An interactive visualisation that provides access to a selection of objects from the 'Lipperheidesche Kostümbibliothek'. Different texts are juxtaposed with a graph navigation that allows the user to browse the paintings, prints, drawings and clothes in the collection. Instead of showing the collection in its entirety, it is accessible through a curated selection of essays designed to provide a better understanding of it, combining editorial approaches with free exploration and user-driven granularity.
Interested in adapting it to your collection?
Go to the refa-reader-template repository to configure it.
- In the Terminal, go to your production folder:
cd path/to/my/folder
- Clone this repository
git clone https://github.com/uclab-potsdam/refa-reader.git
cd refa-reader
- Install dependencies with yarn
yarn
- Run and build the development environment.
yarn dev
yarn build
- Preview the production build with
yarn preview
.
Essays need to be inserted inside the src/route/texts/
folder.
Every markdown contains metadata to customize the layout:
---
title: The title of the essay.
date: "2023-12-14"
color: "blue"
author: "Name of the Author"
isPublic: true // false
lang: en
description: "The description which appears in the home page"
cover: "https://example.com/image.jpg"
---
To configure the graph visualisation it is needed to customise the setup file in the src/setup.json
.
{
"title": "The title of the Website",
"api": "https://exampe.com/api", // The link to a Omeka S Api
"publicSite": "", // The link of an Omeka-S collection
"languages": ["en"],
"description": {
"en": "Text to render in the homepage"
},
"mainCategories": [
// It is possible to customise the column layout by specifing a category and properties as a js Object.
{
"key": "Category title",
"props": [
"is refered to it",
"shows features of",
]
},
...
],
"descriptionSeo": "metadata field for SEO",
"imageSeo": "image path SEO"
}
``