In 2016, I worked for ITEM, the department of literature of CNRS, to optimize and modernize their Archives Zoliennes WordPress website, the online place to browse Emile Zola’s graphical archives.

In 2018, I was asked to develop a new website for ITEM, in order to migrate all Zola’s transcribed letters from Omeka CMS to WordPress.

Let’s dive into this interesting challenge that involved front and back-end WordPress engineering.

What problem did I solve?

ITEM started the tremendous task of organizing and digitalizing all of available Emile Zola’s letters. This work started a few years ago and thousands of letters were already transcribed, scanned and stored in Omeka alongside with precious metadata.

The team really enjoyed managing scientific content with WordPress on Archives Zoliennes and decided to use this CMS for Correspondances Zoliennes. Managing content items in Omeka was proved to be cumbersome and complex and it was necessary to opt for a better solution. WordPress was the perfect answer to manage content in a centralized back-end and to display letters on the front-end.

The new homepage of

Technical details

Working with Omeka API

More than 1500 letters were already stored in a Omeka-powered website created by the client. Importing this data into WordPress was crucial.

A custom WordPress plugin was written from the ground-up, communicating with Omeka database via its public Rest API and importing all the existing pieces of content:

  • collections, tags, contributors, creators and languages were imported in new custom taxonomies for proper content organization
  • transcribed letters were imported in a new custom post type
  • scanned letter images were also imported in WordPress and assigned to their corresponding letter post
  • all the existing pieces of data (date, sources, publisher, etc.) are assigned to the letter post via ACF-powered custom meta fields
A better way of managing letter metadata in a tabbed metabox

A custom back-end experience

Alongside with all those fields created to store many pieces of metadata, the administration of a letter transcription needed some improvements for easier management.

The text of a transcribed letter contains notes (primary and secondary) which store complementary explanations of Zola’s text.

A custom back-end JavaScript logic has been developed in order to link the notes fields with the transcription WYSIWYG main editor. Dynamic markers are automatically added in the transcription editor when a note is created, and these markers are transformed on the front-end side of the site to mention and link to notes anywhere in the content of a letter text.

A tailored front-end interface

With so many meta data linked to one single letter, we needed to think of a way to properly display it. Inspired by Van Gogh Letters website, the single letter page offers the visitor a customizable layout displaying the letter images, transcription and notes.

Notes and transcription text are smartly linked with anchors and tooltips. The 3-columns layout can be reordered to let users emphasize images, transcription and notes the way they want.

The front-end side of a letter

Complementary meta data is displayed below the main section.

ipsum fringilla dapibus ante. facilisis efficitur. Phasellus felis consectetur sit tempus