Criminal photo archives: is Omeka-S the right tool?

[FedeS]

Hi Omeka team,
we should start a new project and we are considering using, instead of the usual Wordpress, Omkea-S.
Sorry if we refer to Wordpress but it is easier for us to describe the concepts using the terms we are already familiar (please correct us with the proper names!) with and sorry if some parts are not accurate, we are just getting into this world…

We have read with interest this and that and other documentation but we still have some doubts, particularly two.

In WordPress it is possible to add “post types,” either by changing the value on the dedicated column of the post table, or by creating custom tables: this can be done with plugins or directly from code. Between these tables it is then possible to create relationships.
In our case, we should support the study of criminal images: so we will have a repository of images, related to one or more crimes and which are part of one or more dossiers (plus other secondary relationships and relationships with custom taxonomies).
We report here a draft of the reports we would like to develop:

Is it possible to obtain these types of relationships?

Also, in WordPress it is possible to extend the standard “post” by adding custom fields: this can also be done with plugins or directly from the code.
In Omeka, from what we understand, it is necessary to create a “Vocabulary” to import.
In our case, we should have a custom vocabulary: e.g. we should have fields for the subjects portrayed, motive of the crime, place of the crime, dates of the trial, etc…
Again, some of these fields we would like to belong to a list (enumerated type) but still be manageable by the editors of the platform (e.g., crime location: it is not possible a priori to compile an exhaustive list, such as private home, park, hotel, etc…) so that it can be searchable: an analog on WordPress could be “taxonomies.”
How is it possible to create and manage custom Vocabularies?

Thanks for the long read,
Federica

[AllanaMayer]

I’m not proficient in Italian, but I imagine the following Omeka structure covers what you have in mind:

Items (nodes) with different classes (types) and individual resource templates (sets of custom fields) for Crimes, Images, Dossiers, Publications, and Archive Files. If I have this right, “processo” is the court case? The trial? That would be another type, presuming you can have many crimes in one trial, and many trials for one crime. I am not clear on what “Esemplare” is.

Perhaps you can use Item Sets for dossiers. Note that items can be in many item sets, item sets can have many items, and item sets can be used as linked relationships in metadata fields, so this is quite flexible.

“Classes” in Omeka are really just a textual label for items (Person, Place, Event, Text, Image, etc.). The real work is done in the resource templates, that allow you to set up the metadata profiles for each kind of item. You would compile your Crime, Image, Dossier, Publication, Archive File, and Trial templates from various vocabularies you can find online. If you need to create your own fields, the Custom Ontology module can help. But if you can find existing vocabulary fields that are conceptually broader than what you want, you can write in custom labels to display on your Omeka site to make them clearer for your purposes.

I can see People in your diagram, but you don’t seem to have much need to describe them (just a “descrizione” field, which I imagine is just a short biography?). If you instead are simply using these as controlled terms for browsing, you will want the Metadata Browse module. If your people only exist relative to their roles in the other nodes (investigator, criminal, journalist, etc.) you will need to make these roles explicit with the metadata fields and fill in names as values. You don’t want the crimes to have simply a “list of related people” and then hide their roles away on their profiles. Note also that if these people are listed elsewhere (e.g. if you were to create nodes for them in Wikidata) you could link there and host the descriziones there.

For your crime location example, I imagine you will probably have relatively clean and regimented data upon bulk import. After that, you can use the Value Suggest module to display all the already-filled-in properties in the installation to help you describe items further while keeping the data clean.

You don’t seem to have a need for Mapping (locating crimes around the city, e.g.) but you obviously know you will need Numeric Data Types for date formatting. You may wish to do “Location” as browseable metadata values, as above, rather than locations with actual GPS coordinates on a map?

I hope this is enough to get you started - let us know any further information that might help work out your data model.

[rickyx]

Hi Allana,
am a colleague of Federica.

Thank you for your detailed reply, we will study the resources you have linked in detail.

Thank you,
Riccardo

ps. ‘Esemplare’ is a specimen of the species: e.g. a photo of the crime appears in the police file and in a magazine, cropped, and in the newspaper, but in black and white: the image is the same but in different specimens. For the rest, your Italian is perfect!

[FedeS]

Hi Allana,
your reply was very exhaustive. Thanks!
It seems that Omeka S seems to be the perfect tool, so I’ve started the Omeka S course yesterday.

Thank you very much,
Federica