Examples of good, bad, and desired user experiences

Owned by Rain Michaels (Unlicensed)
Last updated: Sept 17, 2015 by Angela Spinazzè (Unlicensed)
6 min read

Please use this wiki page to let us know about examples of effective uses of ArchivesSpace, poor uses of ArchivesSpace (or poor experiences within it), and desirable experiences from other products you have encountered. Please include direct links to examples, as well as specific details about why the example is in this list.

Good ArchivesSpace experiences

  • The ability to publish only certain things to the PUI at a very granular level.
    • Specifically, I want to describe everything we have in our holdings, but I don't want everyone who has access to the PUI to see everything what we have.  
    • Example - Board of Trustees materials might have sensitive files (say on executive pay). I may want the Board admin and the Board chairman to link to e-version of what we have but they don't want everyone at the company to know what we have.
  • As the archivist, I like the ability to sort by date created or modified.  It useful because I can see something that I recently modified and pushed to the PUI side.  It could also be useful for users that want to know what is "new" in the database.  However, I don't think it needs to necessarily be called "created" nor "modified".  The distinction may be confusing to end users. Maybe they could be combined into something called "new" or "recently modified". 
  • The ability to link from the "names" to all the associated collections is wonderful.  Also the ability to link from the collection to the names is also very useful.  Context and linking is very important for discovery. Also linking from collection to collection is great.
  • The search bar on all pages is very useful and I like how the advanced search is neatly hidden but easy to get to.
  • In the components section - the "record tree" is very useful. Especially for very complicated hierarchies
  • Breadcrumbs showing where you are very useful and the fact they are hyperlinked so you can go back to a higher part of the hierarchy is great.

Poor ArchivesSpace experiences

  • Jargon such as "Digital Objects", "Accessions", "Classifications".  I don't think folks outside of the archives/ library professional sphere know what this means.  For example in the military, "Classification" means something entirely different.  (yes, we need sensible defaults in ALL of these categories!) 
  • Location of container list tree at very bottom of page and default of collapsed view (when tree has a hierarchy); hard to find, confusing to use.
  • Relationship of container list tree to individual components; when a file (for example) is clicked on, the page you are redirected to feels like a dead end (and layout of information on this page is confusing); even with the breadcrumb trail, the context of this page within the larger Resource Record is confusing.
  • When instances are added at the Resource Record level on the staff side, there is not a way to suppress these from view on the PUI; these then appear above the container list tree, dominate the entire record view (more confusion), and make public some info (such as shelving locations) that may be preferred to not be published.
  • Search results are returned in an unclear order, difficult to locate anything specific via searching.
  • Search results include components, but without any context- the collection they are part of is buried within lists of folder titles, sometimes pages later. Also, it's common for archival objects in different collections to have the same name (consider the term "scrapbook"), and without any context, users can't immediately tell the difference between 10 sequential results for "scrapbook."
  • Relationship between left-hand facets and results displayed–often not accurate (i.e. clicking on some facets does not limit results displayed to only those results related to that facet–example: selecting a facet for a repository displays results from other repositories in addition to the one clicked on)
  • In main horizontal navigation bar, choices that do not have anything published on the staff side still appear as options (i.e. if no Accession are marked as published, Accessions still appears as an option) 
  • Also, accessions were not available through the public interface in Archon, but when they were imported they became public in ArchivesSpace. If users find them in both places, they are going to be confused as to which record to look at.
  • "Include components" checkbox above the facets has no context; non-staff users will generally not understand what this means or why/when/etc. to use it.
  • The Sort order option of "Record Type" is meaningless to users.
  • There is no way to know if a keyword search searches all areas of a resource record (notes, components, instances, etc.).  Seconded!!!  We discovered, for instance, that everything in the JSON record is searched, including things like the "lock_version" and other database ids, which make searching for numbers nigh but impossible.
  • The inability to easily unpublish a record, e.g. if a finding aid is being updated, there is no way to "hide" it while changes are being made. This could end up being confusing for users.
  • Users cannot browse alphabetically with any ease, though those used to Archon may expect to be able to.
  • There is no "Help" for users. 
  • Too many clicks to get to the actual image digital object

Desirable user experience examples

  • Librarything shows a great example of discovery of materials (in this case books) through authors. The authors names are not regularized which bothers me but I think ASpace could leverage this functionality with the "names". https://www.librarything.com/author/rowlingjk
  • BBC programmes uses linked data and has a slick search interface. I like the use of the word "categories"  instead of "classification" in other words they limit the jargon.  It's very visual to the point of being a bit busy but I think some degree of using images is important with so much visual media in archives. http://www.bbc.co.uk/programmes Look at the examples below the search bar to get a full idea of the capability for showing dates, etc.
  • Princeton and BYU finding aid sites provide different approaches to component-level displays with contextual browsing, allowing users to directly access component descriptions while still being able to see what materials surround a particular folder or item in a collection. http://findingaids.princeton.edu and http://findingaid.lib.byu.edu
  • NYPL:  As far as the single-scroll view of finding aids go (here's an example: http://archives.nypl.org/mss/22833#detailed), I think they have one of the best.  They also have a great landing page: http://archives.nypl.org/ (and a whole host of other things like to like, including the URL structures, the fact that the collection's arrangement is always clearly visible). I also like that searching returns collection names first, and then, under each collection, indicates how many matches for a search term were found in that collection, which a user can then choose to view at the component level.
  • Rockefeller Archive Center: http://dimes.rockarch.org/xtf/search  Great implementation of the XTF bookbag feature.  Also note that they provide a lot of useful information to researchers, like access restrictions notes, directly in the search result screens.
  • Archives Portal Europe:  https://www.archivesportaleurope.net/.  The scale alone is impressive.  Also note the lazy-loading context trees.
  • Princeton Theological Seminary: I like the data visualization touches that they've added by doing entity recognition on the narratives in the description.  Here's an example: http://manuscripts.ptsem.edu/collection/32  Also, they have a nice way to view by locations, http://manuscripts.ptsem.edu/?map=true, and searching works well (including highlighted text, etc.).
  • UMichigan's Bentley Image Bank (e.g. http://quod.lib.umich.edu/b/bhl/x-bl001009/bl001009) has simple sharing and citation options at the top of the screen, and also provides the ability to smoothly zoom in on digital objects.
  • University of North Texas:  A great digital library, in general.  I also really like their use of stats (extracted from G.A.) throughout the application. Examples: http://digital.library.unt.edu/stats/, http://digital.library.unt.edu/ark:/67531/metadc13/stats/
  • WebGL in general: Some cool stuff here, for example, http://acko.net/blog/on-webgl/.  I often wonder if the context tree view is the best/only way to let users navigate a collection... and I also think it would be an interesting experiment to have an interface that allows a researcher to browse the archival stacks.
  • Museum Victoria: http://collections.museumvictoria.com.au/  I like how the home page uses a cross between info graphics and images.  Under articles - these very much like "finding aids" and they just look awesome - Very usable.  I like that the citation is right up front.  And here's the code link on github https://github.com/museumvictoria/collections-online
  • The Oriental Institute: Search our collections. Sharing this as an example of public user interface for access to museum, library, and archival holdings. Some of the issues we have recently been discussing, facets, labels, breadcrumbs, etc. are also present here and they use Solr to drive this as well.

Other

  • No entries yet