Upload folder using huggingface_hub

LICENSE

@@ -1,21 +1,71 @@
-MIT License
-
-Copyright (c) 2026 Noé Flandre
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.
+# Herero Language Dataset - Multi-License Collection
+
+This repository is a **redistribution of materials from multiple upstream sources under different licenses**. It is a **collection**, not a derivative work. Each data source retains its original license.
+
+## Important Notice
+
+**This repository is a redistribution of materials from multiple upstream sources under different licenses. Each example remains subject to its original source license. Users are responsible for checking per-row licensing before reuse.**
+
+## Source Licenses
+
+| Source | License | Type |
+|--------|---------|------|
+| FineWeb-2 | ODC-By 1.0 | odc_by |
+| Omnilingual ASR | CC BY 4.0 | cc_by |
+| GlotCC-V1 | CC BY 4.0 | cc_by |
+| Herero Bible (1849) | Public Domain | public_domain |
+| Storybooks Namibia | CC BY 4.0 | cc_by |
+| FinePDFs | ODC-By 1.0 | odc_by |
+| Wikipedia Incubator | CC BY-SA 3.0 | cc_by_sa |
+
+## License Summaries
+
+### Public Domain (herero_bible_1849)
+No copyright restrictions. Published in 1849.
+
+### CC BY 4.0 - Attribution (omnilingual_asr, glotcc_v1, storybooks_namibia)
+You are free to:
+- **Share**: Copy and redistribute the material
+- **Adapt**: Remix, transform, build upon the material
+
+Under these terms:
+- **Attribution**: You must give appropriate credit to the original source
+- **No additional restrictions**: You may not apply legal terms that restrict others
+
+### CC BY-SA 3.0 - Attribution-ShareAlike (wikipedia_incubator)
+Same as CC BY 4.0, plus:
+- **ShareAlike**: If you remix or transform the material, you must distribute your contribution under the same license
+
+Note: As a mere collection (not an adaptation), CC BY-SA does not apply to the aggregated dataset.
+
+### ODC-By 1.0 - Open Data Commons Attribution (fineweb2, finepdfs)
+You are free to:
+- **Share**: Copy, distribute, use the database
+- **Create**: Produce derivative works from the database
+- **Modify**: Transform, translate, build upon the database
+
+Under these terms:
+- **Attribution**: You must give appropriate credit to the original database
+- **Keep open**: If you publicize a modified version, you must also publish it under ODC-By
+- **No sublicensing**: You may not sublicense the database under different terms
+
+See: https://opendatacommons.org/licenses/by/1-0/
+
+## Per-Row Licensing
+
+Each row in the dataset carries these provenance columns:
+- `source`: The data source name
+- `original_license`: The original license of that source
+- `original_url`: Link to the original source
+- `changes_made`: Description of any transformations applied
+
+**Users must check the `original_license` field for each example to determine applicable terms.**
+
+## Disclaimer
+
+This documentation is provided for informational purposes only and does not constitute legal advice. For questions about licensing, consult the original license texts or seek legal counsel.
+
+## See Also
+
+- `sources.csv` - Full source manifest with attribution details
+- `README.md` - Dataset documentation

README.md

@@ -1,38 +1,18 @@
 ---
-annotations_creators:
-- no-annotation
+license: other
+license_name: multiple-licenses
+pretty_name: Herero Language Dataset
 language:
 - hz
-language_creators:
-- found
-license: mit
-multilinguality:
-- monolingual
-pretty_name: Herero Language Dataset
-size_categories:
-- 1K<n<10K
-source_datasets:
-- original
-tags:
-- herero
-- otjiherero
-- namibia
-- bantu
-- low-resource
-- african-languages
-- pretraining
 task_categories:
-- text-generation
-- fill-mask
-task_ids:
-- language-modeling
+- text-classification
 ---
 
 # Herero Language Dataset
 
-[![GitHub Repository](https://img.shields.io/badge/Code-GitHub-blue)](https://github.com/NoeFlandre/herero-dataset)
+A **multi-license collection** of [Herero (Otjiherero)](https://en.wikipedia.org/wiki/Herero_language) text for NLP research. This is **not** a derivative work—sources are aggregated with original licenses preserved.
 
