performs poorly if it tries to handle any more than that.
- Can I import <INSERT FILE FORMAT> into CollectiveAccess?
CA can import and convert a fairly long list of image, audio, video and document formats. These include the usual suspects like JPEG, PNG, TIFF, Photoshop, MP3, MPEG-4 and WAV and well as many more esoteric formats. You can see a complete list on the supported media formats page.
- Does CollectiveAccess generate derivates automatically?
Yes, CA can convert any supported media format into a web-viewable derivatives at specified sizes. By default JPEGs are created for images and PDFs, MP3 and FLV for audio and MPEG-4 (h.264) and FLV for video. JPEG frame previews are also created for video files. The derivative formats, sizes and qualities (compression or bitrate) can all be customized by editing a configuration file.
- Can CollectiveAccess watermark images?
CA can add visible watermarks to derivative images. Invisible watermarks, like those created by DigiMarc are not currently supported, but we are considering support for it in the future.
- Does CollectiveAccess extract metadata embedded in uploaded media?
Yes. IPTC, XMP and EXIF metadata are extracted from uploaded images and stored in the database. We are working on a system for mapping embedded metadata from uploads into specific fields in your schema, which will make extraction much more useful.
- Can CollectiveAccess embed metadata into media provided for download?
Not yet. We're planning to support this before the middle of 2011.
- Can I ingest media into CollectiveAccess in batches?
Yes. You will have to write a script to do so that takes into account the specific needs of your ingestion process, including the specifics of how media should be mapped to existing data. Send a message to firstname.lastname@example.org if you want example code to work from.
- Can CollectiveAccess search text contained within PDF and/or Microsoft Word files?
Yes, so long as you have the appropriate helper applications installed on your server: ABIWord (for extraction of text from Microsoft Word files) and/or PDFToText (for extraction of text from PDF files) or LibreOffice. You'll also need to make sure you've configured your external_applications.conf file with the proper paths to each helper application.
- What happens when I upload a file to CollectiveAccess that is not one of the supported filetypes?
Only media that is in a supported format can be uploaded. If CollectiveAccess can’t verify the media file, extract metadata and produce preview images then it will reject it. If you need to upload files in unsupported formats to CollectiveAccess you have two options:
- 1. Implement a media processing plugin to handle the format.
- 2. Create a metadata element of type File. File elements simply store uploaded files as-is. No attempt is made to identify, verify or process the incoming file.
- Unless you are a developer, option #2 is the simplest approach.
- 1. Implement a media processing plugin to handle the format.
- Why are you doing this?
We started the project back in 2003 because of the complete lack of non-commercial open-source solutions for museum collections management. In a world with open-source options for pretty much every type of software it seemed ludicrous that museums were still paying out large sums for ancient software with restrictive licensing terms.
- What's the catch? Do I have to buy a "Pro" version to do anything useful?
There is no catch and nothing to buy. Everything we produce is free, open-source software licensed under the GNU Public License version 3 (GPLv3). This includes the cataloguing software and software for making your collections information available publicly via the web, mobile devices and kiosks.
- Do I have to pay for support?
Not unless you want to. We provide free support via the online support forum and email. If your project requires dedicated commercial support there are a number of vendors in North America and Europe providing a broad spectrum of CollectiveAccess-related services. We cannot endorse any of these companies, but given a description of your project we will be happy to provide a list of appropriate vendors for you to contact and evaluate. Note that the CollectiveAccess development team does not request "finders fees," commissions or any other financial incentive from vendors for referrals. Our primary interest is in expanding the active userbase of the software and ensuring users have access to the services they need.
- Do I have to pay for customization?
If you need a feature that's not in the software, and you can't implement it yourself, the first thing you should do is post a feature request on the project issue tracker at http://clangers.collectiveaccess.org/jira. If it's something we are planning to do, or think would be generally useful to our users, we'll give you an idea of implementation time-frame and try to get it done. If it's truly custom work then contact us at email@example.com and we'll try to recommend vendors.
