As discussed on yesterday's call, we should make some clarifications to the spec around the use of symbols, both from an author mark-up perspective, and from a renderer perspective. This is about the rendering side of things. Whilst the spec is primarily about author mark-up, it's important for authors to have some expectation of how things may be rendered, and it's important for UA/extension/AT implementers to be measured that the rendering process is simple.
We want to make clear that, with our Bliss concepts look-up, and mappings to other symbol sets, rendering is relatively simple, and would benefit from these look-ups. Here are some points that we need to ensure are clear:
The Bliss concept ID may map to one rendered symbol, or possibly more than one rendered symbol, in the user's chosen symbol set. This is because the user's chosen set may not have a single symbol to represent that one concept.
There would exist a mapping from Bliss concept ID to the user's chosen symbol set, which would allow a UA to render the one (or more) symbols for that Bliss concept ID.
The mapping would ideally in future be provided along with a given symbol set, but it could be created by anyone.
Maybe we should also note that it shouldn't be assumed that one concept ID maps to one symbol—but either way, that should not have to concern authors.
As discussed on yesterday's call, we should make some clarifications to the spec around the use of symbols, both from an author mark-up perspective, and from a renderer perspective. This is about the rendering side of things. Whilst the spec is primarily about author mark-up, it's important for authors to have some expectation of how things may be rendered, and it's important for UA/extension/AT implementers to be measured that the rendering process is simple.
We want to make clear that, with our Bliss concepts look-up, and mappings to other symbol sets, rendering is relatively simple, and would benefit from these look-ups. Here are some points that we need to ensure are clear:
The Bliss concept ID may map to one rendered symbol, or possibly more than one rendered symbol, in the user's chosen symbol set. This is because the user's chosen set may not have a single symbol to represent that one concept.
There would exist a mapping from Bliss concept ID to the user's chosen symbol set, which would allow a UA to render the one (or more) symbols for that Bliss concept ID.
The mapping would ideally in future be provided along with a given symbol set, but it could be created by anyone.
Maybe we should also note that it shouldn't be assumed that one concept ID maps to one symbol—but either way, that should not have to concern authors.