Atom 1.0
Fourteen feeds
Paste any of these into NetNewsWire, Reeder, Feedly, Miniflux or whatever you use. Each carries the headline, a summary and a link straight to the publisher — full text stays where it was written.
EverythingEvery story, all subjects
AtomThe GalleryNew observatory releases and APOD
AtomLaunch scheduleUpcoming windows as they firm up
AtomLaunch & Flight309 stories · C-band radar
AtomCrewed Spaceflight300 stories · S-band
AtomThe Solar System408 stories · Thermal IR
AtomThe Sun & Space Weather189 stories · He II
AtomStars & Supernovae167 stories · O III
AtomComets & Asteroids95 stories · C₂ Swan
AtomExoplanets & Life123 stories · H₂O
AtomTelescopes & Instruments258 stories · Hβ
AtomSatellites & Orbit331 stories · Ku-band
AtomBlack Holes132 stories · Fe Kα
AtomGalaxies & Cosmology238 stories · Lyman-α
AtomHow to use them
Take the two subjects you care about and ignore the other nine. That is the whole argument for splitting a small site into fourteen feeds rather than publishing one: a reader who wants exoplanets and nothing else should not have to skim launch manifests to find them, and a reader who wants the launch manifest should not be handed cosmology.
Atom 1.0 rather than RSS 2.0, and the difference is not academic. Atom requires a globally unique id per entry and a proper timestamp, which is what stops a reader showing you the same story twice when a headline is corrected — a thing that happens most days here, because publishers revise. Every entry carries the publication's name, the time it was gathered and a link straight to the piece.
Summaries are clipped to about forty words. That is deliberate and it is not a technical limit: a feed that reproduces enough of an article to replace it has taken the publisher's click, and this site is a pointer rather than a substitute. If the summary is enough, you did not need the article; if it is not, the link is right there.
Feeds are rewritten on every gather, which is every three hours, and each one holds the most recent forty entries. Anything older is on the archive rather than in the feed. Nothing here requires an account, sets a cookie, or knows that you subscribed — a feed reader fetching a static file is the last genuinely private thing on the web, which is most of why these exist.
The launch feed behaves slightly differently from the rest: an entry is republished when its window firms up, so a scrubbed launch reappears with the new time rather than sitting in your reader with a date that has passed. The schedule page also publishes calendar files, if a launch belongs in your diary rather than your reader.
Deep Field is one person's side project. This website has no paywall, no newsletter capture and no tracking beyond what the ad network sets. It costs about the price of a coffee a week to run.
Buy me a coffee