-A curated dataset of [Herero (Otjiherero)](https://en.wikipedia.org/wiki/Herero_language) language text for LLM pretraining and NLP research. Herero is a Bantu language spoken by ~250,000 people in Namibia and Botswana.
+**⚠️ WARNING: This repository is a redistribution of materials from multiple upstream sources under different licenses. Each example remains subject to its original source license. Users are responsible for checking per-row licensing before reuse.**
 
 ## Dataset Summary
 
@@ -40,9 +20,8 @@ A curated dataset of [Herero (Otjiherero)](https://en.wikipedia.org/wiki/Herero_
 |--------|-------|
 | Documents | 1,951 |
 | Words | 870,941 |
-| Characters | 5,882,568 |
+| Sources | 7 |
 | Language | Herero (hz) |
-| License | MIT |
 
 ## Sources
 
@@ -52,7 +31,7 @@ A curated dataset of [Herero (Otjiherero)](https://en.wikipedia.org/wiki/Herero_
 | Omnilingual ASR | 437 | 60,485 | CC BY 4.0 |
 | GlotCC-V1 | 20 | 75,982 | CC BY 4.0 |
 | Herero Bible (1849) | 81 | 26,807 | Public Domain |
-| Storybooks Namibia | 46 | 13,246 | CC BY 3.0/4.0 |
+| Storybooks Namibia | 46 | 13,246 | CC BY 4.0 |
 | FinePDFs | 25 | 17,800 | ODC-By 1.0 |
 | Wikipedia Incubator | 70 | 4,243 | CC BY-SA 3.0 |
 
@@ -66,18 +45,21 @@ A curated dataset of [Herero (Otjiherero)](https://en.wikipedia.org/wiki/Herero_
 
 ## Schema
 
+Each row includes provenance columns:
+
 | Field | Type | Description |
 |-------|------|-------------|
-| `id` | string | Unique document ID (source_prefix + number) |
+| `id` | string | Unique document ID |
 | `text` | string | Herero text (NFC normalized) |
 | `source` | string | Data source name |
-| `url` | string | Original source URL |
-| `license` | string | Source license |
+| `original_license` | string | Source license (check this!) |
+| `original_url` | string | Link to original source |
+| `changes_made` | string | Transformations applied |
+| `license_type` | string | License category |
+| `url` | string | Source URL |
 | `word_count` | int | Word count |
 | `char_count` | int | Character count |
 
-Splits are stratified by source to ensure representation in all splits.
-
 ## Usage
 
 ```python
@@ -85,16 +67,24 @@ from datasets import load_dataset
 
 dataset = load_dataset("NoeFlandre/herero-dataset")
 
-train_data = dataset["train"]
-val_data = dataset["validation"]
-test_data = dataset["test"]
+# Check per-row license BEFORE use
+for example in dataset["train"]:
+    print(f"License: {example['original_license']} - {example['source']}")
 ```
 
+## Collection Classification
+
+This dataset is a **collection**, not an adaptation:
+- Text is aggregated from sources without material modification
+- Original text is preserved (only formatting normalized)
+- No translation, rewriting, or merging across sources
+- Source boundaries maintained via `source` column
+
 ## Quality Assurance
 
-- **Deduplication**: SHA-256 content hashing
-- **Normalization**: Unicode NFC normalization
-- **Filtering**: Minimum 50-character threshold
+- **Deduplication**: Within-source SHA-256 content hashing
+- **Normalization**: Unicode NFC normalization (no text rewriting)
+- **Filtering**: Min-length 50 chars
 - **Reproducibility**: Fixed random seed (42)
 
 ## Known Limitations
@@ -104,28 +94,31 @@ test_data = dataset["test"]
 - Historical text uses 19th-century orthography
 - Speech transcripts may contain disfluencies
 
-## Licensing
+## Ethical Considerations
 
-Dataset: MIT
+- No personal data included
+- All content publicly available at time of scraping
+- Attribution preserved via `original_url` column
+- Users must respect individual source licenses
 
-Source content retains original licenses. See the [GitHub repository](https://github.com/NoeFlandre/herero-dataset) for full attribution.
+## Licensing Summary
 
-## Citation
+| License Type | Sources | Requirements |
+|-------------|---------|--------------|
+| Public Domain | 1 | None |
+| CC BY | 3 | Attribution |
+| CC BY-SA | 1 | Attribution + ShareAlike |
+| ODC-By | 2 | Attribution + Keep open |
 