- How is the project funded?
We support the project financially by providing CollectiveAccess-specific development services to partner institutions. Typically, a partner will have a grant to develop some sort of collections related service, digital archive, public facing online exhibition or exhibit floor application. We enter into contracts with these institutions that provide us with funds and the right to freely distribute everything we develop for them under the GPL as part of CollectiveAccess. In return they get what they need to fulfill their grant requirements and the good feeling that comes from sharing with their fellow non-profits. It's a somewhat unusual model, but it works, and most importantly it lets us concentrate on more important things than where the next grant is coming from.
- What institutions are using CollectiveAccess?
A partial list of organizations using CollectiveAccess is on the Who's using it? page of the CA web site.
Configuring the User Interface
- How do I change the sort order of list items?
The sort order of list items will be alphabetical unless you declare the default sort value to “rank” and then assign ranks for each list value. You can do this from the User Interface, or from within a profile configuration.
To do this in the user interface, find the list you wish to re-order under “Lists & Vocabularies.” In the editor for the list, change the “Default sort order” to “by rank” from “by label.” Then return to the hierarchy browser to find the list items. Access each value contained within the list and assign each a numerical rank under “Sort order.” Declare “1” for the first item, “2” for the second, and so on.
If you are establishing rank order within the profile configuration, here is the proper syntax, where defaultSort enables the ranking of values, and Rank sets the sort order for each value:
<list code="object_types" hierarchical="1" system="0" vocabulary="0" defaultSort="1">
<labels> <label locale="en_US"> <name>Item Types</name> </label> </labels> <items> <item idno="artwork" rank="1" enabled="1" default="0"> <labels> <label locale="en_US" preferred="1"> <name_singular>Artwork</name_singular> <name_plural>Artworks</name_plural> </label> </labels> </item> <item idno="archival_material" rank="2" enabled="1" default="0"> <labels> <label locale="en_US" preferred="1"> <name_singular>Archival Material</name_singular> <name_plural>Archival Materials</name_plural> </label> </labels> </item> </items> </list>
- How can I edit the structure of a metadata field?
You can access metadata elements, or fields, for editing by selecting “Administration” from the Manage tab in the global navigation, then selecting “Metadata Elements” from the local navigation. A table containing all metadata elements in your profile configuration will be displayed. Select the edit icon for the element you wish to edit. Clicking the red X icon will delete the element altogether and all the data entered in that field system-wide. Once inside the editor for a specific metadata element you will see a menu of choices. You may change the height or width of the field, the minimum/maximum characters allowed in the field entry, and more.
Be careful not to change an element’s datatype if you are working on a live system with data, unless you are absolutely sure. Here you run the risk of wiping data. If, for example, you change a text field (Datatype: Text) to a List (datatype: List), you will delete all text previously entered in the field.
- How can I make a metadata element repeatable? Required? Restricted by type?
To bind an element to a specific record type, go to the Type Restrictions menu in a metadata element’s editor. You will find drop-down menus to bind the attribute to: Record table (object, lot, entity, etc.) and Record type (photo, document, image, etc.)
“Minimum number of characters”
Setting this to “1” or more will make the element required. Setting it to “0” will make this an optional field. If your element is a list, check “Require value”.
“Maximum number of attributes of this kind that can be associated with an item:”
Setting this to “1” will make this element un-repeatable. To make the element repeatable, select the maximum number of repeats for the element.
Number of placements:
“Minimum number of attribute bundles to show in an editing form.”
This number will determine how many times the element is placed on the editor.
- How can I edit a screen?
If you would like to add, remove, or re-arrange metadata elements in any given screen, you may do so through the User Interface editor. Under Manage, in the global navigation, find the “Administration” tab. There, in the local navigation, you will see a choice for “User Interfaces.” Under “User Interface” you will see a table containing all the Editors in your system. Locate the editor you wish to alter by type. If, for example, you wish to edit a screen in the Object records, select the “Standard Object Editor.” In the editor, you will find all the editable settings – name, code, type restrictions, and screens.
