NASA announced a major Artemis architecture overhaul on Oct. 23, 2026. Starship HLS will now dock with Orion in high Earth orbit rather than lunar orbit, then push Orion to the Moon. Blue Origin's Blue Moon Mk2 will defer in-space refueling and instead use three separate transfer stages launched on New Glenn (four launches total). Both companies remain unassigned to a specific landing mission; NASA will choose based on readiness.
The changes solve the long-standing Orion propulsion shortfall — Orion lacks the delta-v to carry itself and a lander into and out of low lunar orbit. Having Starship do the trans-lunar injection burn reduces Orion's required performance. For Blue Origin, skipping orbital refueling removes a major technical risk and schedule pressure. Both changes are aimed at preserving the 2028 landing date.
NASA confirmed on Oct. 23 that both SpaceX and Blue Origin remain in contention for the first crewed Artemis lunar landing, now targeted for 2028 on Artemis IV. Under the revised plan, Starship HLS will dock with Orion in Earth orbit and perform the trans-lunar injection burn, pushing the combined stack to lunar orbit.
NASA's October 2026 HLS timeline targets Artemis III (low Earth orbit docking demo) for Q3 2027, an uncrewed Starship lunar landing demo for Q4 2027, and Blue Origin's Blue Moon Mk2 uncrewed demo for Q1 2028, with the crewed Artemis IV landing in early 2028.
MNASA just tore up its Moon playbook, and both SpaceX and Blue Origin are still in the game.
CBoth of them? I thought this was a winner-take-all contract.
MThat's the old way. The new plan targets 2028 for Artemis IV, the first crewed landing. And instead of picking one lander, NASA keeps both.
CSo it's a bake-off. Two competitors, one Moon. I actually like that.
MMe too, but the details are wild. Starship now docks with Orion in high Earth orbit, then pushes the whole stack to lunar orbit.
CWait, Starship becomes the engine for Orion? That solves the big problem.
MExactly. Orion's propulsion was too weak to drag itself and a lander up from the Moon. Had been a mess for years.
CSo you offload it to Starship. That's actually clever engineering.
MBlue Origin's side is just as interesting. Their Mk2 lander skips refueling entirely and uses three separate transfer stages on New Glenn.
CFour launches total for one landing. That's... a lot of moving parts.
MVersus Starship needing on-orbit refueling. Pick your poison. Eric Berger at Ars Technica says this finally fixes the Orion problem.
CHe's not wrong. But here's my thing: both demo landings are extremely tight. SpaceX uncrewed demo in Q4 2027, Blue Origin Q1 2028.
MThe window is almost nonexistent. If either demo slips, 2028 goes with it.
CSo why not just commit to SpaceX? They're farther along. Starship has flown, it's been caught —
MBut their uncrewed landing demo is also Q4 2027. They haven't landed one on the Moon yet either.
CNo, but — come on, they're clearly ahead on flight heritage.
MHeritage doesn't land on the Moon! Neither company has done it. Keeping both is the smart call.
CIt's also slower. NASA's hedging means more contracts, more oversight, more meetings.
MOr it means a backup when one of them hits a wall. Jared Isaacman said it himself: keep options open as long as possible.
CFine. I'd rather have two shots at 2028 than one shot at 2029.
MHere's a different world entirely: an open-source project recreating the classic Mac OS Platinum desktop for modern PCs.
CPlatinum? The 1990s look with the gray window title bars and the little pencil icon?
MThe one with the tiny triangles instead of scroll arrows, yes. Someone built the whole thing so you can relive it. On Hacker News, one commenter said they use Platinum as a skin in their own projects.
CThere's a certain generation that never got over System 7.
MAnd a younger generation that never saw it. The top reply calls it a month-old 'slopfest' — too soon to trust.
CHarsh. Nostalgia's a powerful drug. I'd install it for ten minutes and remember why we moved on.
MThat's the thing about retro tech: the feeling never survives contact with the actual software.
CSpeaking of old friends, a new version of SpeedCrunch is out. The high-precision calculator for people who think spreadsheet math is for amateurs.
MVersion 1.1. It's been around forever. Engineers love it because it does arbitrary precision — you're not limited to 15 digits.
CIt's the calculator that makes the one on your phone look like it's guessing.
MAnd for the security-minded, a guide on full disk encryption for Fedora with TPM2 is making the rounds. The hardware chip that holds your encryption keys.
CYou encrypt your disk, the keys live in the chip, and if someone pulls your SSD, they get nothing.
MUseful for anyone carrying a laptop through an airport. The guide walks through the whole setup.
CIt's Saturday night, folks. We're talking about encryption and calculators. This is who we are.