Source: /home/coolhand/html/datavis/data_trove/tools/fetchers/cache/submarine_cables.json
Saturn profiled 1 rows across 4 columns. The stats below are deterministic and machine-readable; the prose is a language-model interpretation of those stats (opt-in, added after the fact, never sees raw rows).
!pip install saturn-dissect
import subprocess
subprocess.run([
"saturn", "analyze", "/home/coolhand/html/datavis/data_trove/tools/fetchers/cache/submarine_cables.json",
"--findings", "data-trove-submarine-cable-map.json",
"--llm", "anthropic:default",
])This dataset is a GeoJSON FeatureCollection describing global submarine cables, stored as a single-row JSON file where the entire dataset is packed into nested fields. The top-level metadata columns ('name' and 'type') each contain exactly one value — 'submarine_cables' and 'FeatureCollection' respectively — confirming this is a wrapper structure rather than a flat table. The real analytical content is buried inside the 'features' and 'crs' columns, which were skipped during profiling because they contain complex nested objects (geometries, cable properties, coordinates). To get meaningful insight, the 'features' array needs to be unpacked into individual cable records before any visualization or analysis can begin.
citing: row_count · column_count · columns[name].top_value · columns[type].top_value · columns[features].alerts · columns[crs].alerts
| value | count | share |
|---|---|---|
| submarine_cables | 1 | 100.0% |
| value | count | share |
|---|---|---|
| FeatureCollection | 1 | 100.0% |
| column | kind | null % |
|---|---|---|
| type | categorical | 0.0% |
| name | categorical | 0.0% |
| crs | unknown | 0.0% |
| features | unknown | 0.0% |
This column is a GeoJSON type discriminator field, holding the constant value 'FeatureCollection' across all rows. With n=1, cardinality=1, top_rate=1.0, and entropy of 0.0, it is entirely invariant — it carries zero information. The 'long_tail' and 'imbalance' alerts are technically correct but misleading given the trivial dataset size.
Treatment: Drop before modelling; zero-variance constant column adds no predictive signal.
saturn.columns["type"].stats
| stat | value |
|---|---|
| n | 1 |
| nulls | 0 (0.0%) |
| unique | 1 |
| top_value | FeatureCollection |
| top_rate | 1 |
| cardinality | 1 |
| entropy | 0 |
| entropy_ratio | 0 |
| alert: long_tail | 1 singleton categories |
| alert: imbalance | top value is 100.0% of rows |
type.
| value | count | share |
|---|---|---|
| FeatureCollection | 1 | 100.0% |
This column appears to be a dataset or layer name/identifier, containing exactly one row with the single value 'submarine_cables'. With n=1, cardinality=1, null_rate=0.0, and top_rate=1.0, this column carries zero variance and no discriminative information whatsoever. The alerts for long_tail and imbalance are technically correct but trivially explained by the single-row, single-value nature of the data.
Treatment: Drop this column entirely — zero entropy and a single constant value provide no modelling signal.
saturn.columns["name"].stats
| stat | value |
|---|---|
| n | 1 |
| nulls | 0 (0.0%) |
| unique | 1 |
| top_value | submarine_cables |
| top_rate | 1 |
| cardinality | 1 |
| entropy | 0 |
| entropy_ratio | 0 |
| alert: long_tail | 1 singleton categories |
| alert: imbalance | top value is 100.0% of rows |
name.
| value | count | share |
|---|---|---|
| submarine_cables | 1 | 100.0% |
This column ('crs', likely a Coordinate Reference System or Course identifier) contains only a single row with no null values and no computable statistics, making meaningful profiling impossible. The 'skipped' alert indicates saturn bypassed full analysis, almost certainly due to the trivially small sample size (n=1). No distributional, uniqueness, or type information is available beyond the single observation.
Treatment: Withhold from modelling until more rows are available; re-profile with a full dataset to determine actual type and role.
saturn.columns["crs"].stats
| stat | value |
|---|---|
| n | 1 |
| nulls | 0 (0.0%) |
| unique | — |
| alert: skipped | no profiler for kind=unknown |
This column contains only a single row (n=1) and was skipped during profiling, yielding no computable statistics. With n=1 and null_rate=0.0, there is exactly one non-null value present, but its content, type, and distribution are entirely unknown from this evidence. No meaningful characterisation is possible beyond confirming the column is not empty.
Treatment: Inspect raw value manually; re-profile on a larger sample before deciding on encoding or dropping.
saturn.columns["features"].stats
| stat | value |
|---|---|
| n | 1 |
| nulls | 0 (0.0%) |
| unique | — |
| alert: skipped | no profiler for kind=unknown |
click to copy
@misc{saturn-data-trove-submarine-cable-map-2026,
author = {Steuber, Luke},
title = {Saturn reading: data trove submarine cable map},
year ={2026},
howpublished = {\url{https://dr.eamer.dev/saturn/view/data-trove-submarine-cable-map}},
note = {Profiled with saturn-dissect v0.2.0, prompt saturn-insight-v2, model anthropic:default},
}
Steuber, L. (2026). Saturn reading: data trove submarine cable map. Source: /home/coolhand/html/datavis/data_trove/tools/fetchers/cache/submarine_cables.json. Profiled with saturn-dissect v0.2.0 (saturn-insight-v2, anthropic:default). Retrieved from https://dr.eamer.dev/saturn/view/data-trove-submarine-cable-map