Massimo Piazza

That's the same tendency to max out (at least on paper) that we've seen in the 90s & in 2015, but under different scarcity constraints. Some of that is very real, some of that is not. - In the 90s folks rushed to fill up GEO, bottlenecked by the effective orbital real estate available. (Important to note "effective" because you may have a huge geostationary orbital plane, but at multi-hundred-meter orbit uncertainty covariances there's not much you can safely park there) - In 2015 operators rushed to fill up RF spectrum with LEO constellations. Current filings are nowhere close to maxing out on what's possible from an effective real estate perspective (i.e. you end up w/ plenty of shells available at all sorts of inclinations & altitudes between 350 km and say ~1200 km). It's not about how much we can fit, but rather about how much we can emit & coordinate w/o interfering @ given frequency, beamwidth, sidelobes, etc. - Now we're somewhat back to the 90s scenario, but with a way larger real estate compared to GEO, still limited by the fact that everyone wants the dawn-dusk premium. Interference is basically impossible if all your data is routed through laser links (to the extent that optical comms doesn't even need to be regulated). Given the right amount of coordination, GNSS-like covariances and Stargaze, I'm confident plans for 1M+ satellites are absolutely doable from a safety standpoint. But still, you better max out now. I would in general expect anyone w/ access to internal launch at cost to file for a similarly sized constellation, i.e. waiting for you Rocket Lab.

@elonmusk kudos for not deleting one of the best aerospace memes ever x.com/ulalaunch/stat…

34th launch & landing of the same rocket

The market has no clue how much lasercomms is: - an incredibly complex technology to get done on a prototype, let alone in a mass-manufacturing context (i.e. buys them years on schedule vs. in-house R&D) - a non-negotiable element in basically every single large satellite constellation being designed today (i.e. strongly affects performance, but also substantially drives down capex/opex by lowering burden on ground segment) Either way, information asymmetry always feels good. My theory is that the German government actually pressured Rheinmetall to make a bid, and they declined, which would be very short-sighted (especially at that valuation).

🚨 $RKLB receives regulatory approval to acquire Mynaric. The transaction is expected to close in April.

@pronounced_kyle and today we complain if the link margin drops enough to miss 2.7 seconds of re-entry footage when Starship is wrapped in the coolest plasma sheath color

This is what the SpaceX launch live streams looked like in December 2005. They've come a long way!

@neuralink this is so beautiful

ALS has gradually taken away Kenneth’s ability to speak. Through Neuralink’s VOICE clinical trial, he’s exploring how a brain-computer interface designed to translate thought to speech could help restore autonomy in his daily life. Watch to learn more:

To build a sustained human presence on the Moon, we are building @NASAMoonBase, prioritizing surface operations and scalable infrastructure.  - Frequent robotic landings and mobility testing including MoonFall drones  - Starting in 2027 nearly monthly cadence of equipment and rovers with scientific payloads landing on the Moon.  - Investments in power, communications, and surface mobility  - Scalable infrastructure to support long-term human presence The objective is clear: build the foundation for an enduring lunar base and take the next step toward Mars.

@dpoddolphinpro @SpaceAbhi @NASA Back to pushing the envelope. The Jared effect is very real

BREAKING: @NASA to launch first Nuclear-Powered mission to Mars called Space Reactor-1 Freedom - First EVER mission with nuclear-electric propulsion to travel beyond Earth orbit - Launching NET December 2028 - Features a nuclear fission reactor to generate electricity - This electricity will power ion thrusters for propulsion - Repurposes Lunar Gateway's Power & Propulsion Element (PPE) as the bus - Carrying a payload of 3 Ingenuity-class Helicopters to Mars, called Skyfall 📷 @NASA

the future better looks like the future

Time to monitor the situation harder @vonderleyen

@patrick_oshag @WorkMJ And that’s a good thing. And AI will lower the entry barrier as much as it will increase competition, i.e. that’ll remain flat, and that’s a good thing too.

