probes: 16
This data as json
| id | form_id | form_name | file | kind | corpus_id | written | run | result_count | result_of | result_date | matches_excerpt | note | code |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 16 | form-064 | Period typographic conventions | probes/2026-08/line-064-arginine-titles.py | python | corpus-pubmed-arginine-pre1990-3642 | 2026-09-09 | python3 tools/run_probe.py probes/2026-08/line-064-arginine-titles.py <corpus .md file> | 234 | 3642 | 2026-09-09 | 0 | The August note says 184; the rule it used was not written down. This probe (title carries no lower-case letter) gives 234, almost all from the 1960s. The probe's figure is the one to cite; the excerpt keeps its own. | """Fault Atlas probe — catalogue line 64 (form-064), origin observation. Corpus: PubMed abstracts on arginine, published before 1990-01-01, harvested 2026-07-29 by efetch (3,642 records; see corpora/pubmed-arginine-pre1990-3642.json). The corpus file is a Markdown list: one record per "## n. PMID x · year" heading, the title in bold on the next line. Question: how many titles are written in full capitals (a dated convention)? Run: python3 tools/run_probe.py probes/2026-08/line-064-arginine-titles.py <corpus .md file> Result: 234 of 3,642 titles carry no lower-case letter (232 of them from the 1960s). The August 2026 note counted 184 with a rule that was not preserved; the figure below is the one this probe gives, and it is the one to cite from now on. """ import re CORPUS_PROBE = True LINE = 64 TITLE = "title in full capitals (dated typographic convention)" REC = re.compile(r'^## \d+\. PMID (\d+) · (\d{4})\n\*\*(.*?)\*\*', re.M) def population(text): return len(REC.findall(text)) def probe(text): out = [] for pmid, year, title in REC.findall(text): if re.search(r'[A-Z]', title) and not re.search(r'[a-z]', title): out.append(f"PMID {pmid} ({year}): {title}") return out |