ARTigo

Overview

Art history thrives on working with artworks and their digital reproductions. These are collected online in image catalogues, some of which contain several million objects. To search for images in such directories, descriptive texts are needed: a task you can help us with in a playful way—and at the same time learn a lot about art!

ARTigo is a collaborative platform and an example of open university research in which you participate as a player. All ARTigo games work like this: You are presented with an image. The goal is to describe this image within a certain time with keywords (tags) in such a way that as many points as possible are scored. After you have played a predefined number of rounds, the metadata for the images become visible. In this way, you not only generate art-historical knowledge, but also collect it. Every tag you enter is saved and improves the search for artworks—a win-win situation.

Development setup

Requirements

• docker
• docker-compose

Configuration files

Copy the contents of .env.example to .env and adjust the settings.

Setup process

1. To import data, .csv and .jsonl files should be stored in the ./dump and corresponding .jpg files in the ./media folder.

2. For the OpenSearch container to run properly, increase the virtual memory:

   sudo sysctl -w vm.max_map_count=262144

3. Create an unprivileged artigo user and group:

   sudo addgroup --system --gid 1998 artigo && sudo adduser --system --uid 1999 --ingroup artigo artigo

4. Change the permissions for the dump and upload folders:

   sudo chown -R 1999:1998 dump
   sudo chown -R 1999:1998 media

5. Build and start the container:

   sudo docker compose up --build

6. Install npm:

   sudo docker compose exec web npm install

7. Apply necessary Django operations and import fixtures:

   sudo docker compose exec api python3 manage.py migrate
   sudo docker compose exec api python3 manage.py loaddata sites.json

8. Import the data into api and search:

   sudo docker compose exec api python3 manage.py import_data
   sudo docker compose exec search python3 -m artigo_search --mode client --task insert

Miscellaneous

Code reloading

Hot reloading is enabled for api. To display changes of web (http://localhost:8080/), run:

sudo docker compose exec web npm run build

Alternatively, use serve to enable a hot reloaded instance on http://localhost:8080/:

sudo docker compose exec web npm run serve

Schema updating

To re-generate the OpenAPI 3.0 schema file, use the following command:

sudo docker compose exec api python3 manage.py spectacular --file schema.yml

Database schema migration

To generate database schema migration files, use the following command:

sudo docker compose exec api python3 manage.py migrate

Formatting

Lint can be used to help with standardized formatting:

sudo docker compose exec web npm run lint --fix

UML class diagram

To create an UML class diagram from the Django models, run:

sudo docker compose exec api python3 manage.py graph_models frontend --hide-edge-labels -o /dump/models.png

Contributing

Please report issues, feature requests, and questions to the GitHub issue tracker. We have a Contributor Code of Conduct. By participating in ARTigo you agree to abide by its terms.