You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've filed issue #1304 and #1305 regarding IRI usage and extension keywords.
As discussed in #1140, we are likely to move $dynamic* to using non-IRI identifiers. This is an architectural change from needing to maintain a mapping of IRIs to schemas (or resolve them on demand) to needing to maintain (or resolve on demand) a mapping of an arbitrary keyword+value. For $dynamic*, this involves dynamic scope associations and lookups, but we could conceive of lexical identifier/lookup behavior as well.
This is not about whether we should allow or forbid dynamic behavior. This is about whether we should support assigning identifiers and looking them up in general for extension keywords. If we do think that this is worth supporting, we can further discuss dynamic vs static/lexical behaviors. But there's no need if we can't come up with use cases that make this worth supporting at all.
The text was updated successfully, but these errors were encountered:
I've filed issue #1304 and #1305 regarding IRI usage and extension keywords.
As discussed in #1140, we are likely to move
$dynamic*
to using non-IRI identifiers. This is an architectural change from needing to maintain a mapping of IRIs to schemas (or resolve them on demand) to needing to maintain (or resolve on demand) a mapping of an arbitrary keyword+value. For$dynamic*
, this involves dynamic scope associations and lookups, but we could conceive of lexical identifier/lookup behavior as well.This is not about whether we should allow or forbid dynamic behavior. This is about whether we should support assigning identifiers and looking them up in general for extension keywords. If we do think that this is worth supporting, we can further discuss dynamic vs static/lexical behaviors. But there's no need if we can't come up with use cases that make this worth supporting at all.
The text was updated successfully, but these errors were encountered: