Welcome to AsteriaStar Journal

AsteriaStar spent its first years building an encyclopedia that refuses to say things it cannot support. Field-level provenance on 34,000 scientific values. Live providers that show nothing at all rather than a plausible substitute when an agency cannot be reached. Accuracy figures that are measured against published tables on every build rather than asserted in a footnote.

This publication is the same idea applied to writing.

What it is for

Astronomy reporting has a structural problem. The interesting results arrive as preprints, mission announcements and conference talks — and by the time they reach a reader they have usually been rewritten three times by people who did not read the paper. Each rewrite loses a qualifier. What started as "a candidate signal consistent with, among other explanations, X" arrives as "scientists discover X".

The loss is not usually dishonesty. It is that headlines have no room for the word "candidate", and that the machinery of news does not distinguish between a result that has been checked and one that has not.

So this publication makes that distinction structural. Every article carries an evidence class, displayed at the top, saying what kind of claim it is reporting:

• Peer-reviewed — published after review by other scientists.
• Preprint — posted by its authors and not yet reviewed by anyone else. It may not survive.
• Mission announcement — accurate about what an agency intends. Not a result.
• Observed, measured, modeled, interpreted — different things, often confused.

Those are not filter tags. They are a statement about how much weight the article is asking you to give what follows.

Where the facts come from

Primary sources, and the article names them. NASA, ESA, ESO, STScI, NOIRLab, the IAU, the mission teams themselves, the catalogues astronomy actually runs on — Gaia, SIMBAD, VizieR, the exoplanet archives — and the peer-reviewed literature indexed by ADS.

Not other people's coverage. Not content farms. Not a summary of a summary. When a claim comes from a press release rather than a paper, the article says it came from a press release, because those are different levels of evidence and a reader is entitled to know which one they are being handed.

Every article ends with numbered references, each carrying its publisher, its date, its URL, and the date it was actually read. Where a result exists in the literature, the reference points at the literature — indexed through ADS [2] — rather than at somebody's description of it.

What it will not do

It will not fabricate. No invented quotes, no invented studies, no invented author names, no imagined observations. This is worth saying plainly, in an era when a plausible-sounding citation is easier to generate than a real one.

It will not dress up imagery. Scientific images come from the missions that took them, with the credit, the licence and the source recorded. Artist's impressions are labelled as artist's impressions — they are a legitimate and often necessary illustration, and passing one off as a photograph is not.

It will not quietly rewrite itself. When an article is corrected, the correction is visible on the article and in the corrections log. A publication that edits away its mistakes is asking to be trusted about things nobody can check.

It will not duplicate the encyclopedia. Where an article touches something the platform already documents — a star, a mission, a spacecraft — it links to that page rather than restating it. The encyclopedia is the reference; this is the reporting.

What it will cover

Discoveries and what they actually establish. Missions, including the long unglamorous middles. Observing, with the honest caveat that geometry is not weather. Space weather from the agencies that forecast it. Data releases and what they enable. And the platform itself — what we built, what we connected, and what we got wrong.

The one rule

If a claim cannot be traced to a source that stands behind it, it does not get published — however good the story would have been.

That rule costs stories. It is the only reason to read a publication attached to a scientific platform rather than one attached to an audience.

The standard is not new here — it is the platform's own, written down in its scientific methodology [1], and this publication simply inherits it.