[INTRO] Reading the spec from the catalog side

[Kyoung Min Kim <ka...@gmail.com>]

Hello,

My name is Kyoung Min Kim. I work on metadata catalog tooling in South
Korea, and I've spent the last few weeks reading the core spec and the
ontology draft closely, building small models against them to find where the
documents and the validator disagree.

Two things have come out of that so far. PR #330 makes the validator warn
when a relationship's to_columns doesn't cover a declared key on the target
dataset. And I left a comment on discussion #29 about metric-level filters,
based on a set of models where a conditional metric encoded as
SUM(CASE WHEN ...) gives different results from the same logic expressed as
a WHERE filter, once a time grouping is involved.

The question I keep coming back to, coming from the catalog side, is how
much of a semantic model a downstream engine can reconstruct from an Ossie
document alone, and what has to fall back to custom_extensions. That has
pulled me toward the Ontology and Catalog working groups. I saw the notes
from the ontology sync on 8/11 in the archives. Could someone point me at
how
to join those meetings?

I'd like to keep working on the validation side for now. If there are open
items there that nobody has picked up, I'm interested.

Thanks,
Kyoung Min Kim