Create custom metadata types in Omeka S


#1

Hi, Im currently using Omeka S for some documents, that have very specific properties, that are not found in Dublin Core, or any metadata ontology. Why?. Because we want to publish some of the content of the documents also in the metadata to make it searchable. Is there an easy way of creating these metadata types?

I have found that it is possible in Omeka Classic:

I have figured out, that maybe i can create my own rdf or turtle vocabulary and then import it on omeka, but i would like to avoid this. Is there a simpler solution?


#2

I needed it too, so I created a module to create new non standard properties (with the namespace of the installation), but it will be published only in February.


#3

I think part of the spirit would be that the content of documents would go into Media, probably HTML media. However, I’m not sure that makes it searchable. That might be something for us to look into.


#4

For my part, i need some metadata that are not standardized as rdf, but in some other standards (marc cataloguing), and my users don’t want to lose them. And there are other cases where the user want to keep a quick note about an item in a shareable admin field, so it’s not standardized for that case, but very useful. Another use case is the import of data from an ecm tool. Of course, many fields are converted into rdf properties, but some data are very specific and requires specific fields, like it was very simple to do in Omeka Classic and in any ecm tool in fact.


#5

Daniel_KM, currently im working to digitize a large collection of documents. I can help in the tests of your module. Thank you for the work.


#6

I had to address a similar issue.

Finally, I’ve chosen to use HTML media. I’ve edited my theme to display directly HTML content in the item page, like a property. And I will work with Solr to make this HTML content searchable.


#7

This is an interesting solution! Can you share more details on how did you implement this feature? Thanks!


#8

To display HTML media content in item pages, I’ve added this code in my omeka/site/item/show.phtml

$embedMedia = $this->siteSetting('item_media_embed', false);
/* @var $itemMedia Omeka\Api\Representation\MediaRepresentation[] */
$itemMedia = $item->media();

if ($embedMedia && $itemMedia)
	foreach ($itemMedia as $medium)
		if('HTML' == $medium->ingesterLabel()
				&& stristr($medium->displayTitle(), 'transcription')) {
			echo $medium->render();
			break;
		}

All media I wanted to display had “transcription” in their title (you can remove this condition if you want to display all HTML media).

Dev version of Solr module is now able to index HTML media content. You have to choose “HTML content” — under “Media” category — as field. You also need to use PlainText value formatter in order not to index HTML tags (all tags are removed before indexing).


#9

Thnaks a lot! I’ll try your solution even if I never used Solr module before …


#10

The module Custom Ontology is built to manage specific classes and custom properties.


#11

Hi, i currently “solved” this problem interpreting widly the Dublin Core categories.

Thanks All.