buda-base / blmp-prototype-flow Goto Github PK

The urls at src/api/api.js line #'s 248 and 264 should be local so that blmp is using the local lds-pdi rather than the one at buda1. If the blmp is running on buda1 then it will automatically be using the lds-pdi on buda1.

Or perhaps some other configuration approach that allows to easily switch where the lds-pdi is.

When a string is tagged as English (in RDF), it should be made clear to the browser so that the English spell checker gets triggered by the browser. Same for other languages, maybe integrating custom spell checkers in the future.

We should integrate a login facility for users. In a first time it should just allow users to log in and display their name, and in a second time when the server will be ready, the auth token will be transmitted. We'll be using Auth0 which is an OpenIDConnect (or oidc) which is a layer above OAuth2. These are all widespread standards and some high level libraries exist, see this one integrating Auth0 and React. It's not an emergency but that would be an interesting first step in the low-level plumbing.

• Create a new GitHub project
• Include the current GitHub issues
• Raphael to review his Trello tasks, and add them in this GitHub project
• Nicolas to review the email sent to Raphael based on this discussion between him and Chris (check green highlights), clarify each task, and add them in this GitHub project
• Nicolas to check the previous GitHub project, clarify each task, and add them in this GitHub project
• Elie to review his notes and clarify each task, and add them in this GitHub project
• Elie / Chris to prioritize each task

The wikidata editor is a possible source of inspiration. It doesn't have special qualities, but has the advantage of being something similar to what we want to achieve and is used in production...

example : https://www.wikidata.org/wiki/Q445460

Chat with Elie

• Develop a library using the same API and data model than the RDF JavaScript Libraries which would allow us to create patches
• independant library, published on npm
• the idea would be to make it generic so that we could give it to organization behind RDF JavaScript Libraries so that they could take care of its maintenance
• to handle our specificities, like the "taskid" in the header, we could have a method which handles what needs to be put in the header
• could be pretty simple using a naive approach
• its API should be pretty similar than the Java library. Chris knows the code, she will be able to guide me on the most interesting fonctions

The right panel should display information only when clicking on resources on the left panel which are:

• entities (type :Entity)
• or which are descendent of entities

Entities are all resources which are of the following types: Agent, Etext, Item, Lineage, Place, Topic, Volume, Work.

X is a descendent class of Y:

• if there exists: X1, X2, ... XN avec N >= 0
• such as: X rdfs:subClassOf X1 rdfs:subClassOf X2 rdfs:subClassOf ... rdfs:subClassOf XN rdfs:subClassOf Y

Different ways to express the same concept:

1. object properties which range is a class which is a descendant class of :Entity (or which has for ancestor :Entity)
2. object properties which range is not a descendant class of :Facet or :Event
3. objects which class is not a descendant of :Facet or :Event