Q1, run 2026-11-06-0519-sonnet-query

← back to the run

How many distinct people do the author records named "Mara Ellison", "M. Ellison" and "Ellison, Mara J." stand for?

Tests: Identity: A01, A02 and A03 share an ORCID; A04, also 'M. Ellison', has another.

Expected: count = 3

✅ D1 sql run 1 graded by code · query session

3. A01, A02, A03 and A101 share ORCID 0000-0002-9841-6229, while A04 (0000-0004-8662-8360) and A102 (0000-0008-4738-8328) are different people.

JSON: {"count": 3}

❌ D1 sql run 2 graded by code · query session

1. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229, so they are one person. The other "M. Ellison" records (A04, A102) have different ORCIDs.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D1 sql run 3 graded by code · query session

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229. A04, A101 and A102 are other records named "M. Ellison", and A101 shares the same ORCID.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D1 sql run 4 graded by code · query session

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person (A101 shares it too).

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D1 sql run 5 graded by code · query session

1. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D2 sql+markdown run 1 graded by code · query session

1. A01, A02, A03 and A101 share ORCID 0000-0002-9841-6229, so they are one person. A04 and A102 are also named M. Ellison but have different ORCIDs.

JSON: {"count": 1}

Why: count: expected 3, got 1

✅ D2 sql+markdown run 2 graded by code · query session

3 people. A01, A03 and A02 share one ORCID (0000-0002-9841-6229), so they are one person. The other records named "M. Ellison" (A04, A102) have different ORCIDs and are different people, so the three names cover at most that person plus A04 and A102 if they are counted.

JSON: {"count": 3}

❌ D2 sql+markdown run 3 graded by code · query session

1. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229, so they are one person.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D2 sql+markdown run 4 graded by code · query session

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D2 sql+markdown run 5 graded by code · query session

1. A01, A02, A03 (and A101) share ORCID 0000-0002-9841-6229, so they are one person.

JSON: {"count": 1}

Why: count: expected 3, got 1

✅ D3 sql+executed run 1 graded by code · query session

3. The names cover A01 (which merges A01, A02, A03 and A101), A04 and A102, who have different ORCIDs.

JSON: {"count": 3}

✅ D3 sql+executed run 2 graded by code · query session

3. The names resolve to A01 (Mara Ellison, merging A01/A02/A03/A101), A04 and A102. A04 and A102 are both "M. Ellison" but have different ORCIDs.

JSON: {"count": 3}

✅ D3 sql+executed run 3 graded by code · query session

3. A01 covers "Mara Ellison" and "Ellison, Mara J." (author records A01 A02 A03 A101). "M. Ellison" is two different people, A04 and A102, with different ORCIDs.

JSON: {"count": 3}

✅ D3 sql+executed run 4 graded by code · query session

3. "Mara Ellison" and "Ellison, Mara J." both fall under A01, whose merged author records are A01, A02, A03 and A101. "M. Ellison" is two different people, A04 and A102, who have different ORCIDs.

JSON: {"count": 3}

❌ D3 sql+executed run 5 graded by code · query session

1 person. A01 merges author records A01, A02, A03 and A101, which I take to cover all three name variants. The other "M. Ellison" people, A04 and A102, have different ORCIDs. I could not see which author ID carries which name string, so this is an inference.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D4 sparql run 1 graded by code · query session

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person. A101 ("M. Ellison") also carries that ORCID, while A04 and A102 are different people.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D4 sparql run 2 graded by code · query session

1. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D4 sparql run 3 graded by code · query session

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D4 sparql run 4 graded by code · query session

1. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229, so they are one person (A101 shares it too). A04 and A102 are different people despite the name "M. Ellison".

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D4 sparql run 5 graded by code · query session

1. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D5 sparql+ontology run 1 graded by code · query session

1 person. A01, A02 and A03 share ORCID 0000-0002-9841-6229. A04 and A101 are separate records: A101 shares that ORCID, A04 does not.

JSON: {"count": 1}

Why: count: expected 3, got 1

✅ D5 sparql+ontology run 2 graded by code · query session

3. The records are one person (A01, A02, A03, A101, ORCID 0000-0002-9841-6229) and two different "M. Ellison" people (A04 and A102, with different ORCIDs).

JSON: {"count": 3}

❌ D5 sparql+ontology run 3 graded by code · query session

1. A01, A02, A03 (and A101) share ORCID 0000-0002-9841-6229, so they are one person.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D5 sparql+ontology run 4 graded by code · query session

1. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229, so they are one person.

JSON: {"count": 1}

Why: count: expected 3, got 1

❌ D5 sparql+ontology run 5 graded by code · query session

1 person. A01, A02 and A03 all carry ORCID 0000-0002-9841-6229, so they are the same person.

JSON: {"count": 1}

Why: count: expected 3, got 1