A crazy image

@SciGuySpace I absolutely love that this is a real pic taken before the release of Oppenheimer

Jeff Bezos joins the gold rush to Terminator SSO (and permanent sunlight in space near Earth). arstechnica.com/space/2026/03/…

Absolutely agree. Dawn-dusk is critical now to compensate for launch costs and enable cost competitiveness w/ terrestrial data centers by the end of the decade. It becomes less relevant as Starship (and maybe New Glenn) achieve high cadence. But still a relevant control knob that will affect margins & pricing pressure, especially if one player manages to secure a bigger dawn-dusk chunk than others.

@squareospace Correct. As the architectures get better they can expand off Dawn Dusk as well but if it’s anywhere in your 10 year plan you should probably not wait. China is (will be) going as well.

That's the same tendency to max out (at least on paper) that we've seen in the 90s & in 2015, but under different scarcity constraints. Some of that is very real, some of that is not. - In the 90s folks rushed to fill up GEO, bottlenecked by the effective orbital real estate available. (Important to note "effective" because you may have a huge geostationary orbital plane, but at multi-hundred-meter orbit uncertainty covariances there's not much you can safely park there) - In 2015 operators rushed to fill up RF spectrum with LEO constellations. Current filings are nowhere close to maxing out on what's possible from an effective real estate perspective (i.e. you end up w/ plenty of shells available at all sorts of inclinations & altitudes between 350 km and say ~1200 km). It's not about how much we can fit, but rather about how much we can emit & coordinate w/o interfering @ given frequency, beamwidth, sidelobes, etc. - Now we're somewhat back to the 90s scenario, but with a way larger real estate compared to GEO, still limited by the fact that everyone wants the dawn-dusk premium. Interference is basically impossible if all your data is routed through laser links (to the extent that optical comms doesn't even need to be regulated). Given the right amount of coordination, GNSS-like covariances and Stargaze, I'm confident plans for 1M+ satellites are absolutely doable from a safety standpoint. But still, you better max out now. I would in general expect anyone w/ access to internal launch at cost to file for a similarly sized constellation, i.e. waiting for you Rocket Lab.

“It is crazy we have to say this out loud, but we cannot have a GDPR-like solution for modern company building and investing.” ‘Another GDPR’ should scare the shit out of anyone who cares about European startups. linkedin.com/posts/eu-inc_h…

@beffjezos @AetheroSpace sick! has anyone ever done TID/SEE testing on the chip?

Handing off one of our first XTR-0's to @AetheroSpace TSUs in space soon? 🤔🌌 One step closer to the Dyson swarm 🚀🔥

We strive to always test as we intend to fly. This enables us to derisk components and subsystems before launch. To that end, here is a DeepSpace-2 test article in a vacuum chamber getting prepared for a hotfire thruster test under the control of our full avionics suite.

@KobeissiLetter Jevons entered the chat

Software job postings are rebounding: US software development job postings on Indeed have increased +6.5% since the start of the year, to the highest since January 2024. Since May 2025, openings in this sector have risen +16.7%. As a result, the 21-day moving average is up to the highest since July 2024. By comparison, overall Indeed job postings remain near the lowest since February 2021. That said, software listings are still -69% below their March 2022 peak. At the same time, total openings are -35% below their 2022 high. Has the software job market bottomed?

Jensen announcing a rad-tolerant & high-temp space chip at this year’s GTCs wasn’t on my bingo card tbh. What a time to be alive

also kind reminder that a GW-class data center has the equivalent water footprint of a handful of burger joints. you know... kinda worth it when one of the 6859367e+24 use cases is: curing cancer

things like this are proof that not being a techno-optimist is utterly immoral

we absolutely need coverage on science corner @friedberg @theallinpod things like this are proof that not being a techno-optimist is utterly immoral

You can just do things (genetically sequence your dog’s tumors and design a bespoke mRNA cancer vaccine to save her life)