-```bibtex
-@dataset{herero_dataset,
-  title = {Herero Language Dataset},
-  author = {Noé Flandre},
-  year = {2026},
-  url = {https://huggingface.co/datasets/NoeFlandre/herero-dataset}
-}
-```
+**You must check the `original_license` column for each example.**
+
+## Files
+
+- `data/train.parquet` - Training split
+- `data/validation.parquet` - Validation split
+- `data/test.parquet` - Test split
+- `sources.csv` - Full source manifest
 
-## Acknowledgments
+## Disclaimer
 
-- Wikimedia Foundation (Wikipedia Incubator)
-- Meta AI Research (Omnilingual ASR)
-- CIS LMU Munich (GlotCC)
-- HuggingFace (FineWeb-2, FinePDFs)
-- African Storybook / Saide
-- Rhenish Missionary Society
+This documentation is provided for informational purposes only and does not constitute legal advice. Uncertain licensing cases were reviewed conservatively. For legal questions, consult the original license texts or seek professional counsel.

data/test.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f99987a892cdd6ce0412c16d661581ca71dfad35071ff3047737a88cb9361600
-size 173533
+oid sha256:219eb2bcdb9fe2de11391076602dbbb8449aaf7db4282d8b00c3ca5523e7dacf
+size 176526

data/train.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0db2e66cd58a28ed484ee32144d7f7044631bbf171f8d53914749f847735fcc
-size 3177010
+oid sha256:573df50c75daedbd330896a1ed74f19ab293e0c46e0e6ea733355e41760dc617
+size 3181501

data/validation.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08a15ac7b67f526c9f473c0c18c6b57c9b651d32f5172f133fd1bc334585584c
-size 204546
+oid sha256:3b0484b22cb8a69e77bebdbd6b7abc2e43bff5149a38b0d396476bb91bfa2608
+size 207542

sources.csv

@@ -0,0 +1,8 @@
+source,original_license,license_type,original_url,attribution_text,changes_made,is_public_domain,docs,words
+fineweb2,ODC-By 1.0,odc_by,https://huggingface.co/datasets/HuggingFaceFW/fineweb-2,FineWeb-2 dataset by HuggingFace. See: https://huggingface.co/datasets/HuggingFaceFW/fineweb-2,Filtered to Herero (hz) language subset. Unicode NFC normalization. Min-length 50 chars.,False,1272,672378
+omnilingual_asr,CC BY 4.0,cc_by,https://huggingface.co/datasets/facebook/omnilingual-asr-corpus,Omnilingual ASR Corpus by Meta AI Research. See: https://huggingface.co/datasets/facebook/omnilingual-asr-corpus,Filtered to Herero language. Text extracted from transcripts. Unicode NFC normalization.,False,437,60485
+glotcc_v1,CC BY 4.0,cc_by,https://huggingface.co/datasets/cis-lmu/GlotCC-V1,GlotCC-V1 by CIS LMU Munich. See: https://huggingface.co/datasets/cis-lmu/GlotCC-V1,Filtered to Herero language via GlotLID. Text extracted from CommonCrawl. Unicode NFC normalization.,False,20,75982
+herero_bible_1849,Public Domain,public_domain,https://archive.org/stream/OmahungiOaEmboRaJehovaNaOmaimpurir/,Omahungi Oa Embo Ra Jehova Na Omaimpuriro Mo Otjiherero (1849) by Rhenish Missionary Society. Public Domain.,Text extracted from archive scan. Chapter/section segmentation. Unicode NFC normalization.,True,81,26807
+storybooks_namibia,CC BY 4.0,cc_by,https://storybooksnamibia.net/stories/hz/,Storybooks Namibia (Otjiherero) by African Storybook/Saide. See: https://storybooksnamibia.net,Scraped from storybooksnamibia.net. HTML cleaned. Unicode NFC normalization.,False,46,13246
+finepdfs,ODC-By 1.0,odc_by,https://huggingface.co/datasets/HuggingFaceFW/finepdfs,FinePDFs dataset by HuggingFace. See: https://huggingface.co/datasets/HuggingFaceFW/finepdfs,Filtered to Herero (hz) language subset. PDF text extraction. Unicode NFC normalization.,False,25,17800
+wikipedia_incubator,CC BY-SA 3.0,cc_by_sa,https://incubator.wikimedia.org/wiki/Wp/hz,Wikipedia Incubator (Herero). See: https://incubator.wikimedia.org/wiki/Wp/hz,Scraped via MediaWiki API. HTML templates removed. Unicode NFC normalization. MINIMAL PROCESSING - original text preserved.,False,70,4243