(rai-site) branch main updated: Adding Sourcelume article

[ak...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

andrewmusselman pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/rai-site.git


The following commit(s) were added to refs/heads/main by this push:
     new 5aff34d  Adding Sourcelume article
5aff34d is described below

commit 5aff34d2f3bd0addcdc5888a6439da1ae30af30e
Author: Andrew Musselman <an...@gmail.com>
AuthorDate: Tue Oct 27 09:17:28 2026 -0700

    Adding Sourcelume article
---
 content/blog/tiny-model-full-provenance.md | 42 ++++++++++++++++++++++++++++++
 content/pages/whats-new.md                 |  7 ++++-
 2 files changed, 48 insertions(+), 1 deletion(-)

diff --git a/content/blog/tiny-model-full-provenance.md b/content/blog/tiny-model-full-provenance.md
new file mode 100644
index 0000000..c086ab2
--- /dev/null
+++ b/content/blog/tiny-model-full-provenance.md
@@ -0,0 +1,42 @@
+Title: Tiny model, full provenance — where Apache Sourcelume is, and where it is heading
+Date: 2026-10-27
+Byline: Achim Nierbeck and Jamie Goodyear
+Topics: sourcelume, provenance, projects
+Summary: Apache Sourcelume now records provenance for models as well as datasets, and a sample project trains a tiny model that writes its own training provenance record as part of the run.
+license: https://www.apache.org/licenses/LICENSE-2.0
+
+Ask anyone who ships a language model a simple question: which data was this trained on, and what license comes with it? The honest answer today is a shrug with a bibliography. Training corpora are assembled from dozens or hundreds of sources, licensing information is missing, wrong, or unverifiable — a 2024 audit found license-omission rates above 70% across popular dataset-hosting sites — and what a producer does publish is usually a claim you have to take on faith: a paragraph in a pa [...]
+
+## What Sourcelume is
+
+That is the gap [Apache Sourcelume](https://sourcelume.apache.org/) is built for. It is not a certification authority and it does not decide whether anyone's licensing claims are true. It is plumbing: a versioned, JSON-LD metadata specification for dataset origin, custody chain, and licensing; a reference registry built on Apache Atlas that stores and serves those records through plain REST APIs; and tooling that turns "documenting your data" from a favor into a build step. The point is  [...]
+
+## Datasets, and then models
+
+The dataset side of this is already real. The spec's `ProvenanceRecord` has been exercised against a public example we could find: the [DFM Mímir training corpus](https://huggingface.co/danish-foundation-models/DFM-Mimir). We verified dataset by dataset, straight from the report's own tables. But records about datasets were only ever half of the question.
+
+On the model side, the situation is inverted: model producers know exactly what they trained on, but there has been no neutral place to say it. When a downstream consumer recently asked whether they could fetch structured provenance for a model — machine-readable, not prose — the piece that would answer them did not exist. The mental model that emerged in the discussion was an SBOM for models: software bills of materials let a shipped artifact declare its ingredients; a model should be a [...]
+
+## A record you can watch being born
+
+Which brings us to the demo, because a record type is only as real as a record you can watch being born. The [`apache/sourcelume-sample`](https://github.com/apache/sourcelume-sample) repository trains a character-level language model of 71,891 parameters, on a corpus every reader knows is public domain: *Alice's Adventures in Wonderland*, Project Gutenberg ebook #11, 1865. Then, in the same run, the trainer writes the model's provenance record: a `TrainingProvenanceRecord` pointing at th [...]
+
+This is where the sober result lands. Not a claim restated in a PDF paper — the run's own tokens, shares, and loss now sit in Sourcelume, written by the trainer that produced them. And as the sample shows, the whole thing drops straight into existing build tooling. Nothing is reconstructed by hand after the fact: the record is written where the data is already being used, in the same pipeline that trains the model.
+
+Our sample code lives at [github.com/apache/sourcelume-sample](https://github.com/apache/sourcelume-sample) for those who would like to dive in.
+
+## What it does not do yet
+
+As it stands today, Sourcelume can show the data that was deposited in it. What it does not do yet — and where community discussion still needs to happen — is verify that any of it is true. Today, Sourcelume's job is exactly this: information can be filed and fetched back by machines. Our community is working towards its initial release goals, with attestation features becoming our next target.
+
+## Getting involved
+
+Sourcelume is an open project, and there is room to help shape what comes next. Whether you're interested in provenance metadata, dataset licensing, model training, API design, or responsible AI more broadly, we welcome people who want to help build and test the specification, contribute to the reference implementation, improve the examples and documentation, or participate in discussions about how provenance claims should be represented and eventually verified.
+
+If this work interests you, join the Apache Sourcelume community. Explore the code and examples, try the sample training pipeline, share your feedback, or get involved in the discussions around the project's next milestones. The goal is not just to make provenance machine-readable, but to build an open, community-driven infrastructure that makes it easier to understand where datasets and models come from — and what we can actually verify about them.
+
+Apache Sourcelume is an ASF Top-level Project. For more information, visit [sourcelume.apache.org](https://sourcelume.apache.org/).
+
+---
+
+*Achim Nierbeck is an ASF Member and a member of the Apache Karaf and Apache Sourcelume PMCs. Jamie Goodyear is an ASF Member and Vice President, Apache Sourcelume.*
diff --git a/content/pages/whats-new.md b/content/pages/whats-new.md
index 12255af..9f9980b 100644
--- a/content/pages/whats-new.md
+++ b/content/pages/whats-new.md
@@ -1,8 +1,13 @@
 Title: What's New
 license: https://www.apache.org/licenses/LICENSE-2.0
 
-News items will be shared on the [general@rai.apache.org](https://lists.apache.org/list.html?general@rai.apache.org) list — [subscribe](mailto:general-subscribe@rai.apache.org) today!  Also be sure to check out the [Blog](/blog/) for team perspectives, deeper dives into RAI-related Projects, and more.
+News items will be shared on the [general@rai.apache.org](https://lists.apache.org/list.html?general@rai.apache.org) list — [subscribe](mailto:general-subscribe@rai.apache.org) today!
 
+Our own writing lives on the [blog](blog/); this page collects it alongside coverage elsewhere.
+
+- [Tiny model, full provenance](blog/tiny-model-full-provenance.html) — Apache Sourcelume
+  records provenance for models as well as datasets, demonstrated by a sample that trains a
+  tiny model and writes its own training provenance record in the same run
 - [Responsible AI at Community Over Code, Glasgow](blog/responsible-ai-at-community-over-code-glasgow.html)
   — the hackathon, the roundtable, and where the platform actually is
 - Triumphs, Tooling, TLPs, and talks: [What the ASF Responsible AI Initiative Has