Skip to content

Commit cf79792

Browse files
justint-dbIsaac
andcommitted
[core-spec] Mark spec layers doc as informational; define Layer 3 by Layer 2 equivalence
Address review feedback: update the status line, state up front that this is an informational overview pointing to the normative layer specs, and clarify that Layer 2 is the semantic target for Layer 3 rather than a required execution path. Co-authored-by: Isaac <no-reply@databricks.com>
1 parent fc3c182 commit cf79792

1 file changed

Lines changed: 18 additions & 12 deletions

File tree

‎core-spec/spec_layers.md‎

Lines changed: 18 additions & 12 deletions
Original file line numberDiff line numberDiff line change
@@ -19,13 +19,17 @@
1919

2020
# Ossie Specification Layers
2121

22-
**Status:** Draft for the Ossie Semantics Working Group.
22+
**Status:** Informational / Architecture Overview (Draft).
2323
**Contributors:** Justin Talbot, Will Pugh, Chris Eubank.
2424

2525
---
2626

2727
## Overview
2828

29+
This document is an informational architecture overview. It describes how the Ossie
30+
specifications fit together and links to the specification for each layer. It does not
31+
define normative requirements itself; those are in each layer's specification.
32+
2933
Ossie is organized into four specification layers, each addressing a distinct concern:
3034

3135
| Layer | Name | Role |
@@ -40,11 +44,9 @@ use to write it. Layers 2 and 3 are query interfaces that sit on top of a deploy
4044
they differ in how much SQL knowledge they require from the consumer. Layer 4 is a separate
4145
workstream covering ontological concepts that SQL does not naturally express.
4246

43-
Each layer can be adopted independently. Layers 2 and 3 compose: a wide-table query (Layer 3)
44-
can be rewritten as a SQL measures query (Layer 2), so correctness guarantees flow through.
45-
46-
This document describes each layer at a high level. Each is specified in full in its own
47-
document (linked below).
47+
Each layer can be adopted independently. Layers 2 and 3 compose: the meaning of a wide-table
48+
query (Layer 3) is defined by an equivalent SQL measures query (Layer 2), so correctness
49+
guarantees flow through.
4850

4951
---
5052

@@ -94,8 +96,8 @@ is required from the consumer.
9496

9597
This interface trades explicit query control for simplicity, making it accessible to ad-hoc
9698
users, AI agents, and BI tools without their own data model. Measure correctness is inherited
97-
from Layer 2: a wide-table query is rewritten into a SQL measures query (with heuristic join
98-
and filter choices), so the same grain-safe evaluation guarantee holds.
99+
from Layer 2: a wide-table query is defined as equivalent to a SQL measures query (with
100+
heuristic join and filter choices), so the same grain-safe evaluation guarantee holds.
99101

100102
**Specification:** *(open PR: [apache/ossie#246](https://github.com/apache/ossie/pull/246))*
101103

@@ -112,10 +114,14 @@ group (`#ossie-ontology-wg`) and is a separate workstream from the query interfa
112114

113115
## Composition
114116

115-
Layers 2 and 3 compose: a wide-table (Layer 3) query is rewritten into a SQL measures
116-
(Layer 2) query with heuristic join and filter choices. This means correctness guarantees
117-
flow from Layer 2 through Layer 3—a consumer using the declarative interface gets the same
118-
grain-safe measure evaluation as one writing SQL directly.
117+
Layers 2 and 3 compose: the meaning of a wide-table (Layer 3) query is defined by an
118+
equivalent SQL measures (Layer 2) query, with heuristic join and filter choices. This means
119+
correctness guarantees flow from Layer 2 through Layer 3—a consumer using the declarative
120+
interface gets the same grain-safe measure evaluation as one writing SQL directly.
121+
122+
Layer 2 is the semantic target for Layer 3, not a required execution path. An implementation
123+
does not have to translate Layer 3 queries into Layer 2 SQL text; it only has to return the
124+
same results as the equivalent Layer 2 query.
119125

120126
Providers may support one query interface layer or both, at varying compliance levels. The
121127
layered model makes partial adoption coherent: a tool that only needs SQL measures adopts

0 commit comments

Comments
 (0)