saturn·

← all readings

data trove world languages integrated

saturn notebook · generated 2026-06-22 Report Notebook

Overview

Source: /home/coolhand/html/datavis/data_trove/data/linguistic/world_languages_integrated.json

Saturn profiled 7,130 rows across 8 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).

[2]:
!pip install saturn-dissect
import subprocess
subprocess.run([
    "saturn", "analyze", "/home/coolhand/html/datavis/data_trove/data/linguistic/world_languages_integrated.json",
    "--findings", "data-trove-world-languages-integrated.json",
    "--llm", "anthropic:default",
])

Summary confidence: medium

This dataset is a reference catalogue of 7,130 world languages, each identified by a unique ISO 639-3 three-letter code alongside its name and several linked data sources (Glottolog, Joshua Project, speaker counts, and more). Every row is distinct — no duplicate language codes or names — making this primarily a lookup/reference table. Two things stand out for closer inspection: first, the name column reveals notable clusters around directional qualifiers (Southern, Northern, Western, Eastern) and language families like Zapotec (58), Mixtec (52), and Naga (49), suggesting rich geographic and genealogical structure worth exploring. Second, 'sign' appears 152 times in language names, indicating a surprisingly large representation of sign languages across the world's documented tongues.

citing: row_count · column_count · columns[0].n_unique · columns[1].n_unique · columns[1].stats.len_max · columns[1].stats.len_mean · columns[1].top_words · columns[1].stats.one_word_rate

Out[4]:

saturn.schema() · 8 columns

column kind n null% unique alerts
iso_639_3 text 7,130 0.0% 7,130 near_unique one_word short_text
name text 7,130 0.0% 7,130 near_unique one_word
joshua_project unknown 7,130 0.0% — skipped
glottolog unknown 7,130 0.0% — skipped
language_history unknown 7,130 0.0% — skipped
us_indigenous unknown 7,130 0.0% — skipped
speaker_count unknown 7,130 0.0% — skipped
data_sources unknown 7,130 0.0% — skipped
Fig 1.
name · Top words in language names — look for dominant directional terms (Southern, Northern) and prolific language families like Zapotec and Mixtec.
Show data table
charscount
1 – 225
2 – 3195
3 – 4761
4 – 51106
5 – 61096
6 – 7847
7 – 8529
8 – 9351
9 – 10278
10 – 12238
12 – 13213
13 – 14215
14 – 15191
15 – 16202
16 – 17149
17 – 1895
18 – 1975
19 – 2072
20 – 2166
21 – 2270
22 – 23154
23 – 2450
24 – 2526
25 – 2630
26 – 2723
27 – 288
28 – 297
29 – 3010
30 – 3114
31 – 326
32 – 348
34 – 352
35 – 3610
36 – 371
37 – 384
38 – 390
39 – 401
40 – 410
41 – 421
42 – 431
Fig 2.
name · Distribution of language name lengths — most names are short (median 7 characters) but a long tail reaches up to 43 characters.
Show data table
charscount
1 – 225
2 – 3195
3 – 4761
4 – 51106
5 – 61096
6 – 7847
7 – 8529
8 – 9351
9 – 10278
10 – 12238
12 – 13213
13 – 14215
14 – 15191
15 – 16202
16 – 17149
17 – 1895
18 – 1975
19 – 2072
20 – 2166
21 – 2270
22 – 23154
23 – 2450
24 – 2526
25 – 2630
26 – 2723
27 – 288
28 – 297
29 – 3010
30 – 3114
31 – 326
32 – 348
34 – 352
35 – 3610
36 – 371
37 – 384
38 – 390
39 – 401
40 – 410
41 – 421
42 – 431
Fig 3.
iso_639_3 · All ISO 639-3 codes are exactly 3 characters long, confirming full standard compliance across all 7,130 entries.
Show data table
charscount
2 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 37130
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 40
Fig 4.
Per-column null rate across the corpus. Columns are ordered by input position.
Show data table
columnkindnull %
iso_639_3text0.0%
nametext0.0%
joshua_projectunknown0.0%
glottologunknown0.0%
language_historyunknown0.0%
us_indigenousunknown0.0%
speaker_countunknown0.0%
data_sourcesunknown0.0%

iso_639_3 text identifier

This column contains ISO 639-3 language codes — the international standard three-letter identifiers for individual human languages. Every value is exactly 3 characters long (len_min=3, len_max=3, len_mean=3.0), all lowercase, and completely unique across all 7,130 rows with zero nulls or duplicates. The 7,130 distinct codes is strikingly close to the total number of living languages catalogued by ISO 639-3 (~7,000+), suggesting this may be a near-complete reference table of language codes.