Select the screen you wish to edit by clicking on the Edit icon. You may add, remove, or rearrange elements in the screen content menu by clicking + dragging from the “Available editor elements” column and the “Elements to display on this screen” column. More info here.
- How can I restrict a Relationship bundle placement to a specific record type?
Unlike metadata elements, you cannot restrict Relationship bundles to a specific record type per se. Instead, if you need to restrict Relationship bundles to a specific record type, you actually make the restriction at the Screen level in the User Interface. Then you repeat the same screen, without the bundle, for the other record types.
Say, for example, your system has two basic object types: Photographs and Objects. You wish for “Related Photographer” (ca_entities) to appear on the Basic Info screen for Photograph records but not Object records. In the profile configuration, you will make two Basic Info screens – one for each type. These screens will be nearly identical, except the screen for Photographs will contain the related entity bundle and the screen for Object will not. Here is the syntax to restrict a Screen to an object type, where “type” equals the item idno for the object type you are restricting the screen to:
<screen idno="basic_photograph" default="1"> <labels> <label locale="en_US"> <name>Basic info</name> </label> </labels> <typeRestrictions> <restriction code="TypePhotograph" type="photograph"/> </typeRestrictions> <bundlePlacements> <placement code="idno"> <bundle>idno</bundle> </placement> <placement code="preferred_labels"> <bundle>preferred_labels</bundle> <settings> <setting name="label" locale="en_US">Title</setting> </settings> </placement> <placement code="ca_entities"> <bundle>ca_entities</bundle> <settings> <setting name="label" locale="en_US">Related Photographer</setting> <setting name="restrict_to_types">individual</setting> <setting name="restrict_to_relationship_types">photographer</setting> </settings> </placement> </bundlePlacements> </screen>
Then, to create a Basic Info screen for Object records, you would repeat the screen with a new restriction, and leave out the relationship bundle:
<screen idno="basic_object" default="1"> <labels> <label locale="en_US"> <name>Basic info</name> </label> </labels> <typeRestrictions> <restriction code="TypeObject" type="object"/> </typeRestrictions> <bundlePlacements> <placement code="idno"> <bundle>idno</bundle> </placement> <placement code="preferred_labels"> <bundle>preferred_labels</bundle> <settings> <setting name="label" locale="en_US">Title</setting> </settings> </placement> </bundlePlacements> </screen>
- How can I add a list item value to a drop-down menu, checklist, or hierarchy browser?
You may add or remove list item values within the “Lists & Vocabularies” menu, accessible from the Manage Tab in the Global Navigation. Note: This is only available to users with the access privileges to manage lists and vocabularies. Here you will find a hierarchy browser including all lists in your configuration. To add a value to a specific list or vocabulary, locate the list in the first facet of the hierarchy browser. Select the arrow icon to the right of the list. You will see the menu above the browser change to read, “Add under [the list you selected] new [drop-down menu] followed by a plus icon. Select the type of list value from the drop-down menu – concept, facet, guide term, or hierarchy name and click on the plus icon.
This will take you to an editable form for the list value, were you will assign an item name, description, unique item code, and other values. Click Save for your item to appear in the list.
If you are adding list values in a profile configuration, the syntax is this:
<list code="object_types" hierarchical="1" system="0" vocabulary="0"> <labels> <label locale="en_US"> <name>Item Types</name> </label> </labels> <items> <item idno="artwork" enabled="1" default="0"> <labels> <label locale="en_US" preferred="1"> <name_singular>Artwork</name_singular> <name_plural>Artworks</name_plural> </label> </labels> </item> <items> </list>
1. Can I change a record’s type after it is saved or duplicated?
Yes, providing that your login is associated with a role that has "change type" privileges for the kind of record you're working on. If you have privileges you'll see the "Change Type" icon in the editor inspector. Note that when changing types any fields in the record that are not supported in the new type will lost.
