Add Interpretable pictograms to Cboard

[lutangar]

After diving deeper into the AAC (Alternative Augmentative Communication) ecosystem I figured out a few things :

• it's a very dynamic ecosystem (see :uk: https://opendirective.net/posts/2020-04-17-the-open-communication-symbols-ecosystem/ for example)
• it's embracing (and pushing forward) open-source and interoperability (see OpenAAC manifesto :uk: https://www.openaac.org/#two)

It also made me realize what should matters above else is how the various boards are designed regardless of the technology :

• symbol
  • how relevant it's to the targeted audience (see :uk: https://blog.globalsymbols.com/en/home/ for example)
  • adequation of the vocabulary (the word) associated with it
    • the nature of the word
      • and thus the color associated with it (see :fr: http://documentation.apflab.org/crnt/api/article/article-display.php?r=06213colorcodingsystems)
  • its place on a board -- literally -- where it's spatially placed
  • how it relates to other symbols
    • in the same area of a board
    • on the board
• board : in the end it is what's paramount, no matter how beautiful your icons are, it's how they are arranged into boards, and how they will be linked together, how deep etc. will determine how efficiently one will be able to find a concept.

In fact the methodology I'd suggest to achieve this would be to design boards first and symbols will follow.

• The first milestone would then be a PDF where each pages would be a board which could be printed for a workshop.
• Second milestone would then be to deliver an _OBZ archive, allowing these boards to be used in any compatible AAC software
  • The OBF files might be hand crafted at first but the maintenance process should then be reversed to be automated and use the OBF files as the single source of truth. Of course graphic designer mustn't be forgotten in the process and they should be trained and equipped with preview tool to ease there work.
• Third milestone import the OBZ archive (and automate the process) into our self-hosted instance of our favorite AAC software CBoard.

Note 1 : You may find a list of open-source board here : https://www.openboardformat.org/examples

Note 2 : If Interpetable boards are to be exposed at https://data.grandlyon.com/ then they should be exposed as OBF files and images hosted there as well. I'll make a proper issue for that. EDIT : see issue https://github.com/interpretable/api/issues/8

[lutangar]

I actually worked on an "import" script on a dedicated repository : https://github.com/interpretable/import

There are still manual steps to be taken but If we decide to stick with Google Drive the export and the images download should be automated. For telling the truth I'd rather to with an hosted NextCloud solution.

At first I thought we needed to load our data into CBoard API database. In fact we build a "seed" file ./src/api/boards.json in the CBoard format which is then loaded in the CBoard app local state. I don't think we can do the same thing with the obf format at the moment.

You may have a preview of the result but the datasource clearly need better classification to be usable in CBoard : https://docs.google.com/spreadsheets/d/1EK-Lvo608LWAjOEJIbMyla5pPgknCRWXEzkcP4U6bFQ/

Also a good news for the illustration team is we should stop renaming the image files for now and focus on the pictograms links to a board.

See for yourself : https://interpretable.projets.erasme.org/board/Racine