Treatment: Use as a primary key or join key; left-join to enrich with language metadata (name, family, region).

anthropic:default · confidence high
Out[10]:

saturn.columns["iso_639_3"].stats

statvalue
n7,130
nulls0 (0.0%)
unique7,130
len_min 3
len_max 3
len_mean 3
len_median 3
len_p95 3
word_mean 1
word_median 1
n_empty 0
n_duplicates 0
duplicate_rate 0
vocab_size 7,130
readability_flesch_mean 120.4
emoji_rate 0
url_rate 0
one_word_rate 1
allcaps_rate 0
boilerplate_rate 0
alert: near_unique100.0% of rows are unique strings
alert: one_word100.0% rows are a single word
alert: short_text95th-percentile length under 20 chars
Fig 5.
Character-length distribution for iso_639_3.
Show data table
charscount
2 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 37130
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 30
3 – 40

name text label

This column contains names of human languages or dialects, evidenced by the dominant top words ('language', 'sign', 'zapotec', 'mixtec', 'naga') and directional qualifiers ('southern', 'northern', 'western', 'eastern', 'central') typical of language taxonomy naming conventions. All 7,130 rows are unique with zero duplicates and zero nulls, making this a near-perfect identifier. The one-word rate of 72.9% and a mean word count of 1.37 reflect the compact, often single-token nature of language names, while the max length of 43 characters accommodates longer multi-word dialect names. The vocabulary size of 7,124 against 7,130 rows confirms near-total uniqueness with only minimal token reuse.

Treatment: Use as a human-readable label or entity key; do not encode directly — map to a numeric ID or embed with a language-model encoder if semantic similarity is needed.

anthropic:default · confidence high
Out[13]:

saturn.columns["name"].stats

statvalue
n7,130
nulls0 (0.0%)
unique7,130
len_min 1
len_max 43
len_mean 8.994
len_median 7
len_p95 21
word_mean 1.372
word_median 1
n_empty 0
n_duplicates 0
duplicate_rate 0
vocab_size 7,124
readability_flesch_mean 56.15
emoji_rate 0
url_rate 0
one_word_rate 0.7292
allcaps_rate 0
boilerplate_rate 0
alert: near_unique100.0% of rows are unique strings
alert: one_word72.9% rows are a single word
Fig 6.
Character-length distribution for name.
Show data table
charscount
1 – 225
2 – 3195
3 – 4761
4 – 51106
5 – 61096
6 – 7847
7 – 8529
8 – 9351
9 – 10278
10 – 12238
12 – 13213
13 – 14215
14 – 15191
15 – 16202
16 – 17149
17 – 1895
18 – 1975
19 – 2072
20 – 2166
21 – 2270
22 – 23154
23 – 2450
24 – 2526
25 – 2630
26 – 2723
27 – 288
28 – 297
29 – 3010
30 – 3114
31 – 326
32 – 348
34 – 352
35 – 3610
36 – 371
37 – 384
38 – 390
39 – 401
40 – 410
41 – 421
42 – 431

joshua_project unknown label

This column references the Joshua Project, a well-known ethnoreligious people-group classification system commonly used in missiology and demographic datasets. The profiler returned a 'skipped' alert with no computed stats or uniqueness count, meaning the column type was not resolved and no distributional analysis was available. With 7,130 non-null rows and zero null rate, the data is fully populated, but nothing further can be inferred about cardinality, value distribution, or data type from the evidence provided.

Treatment: Resolve column type manually, then assess cardinality to determine whether to treat as a categorical label or foreign key joining to a Joshua Project people-group reference table.

anthropic:default · confidence low
Out[16]:

saturn.columns["joshua_project"].stats

statvalue
n7,130
nulls0 (0.0%)
unique—
alert: skippedno profiler for kind=unknown

glottolog unknown foreign_key

This column contains Glottolog codes, which are standardized four-letter+four-digit identifiers for the world's languages and language families maintained by the Glottolog database. The profiler skipped detailed analysis (kind='unknown', no stats, null_rate=0.0), so cardinality and value distribution are unavailable. With 7130 rows and zero nulls, the column is fully populated, but uniqueness cannot be assessed from the available evidence.

