Using the Search By Metadata Plugin for Omeka

I recently migrated my “Orange County Fallen Heroes” local history project from Omeka version 1x to Omeka 2x. What a difference! One of the best biggest changes I notice in Omeka 2x is the addition of a new plugin — Search By Metadata (download here; documentation here). The plugin was added in April, and is only available for Omeka 2x.

What the Plugin Does

The Search By Metadata plugin is a HUGE timesaver for me. Much like Tags, the plugin links together identical terms, allowing users that click on that hyperlinked term to pull up all Item records that use the same exact term. However, the plugin does this through the existing Dublin Core metadata (or whichever Element Set your site uses), not through descriptive Tags.

How is this a timesaver? On the previous (Omeka 1x) version of my OC Fallen Heroes project, when I wanted to link together identical Dublin Core metadata terms (for example: Branch of military service, or Hometown), I had to enter those terms in the proper [custom created, under “Person” Item Type] Dublin Core fields, but then repeat those terms as Tags since Tags were previously the only way to generate a hyperlinked group of identical terms. And because I wanted those hyperlinked terms to show up in my actual metadata fields for each casualty’s Show Item view, I would copy the corresponding Tag’s URL to the respective metadata field using the HTML mode to display an actual hyperlink. This workaround was a big time pain in the you-know-what, a ton of extra work, and a really bad use of Tags (Tags should not duplicate the content recorded in metadata fields IMO).

Because of the new Search By Metadata plugin, I use very few Tags now.

Usage With the Simple Vocab Plugin

The plugin documentation recommends using the Search By Metadata plugin in conjunction with the Simple Vocab plugin (download here, documentation here) if you routinely use a a standard set of terms for particular metadata fields. I concur with this recommendation.

The Simple Vocab plugin allows site administrators to build and manage a pre-defined controlled vocabulary, to ensure that frequently used terms are not misspelled or formatted in different manners (for example, the terms I use for each military service branch…I don’t want to mistakenly type in Marines sometimes and Marine Corps other times).

I do not use a controlled Simple Vocab for every single metadata field. Right now, I am only using it for a few fields (Hometown, Gender, Branch of Service, Place of Interment, and a few other hidden administrative “to do” fields: Need Photo [yes/no], Need Bio [yes/no], Needs Confirmation [yes/no]). All the different Grades/Ranks used throughout the various wars made it too difficult to apply a controlled vocabulary to those terms, as do the numerous places of death and places of service. Even the Places of Interment are a bit too long a list (to which I have to keep adding), but I found myself frequently misspelling those terms.

Keep in mind that it is far easier to set up a Simple Vocab PRIOR to building out an Omeka site, or while it is still young (see last bullet point under Viewing Terms Already In Use). If you have already published hundreds or thousands of Item records, it can be a lot of work to go back in and clean up those metadata fields in your Item edit view and Simple Vocab settings (you might want to do the clean up in your actual database, after saving a backup, and then set up your Simple Vocab using those refreshed metadata terms). I have about 600 Items in my Omeka site, but only about 60 that are publicly visible. Since I am actually going through each CSV-imported hidden Item to clean them up prior to publishing, I am able to make my Simple Vocab fixes at that time.

How I Use Search By Metadata Versus Tags

With this new Search By Metadata setup in place, I now use Tags only for terms that fall outside of the standard metadata elements I want to capture for each casualty profile. These metadata elements are the standards biographical details that I think are important (and ultimately, findable) for each casualty across all wars: Date of Death, Place of Death, Hometown, Branch, Unit, Rating, etc. Tags are the less common descriptive terms, such as: a battle name, a school or college name, the name of a ship, etc.

Aside from the much easier faster workflow, what this setup does is allow my site visitors to drill down and view specific terms used only in a specific context. When I had to use linked Tags in the old Omeka v1x setup, the linked Tag term “Santa Ana” could have applied to a Hometown, a Place of Birth, Place of Death, or Place of Service. All of those instances got lumped together into a single linked list because Tags cannot be categorized or filtered. Now, my site visitors can choose if they want to display a list of all casualties who called “Santa Ana” their Hometown, any that might have died in Santa Ana, any that might have served in Santa Ana, or any born in Santa Ana.

If you are on Omeka v2x and using the Search By Metadata plugin, I would love to hear about your own use case!