File size: 3,010 Bytes
6818084
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
---
license: other
license_name: proprietary-license-required
license_link: LICENSE
pretty_name: Usenet Corpus 1980–2013  Full (Samples)
size_categories:
  - 10K<n<100K
task_categories:
  - text-generation
  - text-classification
language:
  - en
  - pl
  - nl
  - es
  - fr
  - it
  - de
  - ru
  - ja
tags:
  - usenet
  - internet-history
  - forums
  - conversational
  - sample
---

# Usenet Corpus 1980–2013 — Full (Samples)

A small, browsable **showcase sample** of the *Usenet Corpus 1980–2013* (cleaned)
dataset — long-form, pre-web Usenet posts. This repo is a free preview; the full,
commercially-licensed corpus (408.0M posts, 103B tokens) is at:

- **Full cleaned dataset (gated):** https://huggingface.co/datasets/OwnedByDanes/Usenet-Corpus-1980-2013-Full
- **Threaded companion:** https://huggingface.co/datasets/OwnedByDanes/Usenet-Corpus-1980-2013-Threaded

## What's in this sample

Eleven gzipped JSON Lines files: one per hierarchy (about 5,000 posts each; ALT about 2,000),
plus a **Big 8 combined** and an **all-hierarchies combined** sample (about 9,000 posts
each). The per-hierarchy and combined files are independent draws and overlap.

Posts are selected as **whole conversations (threads) of at least 3 posts** — a
deliberate showcase cut so the sample contains coherent discussions rather than
isolated one-off posts. It is **not** representative of the full corpus's shape,
which is mostly single-post threads; it is curated for readability. Related posts
sit together within each file, and cross-post duplicates are removed within the
sample. These are the same posts as the *Threaded* sample repo, without the three
thread fields — see that repo if you want the conversation structure.

## Record schema

Each line is a JSON object:

| Field | Type | Description |
| --- | --- | --- |
| `id` | string | Stable per-post identifier (`msg-` + hash of the original Message-ID). |
| `group` | string | Newsgroup the post was archived under. |
| `date` | string | Original post date (RFC 2822). |
| `author` | string | Author display name as posted. Email addresses are redacted. |
| `subject` | string | Subject line. |
| `text` | string | Post body. |

## Provenance and safety

Drawn from the cleaned, de-duplicated corpus: binary attachments and
`alt.binaries.*` removed, adult groups excluded, author emails redacted,
Message-IDs replaced with hashes, and content matching ClamAV malware signatures
removed (the corpus passes a whole-file ClamAV scan with no detections). Content is
otherwise preserved as posted — unmoderated historical text that may be offensive,
biased, or wrong, and should be treated as a primary historical source.

## License

**Proprietary.** © 2026 OwnedByDanes LLC. This sample is provided for evaluation
only and conveys no right to use, copy, redistribute, or train on the data. Any use
requires a separate written commercial license from OwnedByDanes LLC. See the
`LICENSE` file. To request a license: **todd@ownedbydanes.com**.