2. How should I decide when to use Collections, and when to use Sets? What is the difference?
Collections contain a defined group of objects. Collections can represent physical collections, symbolic collections of items associated by some criteria, or any other arbitrary grouping, but they are always considered to be authoritative and are system-wide.
Sets, on the other hand, are an ordered grouping of Objects defined by users for a specific purpose. Unlike Collections, Sets are ad-hoc groups of records created by a user for a practical purpose, such as a working checklist for an upcoming exhibition, or a set of Entities for which biographical information is needed. They are typically temporary groupings and thus not meant for collection-level cataloguing. A Set can be restricted to appear only to the user who created it, or the Set can be shared with other users or user groups.
3. How do I format search results?
When you are conducting a search from the Find menu, you can utilize Display Options to not only switch between the default and custom displays, but also change the sort value, results per page, and layout.
Sort: The sort value defaults to ‘relevance,’ which is determined by the number of matches found in the system. By selecting another term from the drop-down, you can sort the search results by other data elements – such as ID number, title, type and so on.
Number per Page: Up to 48 records can be displayed on a single page at one time. The minimum results per page is 12.
Layout: There are 3 layout options – Full, List, and Thumbnail. The List view displays the data values from the selected display in a tabled format. Full, on the other hand, displays a large preview image as well as the data values from the selected display.
You can create custom displays for search results, and then use those results to export tab or comma delimited files. Displays allow you to select exactly which fields from Object records you want to see in search results.
4. How do I create a display?
Open “My Displays” from the Manage tab in the Global Navigation bar. To create a new display, select an object type from the “New display for” drop-down menu. Then, enter some basic information about your new Display:
Enter a unique Display list name for the new Display.
Enter a unique alpha-numeric identifier in the Display code text box.
By selecting Yes under System Form, you can make this search accessible to all users in the system. If you want to limit access to the form, select No. If the new form is not system wide, you may assign read-only or edit access to specific users or groups using the User Access and Group Access elements. Note: This feature should only available to users with administrative privileges.
If you would like display list values to show even when they are null, make sure the Display Setting box is checked.
To add contents to your display, scroll down to the bottom of the Display Basic Info Screen, where you will find two columns. The left column contains all available display items – elements that can potentially be shown in search results. The right column will contain specific elements chosen for the Display. Using your mouse, drag elements from the left column to the right to configure a Display.
5. In Lists & Vocabularies, what is the meaning and function of Concept, Facet, Guide term, and Hierarchy name?
These are distinguishing types for list items analogous to those used in the Getty Art and Architecture Thesaurus (AAT). They are part of the default configuration defined in install/profiles/xml/base.xml and inherited by most installation profiles. Although they may seem ubiquitous they can be overridden with other values at any time and in any configuration profile if desired. As with types for other catalogue records (Eg. object types, entity types, place types), list item types serve both to distinguish items by general function and to allow metadata elements to vary across items. The default values are intended to provide compatibility with a variety of vocabularies in general and the AAT in particular. When the AAT is imported these types indicate the following:
- 1. Concept: a list item that represents a unique concept and may be used for descriptive cataloguing.
- 2. Facet: a list item that is the parent for thematic unit of vocabulary concepts. Facets are primarily an organizational element and not typically used for cataloguing.
- 3. Guide term: a list item that groups those under it for convenience. Guide terms are purely organizational and never used for cataloguing.
- 4. Hierarchy name: A list item that defines a sub-unit of a vocabulary. These are typically high-level organizational elements.
Configuring Pawtucket (front-end)
- I've configured my browse facets in the browse.conf configuration file, but my facets are not appearing!
Browse facets only appear if there is an available record that matches your chosen metadata. For example, you may have configured browse.conf to allow browsing for objects by related entities. If, however, none of the object records in your system have related entities, the browse facet will not appear in the menu. Additionally, Types will not appear if your system contains only a single type. The browse facet will appear as soon as records of another type are created.
This category currently contains no pages or media.