Treatment: Left-join on this code against the Glottolog reference database to enrich with language family, macroarea, and geographic metadata.

anthropic:default · confidence low
Out[18]:

saturn.columns["glottolog"].stats

statvalue
n7,130
nulls0 (0.0%)
unique—
alert: skippedno profiler for kind=unknown

language_history unknown other

This column contains an unknown data type that Saturn skipped during profiling, likely a complex or nested structure (e.g., JSON, array, or serialized object) representing a history of languages associated with each record. With 7,130 rows, zero nulls, and no stats computed, the actual content and distribution are entirely opaque from this evidence alone. The 'skipped' alert means no cardinality, frequency, or value-level analysis was performed, so downstream usability is unknown.

Treatment: Inspect raw values to determine structure (e.g., parse JSON/arrays), then flatten or encode before modelling.

anthropic:default · confidence low
Out[20]:

saturn.columns["language_history"].stats

statvalue
n7,130
nulls0 (0.0%)
unique—
alert: skippedno profiler for kind=unknown

us_indigenous unknown feature

The column 'us_indigenous' likely represents a binary or categorical indicator of US Indigenous identity/ethnicity for 7,130 records. The profiler emitted a 'skipped' alert with no computed stats or uniqueness count, meaning the column's data type or content prevented standard analysis. No nulls are present, but without distribution or cardinality data, the actual value composition (e.g., boolean flags, counts, coded strings) cannot be determined from this evidence alone.

Treatment: Inspect raw values to confirm encoding (boolean, 0/1, string category), then encode appropriately before modelling.

anthropic:default · confidence low
Out[22]:

saturn.columns["us_indigenous"].stats

statvalue
n7,130
nulls0 (0.0%)
unique—
alert: skippedno profiler for kind=unknown

speaker_count unknown feature

The column 'speaker_count' is likely a numeric count of speakers per record (e.g., per document, meeting, or audio segment). No distributional statistics or uniqueness data were computed — the profiler skipped this column, possibly due to an unrecognised dtype or an upstream parsing issue. With 7,130 non-null rows and zero nulls, the data is complete, but no further characterisation is possible from the available evidence.

Treatment: Resolve the 'skipped' profiling alert by checking dtype (may be stored as string or object); cast to integer and re-profile before using as a numeric feature.

anthropic:default · confidence low
Out[24]:

saturn.columns["speaker_count"].stats

statvalue
n7,130
nulls0 (0.0%)
unique—
alert: skippedno profiler for kind=unknown

data_sources unknown other

The column 'data_sources' has 7130 rows with zero nulls, but saturn skipped profiling it entirely — no type was resolved, no unique count was computed, and no statistics were collected. This suggests the column contains a complex or nested type (e.g., JSON, arrays, or mixed structures) that the profiler could not parse. No further characterisation is possible from the available evidence.

Treatment: Inspect raw values to determine structure (e.g., JSON array, delimited string), then parse and flatten before any downstream use.

anthropic:default · confidence low
Out[26]:

saturn.columns["data_sources"].stats

statvalue
n7,130
nulls0 (0.0%)
unique—
alert: skippedno profiler for kind=unknown

How to cite

click to copy

BibTeX
@misc{saturn-data-trove-world-languages-integrated-2026,
  author       = {Steuber, Luke},
  title        = {Saturn reading: data trove world languages integrated},
  year         ={2026},
  howpublished = {\url{https://dr.eamer.dev/saturn/view/data-trove-world-languages-integrated}},
  note         = {Profiled with saturn-dissect v0.2.0, prompt saturn-insight-v2, model anthropic:default},
}
APA
Steuber, L. (2026). Saturn reading: data trove world languages integrated. Source: /home/coolhand/html/datavis/data_trove/data/linguistic/world_languages_integrated.json. Profiled with saturn-dissect v0.2.0 (saturn-insight-v2, anthropic:default). Retrieved from https://dr.eamer.dev/saturn/view/data-trove-world-languages-integrated