Artificial Wasteland artwaste.land

The measured ceiling of the surviving 78rpm record

6 of 7 conditions 6,062 rows · 3.1 MB · 5 files

One row per 78rpm disc side, carrying the highest frequency at which that disc's programme still stands above that same disc's own surface noise. 6,062 discs, drawn at random with a fixed seed from the Internet Archive's Great 78 Project.

The files

FileRowsBytessha256
edges.jsonl 6,062 3,085,541 f7d8b6fbe6547ae3dc81b7c435e94660d3f9d4dea62fcb0ffa70a286569bbec1
README.md n/a 4,782 b139968b83e909622637211acb842d8df2cbda3eb54766874bedd20cddf27051
schema.json n/a 5,001 6264018926eec301d5a3472733ab3c5f6a54a7e15a8c7906c68d2be2a549bf84
sources.json n/a 1,951 e7ca33711c4fa242293ac533da1954eb59e9900a75aa32dc72561c91eba43d14
validate.mjs n/a 15,198 67aceadfaea1f0b4611fa24f550c1b648437f25fbf20f19bf864292700aa69a2

Take the folder and check it yourself: node validate.mjs in the downloaded directory re-runs every row against the schema and every file against the digest above, with no dependencies and no network. The dataset's own full account is README.md.

A row, field by field

Origin is the column that decides whether we may give a field away at all. built is ours; redistributed is a source's own content, and it may only ship where that source's licence permits it.

FieldTypeOriginWhat it is
identifier string redistributed Internet Archive item identifier. One side of one disc.
year integer or null redistributed Item date as the archive records it. See the README: two sides of one disc disagree 1.8 per cent of the time.
edge_hz number or null built The edge: highest frequency in 200 Hz to 12 kHz whose loud-frame excess over quiet-frame level clears the disc's own asymptote by 6 dB. Null means nothing anywhere in the band cleared it.
asymptote_db number or null built The disc's own floor: median loud-over-quiet excess across 13 to 17 kHz.
noise_6k10k_db number or null built Mean quiet-frame level over 6 to 10 kHz, in dB, on the transfer's own scale.
prog_6k10k_db number or null built Mean loud-frame level over the same band.
seconds number or null built Duration of the transfer measured.
frames integer or null built Number of 8192-sample analysis frames at 4096 hop.
sample_rate_hz integer or null built Sample rate the measurement was made at. The statistic is rate dependent; see the README.
measured_sha256 string or null built sha256 of the exact bytes measured. The audio is not kept and not redistributed, so this is how you confirm you hold what we held.
measured_bytes integer or null built Length of those bytes.
condition string or null redistributed The archive's condition grading for the physical disc. A curator's judgement, not a measurement.
diameter_in integer or null redistributed Disc diameter in inches, 10 or 12.
country string or null redistributed The archive's coverage field.
genre string or null redistributed The archive's free-text genre, lowercased and with "fox trot" folded to "foxtrot".
publisher string or null redistributed Label as the archive records it.
playback_turnover_hz number or null redistributed Playback equalisation turnover stated in the item description, if stated.
playback_rolloff_db number or null redistributed Playback equalisation rolloff stated in the item description, if stated. Null does not mean flat.

Where it came from, and how we know we may pass it on

Great 78 Project item audio and item metadata

Internet Archive, with transfers by George Blood, L.P. · https://archive.org/details/georgeblood

Licence
Not established. Recorded from the source, not assumed.
How we know
not established
Accessed
2026-09-04

Read field by field on 2026-09-04: the collection metadata, its parent collection and a representative item carry no licenseurl, no rights field and no public-domain assertion. The Internet Archive asserts no licence and grants no permission, and that silence is recorded here as silence rather than upgraded into permission. No audio is redistributed by this dataset.

The edge measurement

Artificial Wasteland · https://artwaste.land/strata/the-ceiling-the-record-kept/

Licence
CC0 1.0 Universal (public domain dedication)
How we know
publisher statement
Accessed
2026-09-04

Ours, dedicated to the public domain. The generator is research/the-ceiling-the-record-kept/ and the estimator is edge.py.

Required attributionAudio transferred by George Blood, L.P. for the Internet Archive Great 78 Project. The edge measurement, and every column derived from the audio, are by the Artificial Wasteland.

The seven conditions, and what was measured

  1. Bytes 1 data file(s), 3,085,541 bytes, largest 2.9 MiB
  2. Shape 6,062 rows checked, 0 violation(s)
  3. Count 6,062 rows across 1 file(s), all counts agree
  4. Fixity 5 file(s) verified by sha256
  5. Provenance 2 source(s), each with a licence from the table (1 honestly UNKNOWN)
  6. Origin identifier: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; year: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; condition: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; diameter_in: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; country: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; genre: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; publisher: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; playback_turnover_hz: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution; playback_rolloff_db: redistributed from src:internet-archive-great78, whose licence (UNKNOWN) does not permit redistribution
  7. Rebuild research/data-room/emit/record-ceiling.mjs (byte-equality proved by: node research/data-room/publish.mjs --check)

This dataset does not meet all seven. That is printed rather than hidden, and the detail beside each failing condition is the measurement itself, not a summary of it.

← every dataset, and the three we could not publish